You are on page 1of 186

TRANSPORTATI ON

ENGI NEERI NG


ONLI NE LAB
MANUAL



Oregon State University:
Kate Hunter-Zaworski

Portland State University:
Julia Fowler
Kent Lall

University of I daho:
Ty Bardwell
Patrick Bird
Steven Dahl
Cheryl Gussenhoven
Michael Kyte
Melissa Lines
Mark Lovejoy
Josh Nelson
i

About the Manual

This web site (http://www.webs1.uidaho.edu/niatt_labmanual/index.htm) is a laboratory manual
that is designed to supplement the average junior-level course in transportation engineering.
Laboratory exercises are presented for most of the major topics addressed in undergraduate
transportation engineering. Practicing engineers and educational entities view these topics as the
most important ones that civil engineering graduates need to understand.

The lab manual includes a vast collection of help resources, to assist you in completing the
laboratory exercises, and to increase your understanding of these important topics. Each chapter
includes:

discussions of the important theories and concepts.
demonstrations of many of the concepts.
information about how these concepts are applied in professional practice.
example problems with solutions.
links to other web sites, for more information on the topic.
a glossary of terms specific to that topic.

While this list might sound extensive, and it is, this manual is not meant to be a substitute for
lectures or for texts. Additional topics will likely be covered in class, and your text may contain
more detail in its descriptions of certain concepts or calculations.

This lab manual is the product of two years of collaboration between Oregon State University,
Portland State University, and the University of Idaho. All of the individuals who made
contributions to this lab manual, both great and small, are listed below.

Oregon State University:
Kate Hunter-Zaworski

Portland State University:
Julia Fowler
Kent Lall

University of I daho:
Ty Bardwell
Patrick Bird
Steven Dahl
Cheryl Gussenhoven
Michael Kyte
Melissa Lines
Mark Lovejoy
Josh Nelson

The development of this lab manual was funded by:



ii





Learning More

This lab manual contains a tremendous amount of valuable information that is available to you at
any time. To use this information most effectively, however, you should probably develop a
strategy.

Normally, you will be asked by your instructor to complete a laboratory exercise from this manual.
We recommend that you first read the laboratory exercise and then review the related information in
that chapter before you begin working.

For example, lets say that the lab exercise asked you to write a formal evaluation of the current
signal timing conditions at a given signal. You could begin with an overview of the topic, under the
Introduction button. Next, you could review all of the relevant concepts under the Theory &
Concepts button, and then see how these concepts are applied in practice, under the Professional
Practice button. Finally, you could work an example problem or two, under the Example Problems
button, to guarantee that you fully understand the concepts. At any time, feel free to access the
Related Links or the Glossary of Terms.

Even though this web site is called a lab manual, it can be much more. You can use this lab manual
as a supplemental text at any time. If you are having trouble with a particular concept in class, you
can always review that concept in the lab manual, view the Excel demonstration of that concept,
and work an example problem.

We hope that this lab manual will contribute positively to your education in the field of
transportation engineering. We intend to update the site periodically, and add even more content in
the future. Please help us improve this educational resource, by filling out the evaluation form under
"feedback", on the opening page. Tell us what you liked or didnt like about the lab manual, and
anything that you think we should add or change. Thank you for participating in this project.

Acknowledgements

Contributors: This lab manual is the product of two years of collaboration between Oregon State
University, Portland State University, and the University of Idaho. All of the individuals who made
contributions to this lab manual, both great and small, are listed below.

Oregon State University:
Kate Hunter-Zaworski

Portland State University:
Julia Fowler
Kent Lall

University of I daho:
Ty Bardwell
Patrick Bird
iii

Steven Dahl
Cheryl Gussenhoven
Michael Kyte
Melissa Lines
Mark Lovejoy
Josh Nelson

Funding: The development of this lab manual was funded by:








Acknowledgements: The developers would like to thank the following agencies and organizations
for their assistance in the development of this lab manual:

American Association of State Highway and Transportation Officials (AASHTO), Idaho
Transportation Department (ITD), Institute of Transportation Engineers (ITE), Oregon Department
of Transportation (ODOT), Transportation Research Board (TRB), U.S. Department of
Transportation (USDOT).

Quoted Material: The material in the Professional Practice modules of this lab manual was
excerpted from the following publications, with permission:

A Policy on Geometric Design of Highways and Streets, 1994, American Association of State
Highway and Transportation Officials.

Idaho Transportation Department Traffic Manual, 1995, Idaho Transportation Department.

Manual of Traffic Signal Design, 2nd Edition, 1991 Institute of Transportation Engineers.

Special Report 209: Highway Capacity Manual, Third Edition, copyright 1998 by the
Transportation Research Board, National Research Council, Washington, D.C.

Traffic Engineering Handbook, 1992, Institute of Transportation Engineers. Transportation
Planning Handbook, 1992, Institute of Transportation Engineers. (Note: ITE's Traffic
Engineering and Transportation Planning Handbooks will be available in updated versions in
October, 1999.)

Accessibility: The developers of these materials have made a special effort to make these materials
accessible and available to people with disabilities, who use assistive or enhanced computer
technology.


iv

Table of Contents

1. Bus Service Planning ................................................................................................................ 1
1.1. Introduction ........................................................................................................................... 1
1.2. Lab Exercises ........................................................................................................................ 1
1.2.1. Lab Exercise One: Bus Service Planning ................................................................... 1
1.3. Theory and Concepts ............................................................................................................. 2
1.3.1. Evaluation of Demand ................................................................................................ 2
1.3.2. Route and Network Structures ................................................................................... 2
1.3.3. Fare Structure and Payment Options .......................................................................... 3
1.3.4. Preliminary Schedule Design ..................................................................................... 4
1.3.5. Final Schedule Design and Blocking ......................................................................... 5
1.3.6. Importance of Layover Times .................................................................................... 6
1.4. Professional Practice ............................................................................................................. 6
1.4.1. Cycle Time ................................................................................................................. 7
1.4.2. Route Cycle Times ..................................................................................................... 7
1.4.3. Terminal Points .......................................................................................................... 8
1.4.4. Intermediate Time Points ........................................................................................... 8
1.4.5. Blocking ..................................................................................................................... 8
1.5. Example Problems ................................................................................................................. 9
1.5.1. Cycle Time and Number of Vehicles ......................................................................... 9
1.5.2. Vehicle Blocking ...................................................................................................... 10
1.6. Glossary ............................................................................................................................... 11
2. Capacity and Level of Service (LOS) Analysis .................................................................... 13
2.1. Introduction ......................................................................................................................... 13
2.2. Lab Exercises ...................................................................................................................... 13
2.2.1. Lab Exercise: Freeway Analysis .............................................................................. 13
2.3. Theory and Concepts ........................................................................................................... 15
2.3.1. Basic Freeway Section and Ideal Freeway Conditions ............................................ 15
2.3.2. Free-Flow Speed and Flow Rate .............................................................................. 15
2.3.3. Level of Service Criteria .......................................................................................... 16
2.3.4. Determining Flow Rate [d]....................................................................................... 18
2.3.5. Peak Hour Factor ...................................................................................................... 18
2.3.6. Heavy Vehicle Adjustment Factor [d]...................................................................... 19
2.3.7. Free-Flow Speed Adjustment [d] ............................................................................. 20
2.3.8. Determining Level of Service and Density [d] ........................................................ 21
2.3.9. Applications [d] ........................................................................................................ 22
2.4. Professional Practice ........................................................................................................... 22
2.4.1. Basic Freeway Section and Ideal Freeway Conditions ............................................ 23
2.4.2. Determining Flow Rate ............................................................................................ 23
2.4.3. Free-Flow Speed and Flow Rate .............................................................................. 23
2.4.4. Free-Flow Speed Adjustment ................................................................................... 24
2.4.5. LOS Criteria and Capacity ....................................................................................... 24
2.4.6. Determining LOS and Density ................................................................................. 25
2.5. Example Problems ............................................................................................................... 25
2.5.1. Peak Hour Factor ...................................................................................................... 25
v

2.5.2. Heavy Vehicle Adjustment Factor ........................................................................... 26
2.5.3. Calculating Flow Rate .............................................................................................. 27
2.5.4. Free-Flow Speed Adjustment ................................................................................... 27
2.5.5. Determining LOS and Density ................................................................................. 28
2.5.6. Design Application ................................................................................................... 29
2.6. Glossary ............................................................................................................................... 31
3. Geometric Design .................................................................................................................... 32
3.1. Introduction ......................................................................................................................... 32
3.2. Lab Exercises ...................................................................................................................... 32
3.2.1. Lab Exercise One: Geometric Design ...................................................................... 32
3.3. Theory and Concepts ........................................................................................................... 34
3.3.1. Brake Reaction Time ................................................................................................ 34
3.3.2. Braking Distance [d] ................................................................................................ 34
3.3.3. Stopping Sight Distance [d] ..................................................................................... 36
3.3.4. Decision Sight Distance ........................................................................................... 36
3.3.5. Passing Sight Distance [d] ........................................................................................ 37
3.3.6. Horizontal Alignment ............................................................................................... 40
3.3.7. Superelevation and Side-Friction ............................................................................. 40
3.3.8. Minimum Radius Calculations [d] ........................................................................... 41
3.3.9. Design Iterations....................................................................................................... 42
3.3.10. Horizontal Curve Sight Distance [d] ........................................................................ 42
3.3.11. Transition Segments ................................................................................................. 43
3.3.12. Vertical Alignment ................................................................................................... 43
3.3.13. Ascending Grades ..................................................................................................... 44
3.3.14. Descending Grades ................................................................................................... 45
3.3.15. Vertical Curves ......................................................................................................... 46
3.3.16. Crest Vertical Curves [d] .......................................................................................... 47
3.3.17. Sag Vertical Curves [d] ............................................................................................ 48
3.4. Example Problems ............................................................................................................... 50
3.4.1. Stopping Sight Distance ........................................................................................... 50
3.4.2. Passing Sight Distance ............................................................................................. 51
3.4.3. Horizontal Curve Radius Calculations ..................................................................... 52
3.4.4. Horizontal Curve Sight Distance .............................................................................. 53
3.4.5. Transition Segments ................................................................................................. 53
3.4.6. Ascending Grades ..................................................................................................... 54
3.4.7. Crest Vertical Curves ............................................................................................... 54
3.4.8. Sag Vertical Curves .................................................................................................. 55
3.5. Glossary ............................................................................................................................... 56
4. Parking Lot Design ................................................................................................................. 58
4.1. Introduction ......................................................................................................................... 58
4.2. Lab Exercises ...................................................................................................................... 58
4.2.1. Lab Exercise One: Facility Analysis and Design ..................................................... 58
4.3. Theory and Concepts ........................................................................................................... 59
4.3.1. Parking Studies ......................................................................................................... 59
4.3.2. Adequacy Analysis ................................................................................................... 62
4.3.3. Parking Facility Design Process ............................................................................... 62
4.3.4. Entrance Considerations ........................................................................................... 63
vi

4.3.5. Internal Considerations ............................................................................................. 63
4.3.6. Parking Stall Layout Considerations ........................................................................ 64
4.3.7. Exit Considerations .................................................................................................. 67
4.3.8. ADAAG Requirements ............................................................................................ 67
4.4. Professional Practice ........................................................................................................... 68
4.4.1. Parking Studies ......................................................................................................... 68
4.4.2. Types of Facilities .................................................................................................... 69
4.4.3. Types of Operation ................................................................................................... 69
4.4.4. Operational Design Elements ................................................................................... 69
4.4.5. Change of Mode Parking .......................................................................................... 70
4.4.6. Downtown Areas ...................................................................................................... 72
4.4.7. Location .................................................................................................................... 72
4.4.8. Off-Street Zoning ..................................................................................................... 72
4.4.9. Design of Off-Street Facilities ................................................................................. 73
4.4.10. Supplemental Specifications and Implementation ................................................... 74
4.5. Example Problems ............................................................................................................... 75
4.5.1. Adequacy Analysis ................................................................................................... 75
4.5.2. Space Requirements ................................................................................................. 76
4.6. Glossary ............................................................................................................................... 77
5. Roadway Design ...................................................................................................................... 79
5.1. Introduction ......................................................................................................................... 79
5.2. Lab Exercises ...................................................................................................................... 79
5.2.1. Lab Exercise One: Roadway Design ........................................................................ 79
5.3. Theory and Concepts ........................................................................................................... 80
5.3.1. Route Selection and Alignment ................................................................................ 80
5.3.2. Surveys and Maps .................................................................................................... 81
5.3.3. Design Controls and Criteria .................................................................................... 82
5.3.4. Vertical Profile [d].................................................................................................... 83
5.3.5. Cross Section Elements [d] ...................................................................................... 84
5.3.6. Cut and Fill Sections ................................................................................................ 85
5.3.7. Earthwork [d] ........................................................................................................... 86
5.3.8. Designing Bike Lanes .............................................................................................. 87
5.4. Professional Practice ........................................................................................................... 88
5.4.1. Route Selection......................................................................................................... 88
5.4.2. Surveys and Maps .................................................................................................... 89
5.4.3. Design Controls and Criteria .................................................................................... 89
5.4.4. Horizontal and Vertical Alignment .......................................................................... 90
5.4.5. Cross Sections .......................................................................................................... 90
5.5. Example Problems ............................................................................................................... 91
5.5.1. Traffic Volume ......................................................................................................... 91
5.5.2. Vertical Alignment ................................................................................................... 91
5.5.3. Cross Sections .......................................................................................................... 92
5.6. Glossary ............................................................................................................................... 93
6. Signal Timing Design ............................................................................................................. 94
6.1. Introduction ......................................................................................................................... 94
6.2. Lab Exercises ...................................................................................................................... 94
vii

6.2.1. Lab Exercise One: Signal Timing and LOS ............................................................. 94
6.3. Theory and Concepts ........................................................................................................... 95
6.3.1. Basic Timing Elements ............................................................................................ 95
6.3.2. Queuing Theory [d] .................................................................................................. 96
6.3.3. Design Process Outline ............................................................................................ 97
6.3.4. Intergreen Time [d] .................................................................................................. 98
6.3.5. Pedestrian Crossing Time, Minimum Green Interval [d] ....................................... 100
6.3.6. Saturation Flow Rate and Capacity [d] .................................................................. 101
6.3.7. Peak Hour Volume, Peak Hour Factor, Design Flow Rate .................................... 102
6.3.8. Critical Movement or Lane [d] ............................................................................... 103
6.3.9. Cycle Length Determination [d] ............................................................................. 103
6.3.10. Green Split Calculations [d] ................................................................................... 104
6.3.11. Timing Adjustments ............................................................................................... 105
6.3.12. Computing Delay and LOS [d] .............................................................................. 105
6.4. Professional Practice ......................................................................................................... 106
6.4.1. Design Process Outline .......................................................................................... 107
6.4.2. Intergreen Time ...................................................................................................... 107
6.4.3. Pedestrian Crossing Time, Minimum Green Time ................................................ 109
6.4.4. Capacity/Saturation Flow Rate ............................................................................... 110
6.4.5. Peak Hour Volume, Design Flow Rate, PHF ......................................................... 111
6.4.6. Critical Movement or Lane .................................................................................... 112
6.4.7. Cycle Length Determination .................................................................................. 113
6.4.8. Green Split Calculations ......................................................................................... 114
6.4.9. Timing Adjustments ............................................................................................... 114
6.4.10. Computing Delay and LOS, Operational Analysis Outline ................................... 115
6.5. Example Problems ............................................................................................................. 116
6.5.1. Intergreen Time ...................................................................................................... 116
6.5.2. Pedestrian Crossing Time, Minimum Green Interval ............................................ 117
6.5.3. Capacity/Saturation Flow Rate ............................................................................... 118
6.5.4. Peak Hour Volume, Design Flow Rate, PHF ......................................................... 118
6.5.5. Critical Movement or Lane .................................................................................... 119
6.5.6. Cycle Length Determination .................................................................................. 120
6.5.7. Green Split Calculations ......................................................................................... 121
6.5.8. Timing Adjustments ............................................................................................... 121
6.5.9. Computing Delay and LOS .................................................................................... 122
6.6. Glossary ............................................................................................................................. 122
7. Traffic Flow Theory ............................................................................................................. 125
7.1. Introduction ....................................................................................................................... 125
7.2. Lab Exercises .................................................................................................................... 125
7.2.1. Lab Exercise One: Flow Models [d] ...................................................................... 125
7.2.2. Lab Exercise Two: Shock Waves/Queue Formation ............................................. 126
7.3. Theory and Concepts ......................................................................................................... 126
7.3.1. Types of Traffic Flow............................................................................................. 126
7.3.2. Traffic Flow Parameters ......................................................................................... 127
7.3.3. Speed-Flow-Density Relationship .......................................................................... 128
7.3.4. Special Speed & Density Conditions ..................................................................... 129
7.3.5. Greenshields Model [d] ......................................................................................... 129
7.3.6. Time-Space Diagrams [d] ...................................................................................... 131
viii

7.3.7. Shock Waves [d] .................................................................................................... 132
7.3.8. Queuing Theory ...................................................................................................... 133
7.4. Professional Practice ......................................................................................................... 133
7.4.1. Traffic Flow Parameters ......................................................................................... 134
7.4.2. Speed-Flow-Density Relationships ........................................................................ 136
7.4.3. Greenshields' Model ............................................................................................... 138
7.4.4. Shock Waves and Continuum Flow Models .......................................................... 138
7.4.5. Queuing Theory ...................................................................................................... 141
7.5. Example Problems ............................................................................................................. 142
7.5.1. Greenshield's Model ............................................................................................... 143
7.5.2. Shock Waves .......................................................................................................... 143
7.5.3. Traffic Flow Model ................................................................................................ 146
7.6. Glossary ............................................................................................................................. 147
8. Travel Demand Forecasting ................................................................................................ 149
8.1. Introduction ....................................................................................................................... 149
8.2. Lab Exercises .................................................................................................................... 149
8.2.1. Lab Exercise 1: The Gravity Model ....................................................................... 149
8.2.2. Lab Exercise 2: Cross-Classification ..................................................................... 151
8.3. Theory and Concepts ......................................................................................................... 151
8.3.1. Overview of the TDF Process ................................................................................ 152
8.3.2. Description of the Study Area ................................................................................ 152
8.3.3. Trip Generation Analysis ....................................................................................... 155
8.3.4. Multiple Regression Analysis ................................................................................ 156
8.3.5. Experience Based Analysis .................................................................................... 157
8.3.6. Trip Distribution Analysis ...................................................................................... 157
8.3.7. The Logit Model ..................................................................................................... 157
8.3.8. The Gravity Model ................................................................................................. 158
8.3.9. Modal Choice Analysis .......................................................................................... 158
8.3.10. Trip Assignment Analysis ...................................................................................... 159
8.3.11. Results .................................................................................................................... 160
8.4. Professional Practice ......................................................................................................... 160
8.4.1. Zones and Zoning ................................................................................................... 161
8.4.2. Networks and Nodes .............................................................................................. 161
8.4.3. Trip Generation Analysis ....................................................................................... 161
8.4.4. Trip Distribution ..................................................................................................... 162
8.4.5. Modal Choice ......................................................................................................... 164
8.4.6. Trip Assignment ..................................................................................................... 164
8.4.7. Model Calibration and Validation .......................................................................... 165
8.5. Example Problems ............................................................................................................. 166
8.5.1. Cross Classification ................................................................................................ 166
8.5.2. Gravity Model ........................................................................................................ 170
8.5.3. Logit Model ............................................................................................................ 173
8.5.4. Traffic Assignment ................................................................................................. 174
8.6. Glossary ............................................................................................................................. 176
1

1. Bus Service Planning

1.1. I ntroduction

Public transit systems serve many useful functions in the modern world. When designed properly,
public transit can provide an efficient and convenient alternative to private passenger vehicles.
Reducing the use of private automobiles has several benefits, not the least of which are reduced
congestion in transportation networks and fewer harmful chemical emissions.

The principle problem facing transit engineers is the development of transit systems that encourage
patronage, provide dependable and efficient service, and are operable within the budgetary and
political constraints that exist within their districts.

Bus service is the most common form of public transit. Its popularity is based on its flexibility,
expandability, and low cost. Since the bus service planning process can be long and complex, it is
often done with the help of computer software that can try millions of potential route structures,
service schedules, and employee schedules.

This chapter will introduce most of the important concepts in bus service planning, but will fall
short of being a comprehensive guide to the subject. Once you are in practice, your transportation
district and transit authority will be able to provide you with procedures and guidelines that have
been developed from years of experience in the area.

1.2. Lab Exercises

This exercise will help increase your understanding of Bus Service Planning, by presenting a more
complicated problem that requires more thorough analysis.

1.2.1. Lab Exercise One: Bus Service Planning

Your city is considering a new weekday bus route. Your task will be to study the route and
determine the cost of providing this service. You will present your results in a brief report to the
city manager, who must then make a decision on the feasibility of this service.

Your instructor will provide the specific information on the route, including the points to be served
by the route and other relevant information.

This assignment is divided into several parts:

What street segments should be included in this service?
How long does it take to travel this route during different periods of the day?
How many vehicles are required to provide service on this route?
How many bus drivers will be required to operate this service?

Tasks To Be Completed

As you complete the following tasks, consider what information will be useful to the city manager,
who must decide on the feasibility of this service.

Task 1. Your instructor will assign an area in your city that is to be served by a new public transit
route. You should visit the area and document the key areas to be served and the street segments
that are appropriate for bus operations.
2


Task 2. Based on your site visit, determine the length of the route, the average running speed of the
bus during both the peak and off-peak periods, the bus stops that should be included along the route,
the terminal points on the route, the times of operation, and the headways that should be provided.

Task 3. Calculate the number of vehicles required in order to provide this service during the
different service periods.

Task 4. Prepare a headway sheet showing the schedule.

Task 5. Determine the vehicle blocks necessary to serve this schedule.

Task 6. If the average driver cost per hour is $80 (including fringe benefits and overhead), what is
the total annual operating cost for this service (assume 255 weekdays per year).

Task 7. Prepare a brief report summarizing the results of your work, including documentation of
your site visit, computations, results, and conclusions.

1.3. Theory and Concepts

A course in transportation engineering would not be complete without discussing some elements of
Bus Service Planning. Most junior level courses introduce several aspects of Bus Service Planning,
including the topics listed below. To begin learning about Bus Service Planning, just click on the
link of your choice.

1.3.1. Evaluation of Demand

Whether you are contemplating the addition of new routes to an existing bus transit system, or
developing an entirely new bus system in a community, youll need to estimate the number of users
that your new routes will service. Once you have estimated where and when the demand will be
present, you can design your bus transit system to service that demand.

As discussed in the chapter entitled "Travel Demand Forecasting," you can divide the area that you
want to service into regions and conduct trip generation, trip distribution, and mode split analyses of
the region with your proposed route structure in place. This will give you an estimate of the number
of users that will decide to use your new bus route(s) instead of their current means of
transportation.

Once you have this information, you can use existing traffic data and microanalysis of the regions
to determine when the peak travel periods are and what specific destinations are the most common.
For example, if you knew that a school was present in zone "A" and that zone "B" was primarily
residential, you might deduce that a high demand for travel from zone "A" to zone "B" would exist
around 3:00 PM.

Having a firm understanding of the demand for bus service in both the spatial and temporal
dimensions will make the remainder of the bus service planning process much easier.

1.3.2. Route and Network Structures

A network is a system of routes. Routes are individual paths that are taken by transit vehicles.
Routes include a spatial element -the streets and stops that are serviced along the way, and a
temporal element- the time that the bus will arrive and depart from each stop or station.
3


Selecting a network structure is a complicated task, for which there is not a simple solution. There
are, however, a few network structures that have become very common.

Grid networks are common in large cities. These systems tend to be centered on the central business
district with few routes venturing far outside the central business district (CBD). Grid networks
make extensive use of the existing roadways. Where traffic is heavy, deep within the CBD, one or
more exclusive bus lanes may be required in order to provide buses with adequate freedom to move.
Exclusive bus lanes increase the capacity of the system by reducing delays caused by interfering
traffic, but the exclusive bus lanes also reduce the capacity of roadways to handle private traffic and
parking.

Radial networks are also frequently found in modern cities. These systems contain linear routes
from the CBD to outlying suburbs. Commuters who live in the suburbs and work in the central
business district are well served by radial networks, but those who want inter-suburb transportation
are not well served, unless there are direct lines connecting each of the suburbs. Exclusive bus lanes
are occasionally included on radial routes as well.

Many modern cities employ transit systems that are a combination of the grid and radial networks.
These networks transport individuals to and from the suburbs using radial routes and then provide
transportation within the CBD via a grid network. The route structure should serve the needs of the
population; therefore, each communitys needs require special consideration.

1.3.3. Fare Structure and Payment Options

Bus service planning encompasses not only the calculation of where and when buses will arrive, but
also how much each passenger will be required to pay and how the payment will be received.
Poorly designed fare schedules and fare collection procedures can be a source of significant
confusion and delay.

The amount that passengers must pay for a particular trip can be calculated in several ways.

A city may choose to adopt a uniform fare for all routes in the transit network. While this rate
structure is simple, it also penalizes those who travel short distances on the network.
A more equitable solution would be to adjust the fare based on the distance the user traveled on
the network, but this system is prohibitively complex.
Many transit authorities have decided on a compromise that charges users based on the number
of zones that they travel through on a given route. Travel from zone "A" to zone "E" would cost
the user more than the shorter trip from zone "A" to zone "C." This system is reasonably simple
and much more equitable than the uniform fare system.

Fare collection is another complicated issue, for which several solutions have been devised.

The driver can collect fares from each boarding passenger. While simple, this system causes
large delays at every stop, as the driver must interact with each passenger as they board.
To reduce delay, fare collection machines that accept payment from the passengers are
commonly installed near the bus door. These machines allow the bus driver to focus on driving,
and accelerate the boarding process considerably.
Finally, fare card programs are becoming more and more common. These systems allow the
transit user to purchase a magnetic card with a predetermined value. The fare is deducted when
the passenger swipes the card through a reader at the bus door. This system is very efficient. In
4

addition, it allows the transit authority to monitor the transportation habits of the cardholders by
automatically recording the routes, stops, and times at which each card is used.

1.3.4. Preliminary Schedule Design

Designing a schedule can be quite complicated, so preliminary schedule design will be portrayed
here in the form of an example.

Consider a transit route that connects a residential neighborhood to a central business district. The
distance between the neighborhood and the downtown area is 5 miles. The transit vehicles average
12 miles per hour between the two terminal points. The goal is to provide transit service every 15
minutes along the route.

The first step is to determine the time required to travel from one end of the route to the other. The
one-way trip time is given in the equation below:

One-Way Trip Time = Route Length / Average Operating Speed
One-Way Trip Time = 5 miles / 12 mph
One-Way Trip Time = 25 minutes

The total round-trip time is twice the one-way trip time, or 50 minutes.

The next step is to determine the number of vehicles required in order to operate at the desired level
of service. Now suppose that the desired headway is 15 minutes. That is, the frequency of service is
one vehicle every 15 minutes. How many vehicles would be required to provide this service?

Number of Vehicles Required = Total Round Trip Time / Headway
Number of Vehicles Required = 50 minutes / (15 min/vehicle)
Number of Vehicles Required = 3.33 or 4

The revised round-trip time can now be calculated.

Revised Round Trip Time = (Number of Vehicles) (Headway)
Revised Round Trip Time = (4 vehicles) (15 minutes/vehicle)
Revised Round Trip Time = 60 minutes

This leaves 10 minutes for recovery and layover time, since the actual round-trip running time is 50
minutes.

The capacity of the route can also be determined.

Capacity = (Vehicles) (Capacity/Vehicle)
Capacity = (4 vehicles/hour) (75 passengers/vehicle)
Capacity = 300 passengers/hour

Now suppose that the forecasted demand for this transit route is 400 passengers per hour at the peak
loading point. We need to re-estimate the required vehicles because the capacity calculated above is
insufficient to carry this projected demand.

# Vehicles = (400 passengers/hour) / (75 passengers/vehicle)
# Vehicles = 5.33 vehicles/hour
Headway = 60 minutes / 5.33 vehicles
5

Headway = 11.25 = 10 minutes/vehicle

Note that we use an even clock headway of 10 minutes, rather than the cumbersome and
potentially confusing value of 11.25 minutes that we initially calculated.

# Vehicles = (50 minutes + 10 minutes) / (10 minutes/vehicle)
# Vehicles = 6

At this point, we have completed the preliminary calculations in schedule design. The final
computations involve the development of the schedule and the vehicle 'blocks.' These computations
are presented in the 'final schedule design and blocking' discussion.

1.3.5. Final Schedule Design and Blocking

The final computations in schedule design will produce a summary of the activity that will occur on
the route during the period in question. Well continue our example problem, which was introduced
in the 'preliminary schedule design' section, to illustrate the steps and the desired result.

Our preliminary schedule design conclusions were that we needed 6 vehicles running with
10-minute headways to service the demand of 400 passengers/hour between A and B. Lets
assume that these calculations were meant for the morning peak-period of 7:00 a.m. through 9:00
a.m.

First, we list the departure times from A for each vehicle during the peak-period.

Leave A 7:00 7:10 7:20 7:30 7:40 7:50 8:00 8:10 8:20 8:30 8:40 8:50

Next, since we know that it takes 25 minutes for each vehicle to proceed from A to B, we can
record the arrival times. Including 5-minutes of layover time at each terminal A and B, we can
include the departure times as well. Notice that the work so far has been vehicle-independent. We
are only recording the times at which these events should occur, not which vehicle should be at each
station at these times.

Leave A Arrive B Leave B Arrive A
7:00 7:25 7:30 7:55
7:10 7:35 7:40 8:05
7:20 7:45 7:50 8:15
7:30 7:55 8:00 8:25
7:40 8:05 8:10 8:35
7:50 8:15 8:20 8:45
8:00 8:25 8:30 8:55
8:10 8:35 8:40 9:05
8:20 8:45 8:50 9:15
8:30 8:55 9:00 9:25
8:40 9:05 9:10 9:35
8:50 9:15 9:20 9:45

Now that we have a schedule of times, we can try to link together these times into routes that
specific vehicles can follow. For example, if a vehicle were to leave A at 7:00, it would arrive at
A again at 7:55. This vehicle could then start again with the 8:00 shift. Extending this process
leads to the table below.
6


Vehicle Leave A Arrive B Leave B Arrive A
1 7:00 7:25 7:30 7:55
2 7:10 7:35 7:40 8:05
3 7:20 7:45 7:50 8:15
4 7:30 7:55 8:00 8:25
5 7:40 8:05 8:10 8:35
6 7:50 8:15 8:20 8:45
1 8:00 8:25 8:30 8:55
2 8:10 8:35 8:40 9:05
3 8:20 8:45 8:50 9:15
4 8:30 8:55 9:00 9:25
5 8:40 9:05 9:10 9:35
6 8:50 9:15 9:20 9:45

At this point, we can prepare the final vehicle block summary. This summary simply indicates the
times that each vehicle will be in service and the vehicle block that the vehicle will be assigned to.

Vehicle Vehicle Block Time Block Time In Service
A 1 7:00-8:55 1:55
B 2 7:10-9:05 1:55
C 3 7:20-9:15 1:55
D 4 7:30-9:25 1:55
E 5 7:40-9:35 1:55
F 6 7:50-9:45 1:55
Total 11:30

The tables that have been developed in this section are the ultimate result of schedule design.

1.3.6. I mportance of Layover Times

Layover, while mentioned only casually in the schedule design discussion, is an important part of
the schedule. The layover period serves a variety of functions.

First, it provides a window of time to compensate for vehicles that are running ahead of or behind
schedule. The layover can be extended or shortened in order to keep vehicles on schedule.

Next, the layover time provides an opportunity for drivers to relax and prepare for the next run. In
fact, labor unions usually require layovers periods that are a certain percentage of the cycle length.

Finally, layover periods can be used to change drivers, or for other administrative purposes.

1.4. Professional Practice

In order to supplement your knowledge about the various concepts within Bus Service Planning,
and in order to give you a glimpse of how these various topics are discussed in the professional
environment, we have included selected excerpts from professional design manuals.

7

1.4.1. Cycle Time

The following excerpt was taken from the Transit Cooperative Research Program (TCRP) Report
30, page 19.

Cycle time is the number of minutes needed to make a round trip on the route, including
layover/recovery time. Cycle time is important for several reasons, including playing a part in the
formula used for determining the number of vehicles needed to provide a given level of service on a
route.

Since cycle time equals the number of minutes needed to make a round trip, including the
layover/recovery time, the scheduler determines the amount of time it takes to operate or "run" from
one end of the route to the other and back, then adds layover/recovery time to yield the cycle time.

Minimum vs. Available Cycle Time

For many agencies, on some or all routes, the amount of layover/recovery time is often determined
by labor agreement or agency policy. These agreements or policies dictate a minimum number of
minutes that must be built into the schedule for layover/recovery.

Minimum cycle time is the number of minutes scheduled for a vehicle to make a round trip,
including a minimum layover/recovery time as dictated by labor agreement or agency policy . . . .
However, maintaining a constant headway . . . will, in most cases, result in a cycle time other than
the minimum cycle time for the vehicles operating that route . . . . The resulting cycle time (which
includes the additional layover/recovery time) necessary to maintain the 30-minute headways is
now called the available cycle time. In the optimal case, the minimum cycle time would be the same
as the available cycle time. However, maintaining fixed, clock multiple headways often makes that
impossible.

1.4.2. Route Cycle Times

The following excerpt was taken from the Transit Cooperative Research Program (TCRP) Report
30, page 4.

Cycle time is the time it takes to drive a round trip on a route plus any time that the operator and
vehicle are scheduled to take a break (layover and/or recovery time) before starting out on another
trip.

Typical service standards attempt to maximize the length of the route design per cycle time, while
providing for the minimum amount of layover/recovery time allowed. Maximizing route length per
cycle time utilizes equipment and labor power most effectively. However, other considerations
make this optimization difficult to achieve.

Other considerations that make optimization of labor and equipment difficult include:

the need to maintain consistent time between vehicles on a route (headway).
adjusting for changes in ridership and traffic during the day (for example, rush hour vs. non rush
hour).
planning for vehicles to arrive at common locations so that passengers may make transfers to
other routes (timed transfers).

These considerations often require additional layover/recovery time beyond the minimum allowed.
8


1.4.3. Terminal Points

The following excerpt was taken from the Transit Cooperative Research Program (TCRP) Report
30, page 16.

Terminal points are considered the "ends" of a line or route. These are the locations where vehicles
generally begin and/or end their trips and operators usually take their layovers. For that reason,
locations where there is safe parking and restrooms close by are considered desirable locations for
terminal points.

How many terminal points are usually on a route?

Loop routes that operate only in one direction generally have only one terminal point. A basic
end-to-end route with bi-directional service and no branches or short turns generally has two
terminal points, one located at each end of the route. Routes with more complex patterns generally
have more than two terminal points.

1.4.4. I ntermediate Time Points

The following excerpt was taken from the Transit Cooperative Research Program (TCRP) Report
30, page 21.

Intermediate time points are locations along the route, between the terminals, that indicate when the
vehicle will be there. The term "node" is commonly used in computerized scheduling systems to
denote a time point. Generally speaking, on public timetables, these intermediate time points, or
nodes, are timed to be between 6 and 10 minutes apart.

In theory, when intermediate time points are too close together, there is a greater risk that the
operator may arrive early and have to wait or "dwell" at that point to stay on schedule, causing
passengers to become impatient. When time points are more than 10 minutes apart, some agencies
believe that customers are more likely to be confused about when a vehicle will arrive at a
particular stop, given the differences in individual operator driving habits.

Where are intermediate time points typically located?

Physical location considerations also affect the selection of intermediate time points. Major
intersections that are widely recognized and possess good pedestrian amenities like sidewalks and
actuated traffic signals make good time points.

It is a good idea to locate intermediate time points at major trip generator locations such as
shopping centers, hospitals, and government buildings. Time points are also useful at locations
where time is critical, such as major employment centers and intersecting bus routes or rail centers.

1.4.5. Blocking

The following excerpt was taken from the Transit Cooperative Research Program (TCRP) Report
30, page 38.




9

What is "blocking"?

Blocking is the process of developing vehicle assignments. These assignments, or blocks, describe a
series of trips that are "hooked" together and assigned to a single vehicle.

The vehicle trips that are linked together as part of the block may cover more than one route and
may also involve more than one operator during the course of the vehicle workday. However, the
block refers to the work assignment for only a single vehicle for a single service workday.

Why is blocking important?

Blocking is a critical element in the scheduling process because it serves as the basis for both the
costs associated with operating the revenue service vehicle as well as influencing the cost associated
with work assignments for operators.

1.5. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable concepts. The more concerned you
are about your understanding of a topic, the more seriously you will want to approach the example
problem for that topic.

1.5.1. Cycle Time and Number of Vehicles

You have just designed a route that requires 65 minutes to travel round trip without any
layover/recovery time at the route terminals. Your boss indicates that a layover/recovery period of
at least 5% of the round trip time must be included at either end of the route. Adjust the cycle time
to include the layover/recovery time and determine the number of vehicles that will be required to
service this route if the required headway is 15 minutes.

Solution

First, youll need to adjust the cycle time so that it includes the necessary layover/recovery periods.

C = Travel Time + Layover / Recovery Time
C = 65 minutes + 65 (10%) = 71.5 minutes
C = 72 minutes (rounded to the next whole number)

Next, youll need to determine the number of vehicles that are required.

Number of Vehicles = Cycle Time / Headway
Number of Vehicles = 72 / 15 = 4.8 vehicles
Number of Vehicles = 5 vehicles (rounded to the next whole number)

Since the number of vehicles and the headway are set, we should solve for the new cycle time and
determine what layover/recovery period is actually provided.

Cycle Time = Number of Vehicles Headway
Cycle Time = 5 15 = 75 minutes

10

Since the actual travel time is 65 minutes and the total cycle time is 75 minutes, the total
layover/recovery period for each bus is 10 minutes, or 5 minutes at each end. The layover/recovery
period at each end is, therefore, 7.6% of the travel time and the number of vehicles that are required
is 5.

1.5.2. Vehicle Blocking

A two terminal bus route has the following characteristics during the evening peak period (4-6 PM):

cycle time = 90 minutes
layover/recovery time = 5 minutes at each terminal
number of vehicles = 6
headway = 15 minutes.

Develop a vehicle block summary for this time period.

Solution

The easiest place to start is with the departures. Develop a table that indicates when vehicles should
depart from the first terminal.

Leave TP #1 4:00 4:15 4:30 4:45 5:00 5:15 5:30 5:45

Next, determine the travel time between terminals.

Cycle Time = 90 minutes
Layover/Recovery time = 10 minutes total
Travel Time = 80 minutes
Travel Time between terminals = 40 minutes

Now that we know that the travel time between terminals is 40 minutes and the layover/recovery
time is 5 minutes at each terminal, we can develop a table of arrivals and departures for both
terminals.

Leave TP #1 Arrive TP #2 Leave TP #2 Arrive TP #1
4:00 4:40 4:45 5:25
4:15 4:55 5:00 5:40
4:30 5:10 5:15 5:55
4:45 5:25 5:30 6:10
5:00 5:40 5:45 6:25
5:15 5:55 6:00 6:40
5:30 6:10 6:15 6:55
5:45 6:25 6:30 7:10

Now we can hook the trips together and form the initial vehicle blocks.





11

Vehicle Leave TP Arrive TP #2 Leave TP #2 Arrive TP
1 4:00* 4:40 4:45 5:25
2 4:15* 4:55 5:00 5:40
3 4:30* 5:10 5:15 5:55**
4 4:45* 5:25 5:30 6:10**
5 5:00* 5:40 5:45 6:25**
6 5:15* 5:55 6:00 6:40**
1 5:30 6:10 6:15 6:55**
2 5:45 6:25 6:30 7:10**
* Vehicle enters service
** Vehicle leaves service

Now we can create the final block summary.

Vehicle Time Block Time in Service
1 4:00 6:55 2:55
2 4:15 7:10 2:55
3 4:30 5:55 1:25
4 4:45 6:10 1:25
5 5:00 6:25 1:25
6 5:15 6:40 1:25
Total 11:30

1.6. Glossary

Blocking: assigning trips to vehicles so that each vehicle works continuously and proper headways
are maintained.

Cycle Time: the total time required to complete a full cycle. The cycle time includes the running
time and the layover/recovery time.

Exclusive Bus Lanes: roadway lanes that are meant to be used by buses only. These lanes reduce
conflicts with passenger cars and other traffic.

Fare: the amount of money that is charged for riding a transit vehicle.

Headway: the time that should elapse between consecutive buses arriving at stations or terminal
points.

Layover/Recovery Time: the time that transit vehicles should remain stationary at each terminal
point. The layover/recovery time is used for resting, administrative purposes, and for maintaining
proper headways.

Network: a system of routes.

Operating Speed: the average speed at which a transit vehicle can traverse the route in question,
including intermediate stops.

Policy Headway: the headway set by the local transit authority.

Route a specific physical path that a transit vehicle follows.
12


Running Time: the portion of the cycle time that is spent traveling, not in layover/recovery.

Schedule: the temporal path that a transit vehicle follows, or a listing of times at which the transit
vehicle should be located at various places.

Terminal Point: a point at the end of the route, or any other designated point, at which the transit
vehicle may enter or leave service or remain stationary for a few moments for layover/recovery.
There are normally two terminal points on linear routes, one at each end.

Travel Time: see Running Time above.

Vehicle Block Summary: a table listing the time intervals that each vehicle will be in service and
also listing the total time that each vehicle will be in service.






































13

2. Capacity and Level of Service (LOS) Analysis

2.1. I ntroduction

The number of vehicles on our highways increases every year, and transportation engineers are
often faced with the challenge of designing modifications to existing facilities that will service the
increased demand. As part of this work, the engineers must evaluate the capacity of the existing and
proposed systems. In addition, engineers are often required to justify the expense of modifying or
adding facilities by looking at the current and potential levels of service.

Capacity and Level of Service (LOS) are closely related and can be easily confused. To help clarify
the difference between the two, imagine a phone booth that contains ten people. The phone booth
obviously has a capacity of ten or more people, but its likely that the level of service (quality of
service) would be unanimously unacceptable. Capacity is a measure of the demand that a highway
can potentially service, while level of service (LOS) is a measure of the highways operating
conditions under a given demand.

Traffic engineers use capacity and level of service analyses to:

Determine the number and width of lanes needed for new facilities or for expanding existing
facilities.
Assess service levels and operational characteristics of existing facilities that are being
considered for upgrading.
Identify traffic and roadway changes needed for new developments.
Provide base values for determining changes in fuel consumption, air pollutant emissions, road-
user costs, and noise associated with proposed roadway changes.

Capacity and level of service (LOS) are fundamental concepts that are used repeatedly in
professional practice. Because of their obvious importance, this chapter is designed to introduce the
undergraduate engineering student to capacity and level of service (LOS).

2.2. Lab Exercises

This exercise will help increase your understanding of Capacity and LOS Analysis, by presenting a
more complicated problem that requires more thorough analysis.

2.2.1. Lab Exercise: Freeway Analysis

Your State Department of Transportation (DOT) has been given funding for the construction of a
new freeway to relieve the congestion on the existing freeway that extends through the downtown
section of your community. Construction of the new freeway will begin this year and is expected to
take a minimum of three years to complete. In the interim, the DOT has decided to evaluate the
most congested section of the existing freeway for immediate improvements that would keep traffic
flow at less than capacity on this section while the new freeway is under construction. Your
supervisor has given you the task of determining the most economical improvements for the
existing freeway, so that operations do not exceed capacity for the next several years. This
3,000-foot section of existing freeway has the following characteristics:

an interchange density of 1 per mile
free-flow speeds of 58 mph on the upgrade and 62 mph on the downgrade
carries 1600 vehicles per hour during the peak hour (in one direction) on a grade of 5%
four asphalt-paved lanes (two in each direction)
14

11-foot lanes with 1-foot right shoulder lateral clearance to a concrete barrier, a 4-foot median
with a concrete barrier
PHF = 0.85
11% trucks, 4% buses and no RVs
the total right-of-way consists of 64 feet

This assignment is divided into the following parts:

1. What is the existing LOS on this 3,000-foot section of freeway?
2. Given that the anticipated rate of annual growth in traffic volume in the area is expected to be
15%, what will the LOS be in three years?
3. What are possible improvements (and their estimated costs) that can be made to the existing
freeway to delay capacity flow conditions for three years, given that the existing right-of-way
cannot be expanded?

Tasks to be Completed

As you complete the following tasks, you will determine the most economical improvements that
can be implemented on the existing freeway to delay the onset of capacity flow operations.

Task 1. Calculate the free-flow speed and convert volume (vph) to flow rate (pcphpl) for the
existing freeway, in both directions (upgrade and downgrade). You will first need to calculate the
upgrade and downgrade heavy-vehicle adjustment factors in order to convert volume to flow rate.
For selecting the driver population adjustment factor, you can assume that the traffic is mostly
commuters who are familiar with the freeway. Using your calculated free-flow speed, construct an
appropriate speed-flow curve of the same shape as the typical curves on the free-flow speed versus
flow rate graph. The curve should intercept the y-axis at your calculated free-flow speed. The LOS
for the upgrade and downgrade can be determined directly from the graph.

Task 2. Using the anticipated growth rate of 15% per year, determine what the traffic volume and
flow rate will be in three years. Then repeat all of the steps in Task 1 except for calculation of the
free-flow speed to determine what the LOS will be in three years. How many years will it be until
the upgrade section is operating close to capacity?

Task 3. Given that the right-of-way cannot exceed a total of 64 feet, develop possible improvements
that will forestall capacity operations on this section of freeway. At a minimum, you should
evaluate the effects of changing lane widths, lateral clearances and number of lanes (specifically in
the upgrade direction). Setting up a spreadsheet program to calculate free-flow speed and flow rate
will make this task relatively straightforward.

Task 4. Estimating costs may prove to be the most challenging part of this exercise. What you need
to keep in mind is that the term "estimated" means just that - an estimate. The purpose of this
exercise is simply to introduce you to the costs associated with highway improvement projects.
Possible resources include cost data manuals (such as the RS Means Cost Manuals), government
transportation offices, and the civil engineering department at your university.

Task 5. You are to present the data from Tasks 1 through 4 in a report that includes, at a minimum,

1. a summary of existing conditions and LOS,
2. a description of when the freeway will reach capacity with no improvements,
3. a list of possible improvements and the respective effects on LOS, and
15

4. your recommended improvement(s) and estimated costs to implement your recommendation.
Remember to note any assumptions in your report.

2.3. Theory and Concepts

Evaluating the capacity and LOS of a roadway probably seems like a daunting task. In reality, the
calculations are really quite simple. The Theory and Concept links that are located below should
help you navigate through the process with ease. Topics followed by the characters '[d]' include an
Excel demonstration.

2.3.1. Basic Freeway Section and I deal Freeway Conditions

A basic freeway section is a segment where there are no interruptions to the flow of traffic.
Interruptions to traffic flow occur when vehicles enter or leave the freeway. Therefore, a basic
freeway section is one where on or off ramps are not present for at least 1500 feet upstream and
downstream of the section.

In addition to uninterrupted conditions, the "ideal" basic freeway section is defined as having the
following characteristics:

Each lane is 12 feet wide.
There is 6 feet of clearance between the outside and the inside edges of the freeway and the
nearest obstruction that would distract or influence a motorist.
All vehicles are passenger cars (no trucks, buses, or recreational vehicles).
Ten or more lanes (in urban areas only).
Interchanges are spaced every 2 miles or more.
The drivers are regular and familiar users of the freeway section.
The terrain is level, with grades no greater than 2%.

Together, these conditions represent the "highest" (ideal) type of freeway section, which is one with
a free-flow speed of 70 mph or higher and a capacity of 2400 passenger cars per hour per lane
(pcphpl).

2.3.2. Free-Flow Speed and Flow Rate

An understanding of the relationship between speed and flow rate is the key to determining capacity
and LOS for a specific freeway section. In general, freeways are designed to accommodate
relatively large numbers of vehicles at higher speeds than other roadways.

Free-flow speed is the term used to describe the average speed that a motorist would travel if there
were no congestion or other adverse conditions (such as bad weather). The "highest" (ideal) type of
basic freeway section is one in which the free-flow speed is 70 mph or higher. Flow rate is defined
as the rate at which traffic traverses a freeway segment, in vehicles per hour or passenger cars per
hour.

Free-flow speed is actually defined as the speed that occurs when density and flow are zero. Of
course, observing zero density and flow doesnt make much sense. The following scenario
illustrates the relationship between Free-flow Speed and Flow Rate.

Imagine that you are the only motorist on a section of freeway that you travel frequently, the
weather is good and you are driving at a speed that is comfortable for that particular section, say 70
mph. Studies have shown that as long as the number of vehicles traveling per hour per lane on your
16

section of freeway does not exceed a flow rate of 1300, you will likely continue traveling at 70
mph. (This assumes all passenger cars - no trucks, buses or recreational vehicles). Your speed will
start to decrease once the flow rate exceeds 1300 passenger cars per hour per lane (approximately
22 cars per minute, or about 1 car every 3 seconds). If you were traveling at 65 mph, your speed
wouldnt decrease until a flow rate of 1450 passenger cars per hour per lane (pcphpl) has been
reached. The relationship is shown below.



2.3.3. Level of Service Criteria

Six levels of service have been defined for roadways and have been given letter designations of A
through F. LOS A represents the best level of service and LOS F represents the worst. The
following table lists the criteria for each LOS, based on the free-flow speed.

Level of
Service
Maximum Density
(pc/mi/ln)
Minimum Speed
(mph)
Maximum Service
Flow Rate (pcphpl)
Maximum (v/c)* Ratio
Free-Flow Speed = 70 mph
A 10 70.0 700 0.29
B 16 70.0 1,120 0.47
C 24 68.0 1,632 0.68
D 32 64.0 2,048 0.85
E 45 53.0 2,400 1.00
F var var var var
Free-Flow Speed = 65 mph
A 10 65.0 650 0.28
B 16 65.0 1,040 0.44
C 24 64.5 1,548 0.66
D 32 62.0 1,984 0.84
E 45 52.0 2,350 1.00
F Var var var var

17

(continuation of table which at previous page)
Level of
Service
Maximum Density
(pc/mi/ln)
Minimum Speed
(mph)
Maximum Service
Flow Rate (pcphpl)
Maximum (v/c)* Ratio
Free-Flow Speed = 60 mph
A 10 60.0 600 0.26
B 16 60.0 960 0.42
C 24 60.5 1,440 0.63
D 32 58.0 1,856 0.81
E 45 51.0 2,300 1.00
F var var var var
Free-Flow Speed = 55 mph
A 10 55.0 550 0.24
B 16 55.0 880 0.39
C 24 55.0 1,320 0.59
D 32 54.5 1,744 0.78
E 45 50.0 2,250 1.00
F var var var var
*See Terms and Definitions

To illustrate where each LOS falls with respect to speed and flow rate, the chart below shows speed
versus flow rate with corresponding levels of service A through E. LOS F lies beyond LOS E. The
value of the slope of each line that separates the levels of service is the maximum density for that
level of service.



For example, the line drawn for LOS E extends from the end of the free-flow speed lines to the
origin and has a slope of 45.0 pc/mi/ln. Service flow rate E is the value that corresponds to the
maximum flow rate, or capacity. Service flows C or D are usually used for most design or planning
purposes because these levels of service are more acceptable to roadway users. Note that the flow
rate at capacity for a free-flow speed of 70 mph is 2400 pcphpl. This capacity represents ideal
traffic and roadway conditions. Also note that the value of capacity varies with the free-flow speed.

18

2.3.4. Determining Flow Rate [d]

Determining the LOS for a basic freeway section involves two steps:

1. Adjusting a count or estimate of the hourly volume of vehicles to account for the effects of
prevailing traffic conditions. This module addresses step one.
2. Adjusting the free-flow speed for the prevailing design conditions of that section.

The module "Free-Flow Speed Adjustment" addresses step two.

The hourly volume (in vehicles per hour) is changed to an equivalent passenger-car flow rate by
allowing for the effects of heavy vehicles (buses, trucks and recreational vehicles) on traffic flow,
the variation of traffic flow during the hour, and the characteristics of the driver population. The
passenger-car equivalent flow rate is then reported on a per lane basis. Passenger-car equivalents in
passenger car per hour per lane (pcphpl) are determined using the following equation:

v
p
=
v
PBF N f
HV
f
P


Where

v
p
= 15-minute passenger-car equivalent flow rate (pcphpl)
V = hourly volume (vph)
PHF = peak-hour factor
f
HV
= heavy-vehicle adjustment factor
f
p
= driver population factor

Values for the driver population factor, fp, range from 0.85 to 1.0. In general, the value of 1.0 is
used to reflect commuter traffic. Use of a lower value reflects more recreational traffic. The peak
hour and heavy-vehicle adjustment factors are described in their respective modules. The
demonstration for this module uses values derived from the peak-hour and heavy-vehicle
adjustment factor modules.

2.3.5. Peak Hour Factor

Traffic engineers focus on the peak-hour traffic volume in evaluating capacity and other parameters
because it represents the most critical time period. And, as any motorist who travels during the
morning or evening rush hours knows, its the period during which traffic volume is at its highest.
The analysis of level of service is based on peak rates of flow occurring within the peak hour
because substantial short-term fluctuations typically occur during an hour. Common practice is to
use a peak 15-minute rate of flow. Flow rates are usually expressed in vehicles per hour, not
vehicles per 15 minutes. The relationship between the peak 15-minute flow rate and the full hourly
volume is given by the peak-hour factor (PHF) as shown in the following equation:

PBF =
Bouily volume
Peak iate of flow whitin the houi


If 15-minute periods are used, the PHF is computed as:

PBF =
v
4 v
15


Where
19


V = peak-hour volume (vph)
V
15
= volume during the peak 15 minutes of flow (veh/15 minutes)

Typical peak-hour factors for freeways range between 0.80 and 0.95. Lower factors are more
typical for rural freeways or off-peak conditions. Higher factors are typical of urban and suburban
peak-hour conditions.

2.3.6. Heavy Vehicle Adjustment Factor [d]

Determining the adjustment factor for the presence of heavy vehicles is a two-step process:

1. Finding passenger-car equivalents for trucks, buses and recreational vehicles
2. Using the equivalent values and the percentage of each type to compute the adjustment factor,
using the following equation:

f
HV
=
1
1 + P
T
(E
T
- 1) + P
R
(E
R
- 1)


Where

f
HV
= heavy-vehicle adjustment factor
E
T
= passenger-car equivalents for trucks and/or buses
E
R
= passenger-car equivalents for recreational vehicles
P
T
, P
R
= proportion of trucks or buses and RVs in the traffic stream.

Finding ET and ER

There are two methods for finding values of E
T
and E
R
and the choice of methods depends on the
freeway grade conditions.

Method 1: If an extended length of freeway contains a number of upgrades, downgrades and level
segments, but no one grade is long enough or steep enough to have a significant impact on traffic
operations, finding E
T
and E
R
is relatively straightforward. E
T
and E
R
are shown in the following
table for extended general segments where no one grade equal to or greater than 3 percent is
longer than 1/4 mile, or longer than 1/2 mile for grades less than 3 percent.

TYPE OF TERRAIN
CATEGORY LEVEL ROLLING MOUNTAINOUS
E
T
(Trucks and Buses Combined) 1.5 3.0 6.0
E
R
(RVs) 1.2 2.0 4.0

Method 2: There are three tables you can use to find E
T
and E
R
for isolated specific up and down
grades:

1. The first table is used to find the passenger car equivalent for trucks and buses on upgrades that
are more than 1/2 mile for grades less than 3 percent or 1/4 mile for grades of 3 percent or
more.
2. The second table is used to find the passenger car equivalent for recreational vehicles on
upgrades that are more than 1/2 mile for grades less than 3 percent or 1/4 mile for grades of 3
percent or more.
20

3. The third table is used to find the passenger car equivalent for trucks and buses on downgrades
that are more than 4 miles for grades of 4 percent or more.

2.3.7. Free-Flow Speed Adjustment [d]

Determining the LOS for a basic freeway section involves two steps:

1. Adjusting a count or estimate of the hourly volume of vehicles to account for the effects of
prevailing traffic conditions. See the module "Determining Flow Rate" for information on Step
One.
2. Adjusting the free-flow speed for the prevailing design conditions of that section. This module
addresses the second step.

The free-flow speed of a freeway section can be obtained directly by field measurement. If field
measurements are not feasible, the free-flow speed can be estimated by the following equation,
which accounts for the effects of physical characteristics:

FFS = 70 f
LW
f
LC
f
N
- f
ID


Where

FSS = estimated free-flow speed
f
LW
= adjustment for lane width
f
LC
= adjustment for right-shoulder lateral clearance
f
N
= adjustment for number of lanes
f
ID
= adjustment for interchange density

The adjustment factors can be obtained from the tables below.

Table 1. Adjustment Factors for Lane Width
Lane Width (ft) Reduction in Free-Flow Speed, f
LW
(mph)
12 0.0
11 2.0
10 6.5

Table 2. Adjustment Factors for Right-Shoulder Lateral Clearance
Right Shoulder Lateral Clearance (ft)
Reduction in Free-Flow Speed, f
LC
(mph)
Lanes in One Direction
2 3 4
6 0.0 0.0 0.0
5 0.6 0.4 0.2
4 1.2 0.8 0.4
3 1.8 1.2 0.6
2 2.4 1.6 0.8
1 3.0 2.0 1.0
0 3.6 2.4 1.2




21

Table 3. Adjustment Factors for Number of Lanes
Number of Lanes (One Direction) Reduction in Free-Flow Speed, f
N

5 0.0
4 1.5
3 3.0
2 4.5

Table 4. Adjustment Factors for Interchange Density
Interchanges per Mile Reduction in Free-Flow Speed, f
ID

0.50 0.0
0.75 1.3
1.00 2.5
1.25 3.7
1.50 5.0
1.75 6.3
2.00 7.5

2.3.8. Determining Level of Service and Density [d]

Once you have made the appropriate adjustments to the free-flow speed and have calculated the
equivalent passenger-car flow rate, determining LOS for a basic freeway section is as simple as
looking at the table given in the module "Level of Service Criteria and Capacity", or at the graph
below.



If your free-flow speed calculation resulted in a speed other than 70, 65, 60 or 55 mph, you would
construct the appropriate curve on the graph below. The curve would have the same general shape
as those shown and would intersect the y-axis at the estimated (or measured) free-flow speed. The
red line in the figure below presents an example of a curve drawn for an estimated free-flow speed
of 63 mph.

22



Vehicle density is calculated by the following equation:

B =
v
p
S


where

D = density (pc/mi/ln),
v
p
= flow rate (pcphpl), and
S = average passenger-car speed (mph).

2.3.9. Applications [d]

The methodology presented in the modules for this section may be used for the following
applications:

1. Operational analysis. Known or projected design and traffic variables are used to estimate LOS,
speed, and density of the traffic stream, as demonstrated in the module entitled "Determining
Level of Service and Density". This application is used to evaluate impacts of alternative
designs.
2. Design analysis. A forecasted demand volume, known design standards and a desired LOS are
used to determine the appropriate number of lanes needed for a basic freeway section. The
demonstration associated with this module uses a design approach.
3. Planning analysis. A desired LOS is used to determine the number of lanes needed. The
difference between a planning analysis and a design analysis is that the design standards and the
specifics of the demand volume may not be known. This application is beyond the scope of this
training module.

2.4. Professional Practice

In order to supplement your knowledge about the various concepts within Capacity and LOS
Analysis, and to give you a glimpse of how these various topics are discussed in the professional
environment, we have included selected excerpts from professional design manuals.
23


2.4.1. Basic Freeway Section and I deal Freeway Conditions

The following excerpt was taken from Chapter 3, page 3-1, of the 1997 revision of the Highway
Capacity Manual published by the Transportation Research Board.

A freeway may be defined as a divided highway with full control of access and two or more lanes
for the exclusive use of traffic in each direction.

Freeways provide uninterrupted flow. There are no signalized or stop-controlled at-grade
intersections, and direct access to and from adjacent property is not permitted. Access to and from
the freeway is limited to ramp locations. Opposing directions of flow are continuously separated by
a raised barrier, an at-grade median, or a raised traffic island.

Operating conditions on a freeway primarily result from interactions among vehicles and drivers in
the traffic stream and between vehicles and their drivers and the geometric characteristics of the
freeway. Operations can also be affected by environmental conditions, such as weather or lighting
conditions, by pavement conditions, and by the occurrence of traffic incidents.

2.4.2. Determining Flow Rate

The following excerpt was taken from Chapter 3, pages 3-14 and 3-15, of the 1997 revision of the
Highway Capacity Manual published by the Transportation Research Board.

The hourly flow rate must reflect the effects of heavy vehicles, the temporal variation of traffic flow
during an hour, and the characteristics of the driver population. These effects are reflected by
adjusting hourly volume counts or estimates, typically reported in vehicles per hour (vph), to arrive
at an equivalent passenger-car flow rate in passenger cars per hour (pcph). The equivalent
passenger-car flow rate is calculated using the heavy-vehicle and peak-hour adjustment factors and
is reported on a per lane basis, or in passenger cars per hour per lane...

Heavy Vehicle Adjustment Factor

Freeway traffic volumes that include a mix of vehicle types must be adjusted to an equivalent flow
rate expressed in passenger cars per hour per lane. This adjustment is made using the factor fHV.

Adjustments for the presence of heavy vehicles in the traffic stream apply for three vehicle types:
trucks, buses, and RVs. There is no evidence to indicate any differences in performance
characteristics between the truck and bus populations on freeways, so trucks and buses are treated
identically.

2.4.3. Free-Flow Speed and Flow Rate

The following excerpt was taken from Chapter 3, page 3-3, of the 1997 revision of the Highway
Capacity Manual published by the Transportation Research Board.

All recent freeway studies indicate that speed on freeways is insensitive to flow if the flow is low to
moderate. This is reflected in Figure 3-2 [reproduced in the Theory and Concept section entitled
Free-Flow Speed and Flow Rate], which shows speed to be constant for flows up to 1,300 pcphpl
for a 70-mph free-flow speed. For freeways with a lower free-flow speed, the region over which
speed is insensitive to flow extends to even higher flow rates. Thus, free-flow speed is easily
measured in the field as the average speed of passenger cars when flow rates are less than 1,300
24

pcphpl. Field determination of free-flow speed is easily accomplished by performing travel time or
spot speed studies during periods of low flows.

Note that although Figure 3-2 shows only curves for free-flow speeds of 75,70, 65, 60, and 55 mph,
curves representing any free-flow speed between 75 and 55 mph can be obtained by interpolation.
Also, the speed-flow curve representing a 75-mph [not shown in the figure included with this
chapter] free-flow speed, which corresponds with the recent increase in the posted speed limit on
many rural freeway sections throughout the United States, shown by a dashed line, is not based on
empirical field research but was created by extrapolation from the 70-mph free-flow speed curve.
Capacity at free-flow speeds greater than or equal to 70 mph is considered to be 2,400 pcphpl.

2.4.4. Free-Flow Speed Adjustment

The following excerpt was taken from Chapter 3, pages 3-4 and 3-5, of the 1997 revision of the
Highway Capacity Manual published by the Transportation Research Board.

Recent research has found that the free-flow speed on a freeway depends on the traffic and roadway
conditions present on a given facility. These conditions are described in the following sections.

Lane Width and Lateral Clearance

When lane widths are less than 12 feet, drivers are forced to travel closer to one another laterally
than they would normally desire. The effect of restricted lateral clearance is similar. When objects
are located too close to the edge of the median and roadside lanes, drivers in these lanes will shy
away from them, positioning themselves further from the lane edge. This restricted lateral clearance
has the same effect as narrow lanes: it forces drivers closer together laterally. Drivers have been
found to compensate by reducing their speed. The closeness of objects has been found to have a
greater effect on drivers in the right shoulder lane than on those in the median lane.

Drivers in the median lane appear to be unaffected by lateral clearance when minimum clearance is
2 feet, whereas drivers in the right shoulder lane are affected when lateral clearance is less than 6
feet...

Number of Lanes

The number of lanes on a freeway section influences free-flow speed. As the number of lanes
increases, so does the opportunity for drivers to position themselves to avoid slow-moving traffic.
In typical freeway driving, traffic tends to be distributed across lanes according to speed. Traffic in
the median lane or lanes typically moves faster than in the lane adjacent to the right shoulder...

2.4.5. LOS Criteria and Capacity

The following excerpt regarding LOS F was taken from Chapter 3, page 3-10, of the 1997 revision
of the Highway Capacity Manual published by the Transportation Research Board.

LOS F describes breakdowns in vehicular flow. Such conditions generally exist within queues
forming behind breakdown points. Such breakdowns occur for a number of reasons:

Traffic incidents cause a temporary reduction in the capacity of a short segment, so that the
number of vehicles arriving at the point is greater than the number of vehicles that can move
through it.
25

Points of recurring congestion exist, such as merge or weaving areas and lane drops where the
number of vehicles arriving is greater than the number of vehicles discharged.
In forecasting situations, any location where the projected peak-hour (or other) flow rate
exceeds the estimated capacity of the location presents a problem.

Note that in all cases, breakdown occurs when the ratio of demand to actual capacity or the ratio of
forecast demand to estimated capacity exceeds 1.00. Operations immediately downstream of such a
point, however, are generally at or near capacity, and downstream operations improve (assuming
that there are no additional downstream bottlenecks) as discharging vehicles move away from the
bottleneck.

2.4.6. Determining LOS and Density

The following excerpt was taken from Chapter 3, page 3-22, of the 1997 revision of the Highway
Capacity Manual published by the Transportation Research Board.

The level of service on a basic freeway section can be determined directly from Figure 3-4 (this
figure is presented in Theory and Concepts, Determining LOS and Density) on the basis of the free-
flow speed and the flow rate. The procedure is as follows:

Step 1. Define and segment the freeway section as appropriate.

Step 2. On the basis of the measured or estimated free-flow speed on the freeway segment,
construct an appropriate speed-flow curve of the same shape as the typical curves shown in Figure
3-2. The curve should intercept the y-axis at the free-flow speed.

Step 3. On the basis of the flow rate, v
p
, read up to the free-flow speed curve identified in Step 2
and determine the average passenger-car speed and level of service corresponding to that point.

Step 4. Determine the density of flow as B =
v

S


Where

D = density (pc/mi/ln),
v
p
= flow rate (pcphpl), and
S = average passenger-car speed (mph).

The level of service can also be determined using the density ranges provided in Table 3-1 [this
table is presented in Theory and Concepts, LOS Criteria].

2.5. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable roadway design concepts. The
more concerned you are about your understanding of a topic, the more seriously you will want to
approach the example problem for that topic.

2.5.1. Peak Hour Factor

The results of a traffic count taken between 5:00 p.m. and 6:00 p.m. are given below:

26

Time Interval Volume (vehicles)
5:00-5:15 p.m.
5:15-5:30 p.m.
5:30-5:45 p.m.
5:45-6:00 p.m.
900
1000
1200
850
5:00-6:00 p.m. 3950 total vehicles

The peak hour factor needs to be determined for this section of freeway.

Solution

The equation for calculating the peak hour factor for 15-minute periods is as follows:

PBF =
v
4 v
15


Where

V = peak-hour volume (vph) 3950
V
15
= volume during the peak 15 minutes of flow (veh/15 minutes) 1200

Therefore,

PBF =
S9Su
4 12uu
= u.82

2.5.2. Heavy Vehicle Adjustment Factor

A six lane freeway has a flow of 3500 vehicles. This flow consists of 180 trucks per hour, 200 RVs
per hour, 350 passenger buses per hour and the remainder of passenger. Calculate the heavy vehicle
adjustment factor for a mile section of this freeway that has a +4% grade.

Solution

The percentage of trucks is 5% [(180/3500) 100], buses is 10% [(350/3500) 100] and RVs is
6%. The heavy vehicle adjustment factor is calculated using the following equation:

f
HV
=
1
1 + P
T
(E
T
- 1) + P
R
(E
R
- 1)


E
T
= passenger-car equivalents for trucks and/or buses = 3.0 (see Passenger-Car Equivalents
For Trucks And Buses On Specific Upgrades excel demonstration)
E
R
= passenger-car equivalents for recreational vehicles = 2.0 (see Passenger-Car Equivalents
For Trucks And Buses On Specific Upgrades excel demonstration)
P
T
= proportion of trucks/buses in the traffic stream = 5 + 10 = 15%.
P
R
= proportion of RVs in the traffic stream = 6%

Therefore,

f
HV
=
1
1 + u.1S(S -1) +u.u6(2 - 1)
= u.74

27

2.5.3. Calculating Flow Rate

A four-lane highway (two lanes in each direction) has a 2,500-vph peak-hour volume that includes
mostly commuter traffic, 5% trucks and 6% buses. The section of highway that we are interested in
is in rolling terrain. The peak-hour factor has been determined by earlier studies to be 0.95. What is
the passenger-car equivalent (or service) flow rate for this section of freeway?

Solution

The equation for calculating the passenger-car equivalent flow rate is:

v
p
=
v
PBF N f
HV
f
P


V = peak hour volume (vph) = 2,500
PHF = peak-hour factor = 0.95
f
HV
= heavy-vehicle adjustment factor

f
HV
=
1
1 + P
T
(E
T
- 1) + P
R
(E
R
- 1)
=
1
1 +(u.uS + u.u6)(S -1) +u(2 - 1)
= u.82


(for E
T
and E
R
, use the small table for extended general freeway segments)

f
p
= driver population factor = 1.0 for commuter traffic
v
p
= 15-minute passenger-car equivalent flow rate (pcphpl)

v
p
=
2Suu
u.9S 2 u.82 1
= 1.6uS

2.5.4. Free-Flow Speed Adjustment

An existing six-lane freeway in an urban area has the following physical characteristics:

11-ft lanes
2-ft lateral clearance on outer shoulders
interchange density of 1 interchange per mile.

Calculate the free-flow speed for this section of freeway.

Solution

When actual field measurements arent available, free-flow speed is estimated by the following
equation, which utilizes the tables given in the Free-Flow Speed Adjustment theory and concept
subject:

FFS = 70 f
LW
f
LC
f
N
f
ID


where

f
LW
= adjustment for lane width = 2.0
f
LC
= adjustment for right-shoulder lateral clearance = 1.6
28

f
N
= adjustment for number of lanes = 3.0
f
ID
= adjustment for interchange density = 2.5
FSS = estimated free-flow speed = 70 2.0 1.6 3.0 2.5 = 60.9 mph

2.5.5. Determining LOS and Density

An existing freeway has the following characteristics:

8 lanes
Carries a flow of 4000 vph
Trucks make up 8% and RVs 2% of the flow
12-foot lane widths
Interchange density is less than 0.5 per mile
Obstructions within 4 feet of the outside edges of the freeway
PHF is 0.95

We are interested in determining the existing LOS and density as well as the maximum service flow
rate at capacity (LOS E) for a two-mile section of this freeway with a grade of +4%

Solution

In order to determine the existing LOS using the applicable graph or table, we need to calculate (1)
the service flow rate and (2) the adjusted free-flow speed.

1. The service (or passenger-car equivalent) flow rate calculation is as follows:

v
p
=
v
PBF N f
HV
f
P


V = hourly peak hour volume (vph) = 4,000
PHF = peak-hour factor = 0.95
f
HV
= heavy-vehicle adjustment factor =

f
HV
=
1
1 + P
T
(E
T
- 1) + P
R
(E
R
- 1)
=
1
1 +(u.u8)(6 - 1) + u.u2(S - 1)
= u.69

f
p
= driver population factor = 1.0 (assumed)
v
p
= 15-minute passenger-car equivalent flow rate (pcphpl)

v
p
=
4uuu
u.9S 4 u.69 1
= 1.S26

2. The free-flow speed calculation is as follows:

FFS = 70 f
LW
f
LC
f
N
- f
ID


Where

f
LW
= adjustment for lane width = 0
f
LC
= adjustment for right-shoulder lateral clearance = 0.4
f
N
= adjustment for number of lanes = 1.5
f
ID
= adjustment for interchange density = 0
29

FSS = estimated free-flow speed = 70 0 0.4 1.5 0 = 68.1 mph

We then draw the free-flow speed curve of 68.1 mph on the graph as shown below.



At the flow rate of 1526 pcphpl, the LOS is C. Density is calculated as:

B =
v
p
S


Where

v
p
= flow rate (pcphpl) = 1,526, and
S = average passenger-car speed (mph) = 67 (from the graph) and
D = density (pc/mi/ln) = 23.7 pc/mi/ln.

The flow at capacity can be taken from the graph above, at the end of the red line as 2380 pcphpl.

2.5.6. Design Application

A new 1-1/2 mile section of freeway is going to be built in an urban area with the following
characteristics:

5% grade
1.5 interchanges per mile
Uphill traffic volume of 3080 vehicles per hour
5% trucks, no buses and 2% RVs
Estimated peak hour factor if 0.95
Full shoulders
12-foot-wide lanes

How many lanes will be required to provide LOS C for the uphill direction? If we assume the same
traffic and design components, will the downhill lane requirement be the same?

Solution

We want to solve the following equation:
30


N =
v
(v
p
PBF f
HV
f
p
)


Where:

V = hourly volume (vph) = 3,080
PHF = peak-hour factor = 0.95
f
HV
= heavy-vehicle adjustment factor (equation shown below)
f
p
= driver population factor = 1.0 for commuter traffic

f
HV
=
1
1 + P
T
(E
T
- 1) + P
R
(E
R
- 1)


(for E
T
and E
R
use the applicable passenger-car equivalent tables)

For the uphill section,

f
HV
=
1
1 + (u.uS)(9 - 1) + u.u2(4.S -1)
= u.68

For the downhill section,

f
HV
=
1
1 + (u.uS)(1.S -1) +u.u2(1.2 - 1)
= u.97

The value for v
P
can be interpolated from the table or the graph given in the module entitled "Level
of Service Criteria and Capacity" after we've adjusted the free flow speed.

Adjusted Free-flow speed = FFS = 70 f
LW
f
LC
f
N
f
ID


Where

f
LW
= adjustment for lane width = 0
f
LC
= adjustment for right-shoulder lateral clearance = 0
f
N
= adjustment for number of lanes (we'll assume 3 lanes to begin with and come back to
check to see if it agrees with our final solution) = 3.0
f
ID
= adjustment for interchange density = 5
FFS = estimated free-flow speed = 70 0 0 3.0 5 = 62

By interpolation, the maximum service flow rate (v
P
) for LOS C at a free-flow speed of 62 mph is
1480 pcphpl.

Therefore, for the uphill section:

N = 3080 / (1480 0.95 0.68 1) = 3.2 or 4 lanes

Checking the free-flow speed: 70 0 0 1.5 5 = 63.5

And the maximum service flow rate for LOS C for a free-flow speed of 63.5 mph is approximately
1520 vph. Let's confirm that 4 lanes is still appropriate:

31

N = 3080 / (1520 0.95 0.68 1) = 3.1 or 4 lanes (always round up).

For the downhill section, and going back to the 3 lane assumption because the heavy vehicle
adjustment factor is quite a bit larger:

N = 3080 / (1480 0.95 0.97 1) = 2.3 or, rounding up, 3 lanes are needed in the downhill
direction.

2.6. Glossary
Basic Freeway Section: Freeway segments that are outside of the influence of ramps or weaving
sections.
Capacity: The maximum number of vehicles that can reasonably be expected to traverse over a
specific section of roadway, in one direction, during a given time period and under the prevailing
conditions. This is expressed in passengers cars per hour per lane (pcphpl).
Design Conditions: The physical qualities of a basic freeway section such as lane width, shoulder
clearances and density of interchanges (on and off ramps).
Density: The number of vehicles in a one-mile segment of one lane of traffic.
Flow rate: The rate, in vehicles per hour or passenger cars per hour, at which traffic traverses a
freeway segment.
Free-flow speed: The speed of traffic flow that is unaffected by upstream or downstream
conditions. Ideally, free-flow speed is the speed that occurs when density and flow are zero.
Level of Service (LOS): A measure of the operating conditions of a basic freeway section. There
are six categories A through F with F being the least desirable.
Peak-hour factor: The ratio of the number of vehicles during the peak hour to four times the
number of vehicles entering the traffic stream during the peak 15-minute period.
Traffic conditions: The qualities of traffic such as traffic speed, density, vehicle types and traffic
flow rate.
V/C ratio: The proportion of the facilitys capacity being utilized by current or projected traffic.
v/c = rate of flow/capacity.






32

3. Geometric Design

3.1. I ntroduction

There are an infinite number of ways to get from point A to point B. Geometric design is the aspect
of transportation engineering that deals with selecting the best path between those points. A good
geometric design will balance operational efficiency, comfort, safety, convenience, cost,
environmental impact, and aesthetics.

Geometric Design can be a difficult process, and many professionals rely on the assistance of
design manuals. The American Association of State Highway and Transportation Officials has
published a design manual entitled: A Policy on Geometric Design of Highways and Streets, which
provides engineers with a guide to geometric design that is based on years of practical experience
and research.

This chapter is designed to help the undergraduate engineer in his/her studies of the more
quantifiable aspects of geometric design. Our focus here is on the geometric aspect, or the actual
shape of the roadway.

3.2. Lab Exercises

This exercise will help increase your understanding of Geometric Design, by presenting a more
complicated problem that requires more thorough analysis.

3.2.1. Lab Exercise One: Geometric Design

A section of an existing two-lane rural highway will be modified, requiring the connection of two
intersecting one-percent grades with a vertical curve. (See Figure 1). You have been asked to
determine the length of the vertical curve necessary so that sufficient sight distance is provided
along this vertical curve for one vehicle to safely pass another.



This assignment is divided into several parts:

How much distance is required for one vehicle to safely pass another vehicle and return again to
the travel lane?
What is the overall sight distance required with consideration given to an oncoming vehicle?
What length of vertical curve is required to accommodate this sight distance?
How do your computations compare with the AASHTO design standard?
33


Tasks To Be Completed

As you complete the following tasks, you will determine the required sight distance for passing and
the required length of the vertical curve connecting the two grades.

Task 1. Consider the following situation. Vehicles are traveling along a two-lane rural highway. The
speed limit on the highway is 55 miles per hour. A passenger car is following a slower vehicle
traveling at 45 miles per hour. The passenger car would like to travel at the speed limit of 55 miles
per hour. Describe the tasks that must be performed by the driver of the passenger car in order to
safely pass the slower vehicle. These tasks should be defined in terms of both the decisions and the
maneuvers to be made by the driver, from the decision to pass to the maneuver back into the travel
lane.

Task 2. Identify the key events relating to this passing maneuver on a time-distance diagram. Plot
distance on the y-axis and time on the x-axis. Sketch the key events (without regard for
computations yet) relating both the lead (slower) vehicle and the following (faster) vehicle.

Task 3. Calculate the kinematic characteristics (position, velocity, and acceleration) of both the
passing and the passed vehicles for each stage of the passing maneuver that you identified in Task
1. What distance is traveled by each vehicle during these stages? What is the resultant position of
the vehicles at each stage? How much time does each stage consume?

Task 4. Now consider the effects of an oncoming vehicle. What distance is traveled by the
oncoming vehicle during the relevant stages of the passing maneuver? In determining this distance,
you should identify the point of the passing maneuver that can be described as the point of no return
for the passing driver.

Task 5. Using a spreadsheet program, prepare a time-distance plot of the three vehicles involved in
this passing maneuver. This should provide a visual check for you to make sure that your
assumptions and calculations are correct.

Task 6. Integrate the results of Tasks 1 through 5 above to calculate the required passing sight
distance for the passing vehicle. How do your results compare with the AASHTO model and design
information? Why do you suppose that these differences exist? (Note: the standards used by
AASHTO are based on certain kinematic assumptions and simplifications, as well as field
measurements performed a number of years ago; your standard should reflect both your own
reasoning regarding the passing maneuver as well as a clear documentation of this logic).

Task 7. The grades of the section for which you need to design the vertical curve are both one
percent. Using the relationships between sight distance and the length of the vertical curve, compute
the minimum length of the vertical curve required to accommodate this passing maneuver. If the
required length of the curve is more than 2000 feet, then a no-passing zone should be established for
the curve. If the length of the curve is less than 2000 feet, then a passing zone can be established.

Task 8. Prepare a brief report summarizing the results of your work.

Assumptions

The following data will be useful to you in this problem.

1. The acceleration rate for the passing vehicle is 1.47 miles per hour per second.
34

2. The perception/reaction time for a passing maneuver is 1 second.
3. Safe following distance is assumed to be 2 seconds.
4. Minimum clearance between the passing vehicle and the opposing vehicle is 1 second.

3.3. Theory and Concepts

A course in transportation engineering wouldn't be complete without discussing some elements of
Geometric Design. Most junior level courses cover several aspects of Geometric Design, including
the topics listed below. As these discussions are only meant to supplement your learning
experience, please don't forget to read your textbook. To begin learning about these Geometric
Design principles, just click on the link of your choice. Topics followed by the characters '[d]'
include an Excel demonstration.

3.3.1. Brake Reaction Time

The brake reaction time is the amount of time that elapses between the recognition of an object or
hazard in the roadway and the application of the brakes. The length of the brake reaction time varies
widely between individual drivers. An alert driver may react in less than 1 second, while other
drivers may require up to 3.5 seconds.

The brake reaction time depends on an extensive list of variables, including:

driver characteristics such as attitude, level of fatigue, and experience.
environmental conditions such as the clarity of the atmosphere and the time of day
the properties of the hazard or object itself, such as size, color and movement.

To make highways reasonably safe, the engineer must provide a continuous sight distance (see the
stopping sight distance module) equal to or greater than the stopping sight distance. As an integral
part of the stopping sight distance, a value for the brake reaction time must be assumed. Extensive
research has shown that 90% of the driving population can react in 2.5 seconds or less. The brake
reaction time normally used in design, therefore, is 2.5 seconds. The distance traveled during the
brake reaction time can be calculated by multiplying the vehicle's initial speed by the brake reaction
time.

Both the brake reaction time and the braking distance are used in the calculation of the stopping
sight distance. Therefore, it is suggested that you read the braking distance module before
proceeding to the stopping sight distance module.

3.3.2. Braking Distance [d]

The braking distance is the distance that a vehicle travels while slowing to a complete stop. The
braking distance is a function of several variables. First, the slope (grade) of the roadway will affect
the braking distance. If you are going uphill, gravity assists you in your attempts to stop and reduces
the braking distance. Similarly, gravity works against you when you are descending and will
increase your braking distance. Next, the frictional resistance between the roadway and your tires
can influence your braking distance. If you have old tires on a wet road, chances are you'll require
more distance to stop than if you have new tires on a dry road. The last parameter that we will
consider is your initial velocity. Obviously, the higher your speed the longer it will take you to stop,
given a constant deceleration.

The equation used to calculate the braking distance is a child of a more general equation from
classical mechanics. The parent equation is given below.
35


v
I
2
= v
0
2
+2 a u

Where:

V
f
= Final velocity
V
o
= Initial velocity
a = Acceleration rate
d = Distance traversed during acceleration

When calculating the braking distance, we assume the final velocity will be zero. Based on this, the
equation can be manipulated to solve for the distance traversed during braking.

u =
-v
0
2
2a


Notice that the distance will be positive as long as a negative acceleration rate is used.

The acceleration of a braking vehicle depends on the frictional resistance and the grade of the
road. From our knowledge of the frictional force, we know that the acceleration due to friction can
be calculated by multiplying the coefficient of friction by the acceleration due to gravity. Similarly,
we know from inclined plane problems that a portion of the car's weight will act in a direction
parallel to the surface of the road. The acceleration due to gravity multiplied by the grade of the
road will give us an estimate of the acceleration caused by the slope of the road.

The final formula for the braking distance is given below. Notice how the acceleration rate is
calculated by multiplying the acceleration due to gravity by the sum of the coefficient of friction
and grade of the road.

u =
v
2
2g(f + u)


Where:

d = Braking Distance (ft)
g = Acceleration due to gravity (32.2 ft/sec
2
)
G = Roadway grade as a percentage; for 2% use 0.02
V = Initial vehicle speed (ft/sec)
f = Coefficient of friction between the tires and the roadway

The braking distance and the brake reaction time are both essential parts of the stopping sight
distance calculations. In order to ensure that the stopping sight distance provided is adequate, we
need a more in-depth understanding of the frictional force. The value of the coefficient of friction is
a difficult thing to determine. The frictional force between your tires and the roadway is highly
variable and depends on the tire pressure, tire composition, and tread type. The frictional force also
depends on the condition of the pavement surface. The presence of moisture, mud, snow, or ice can
greatly reduce the frictional force that is stopping you. In addition, the coefficient of friction is
lower at higher speeds. Since the coefficient of friction for wet pavement is lower than the
coefficient of friction for dry pavement, the wet pavement conditions are used in the stopping sight
distance calculations. This provides a reasonable margin of safety, regardless of the roadway
surface conditions. The table below gives a few values for the frictional coefficient under wet
roadway surface conditions (AASHTO, 1984).
36


Design Speed (mph) Coefficient of Friction (f)
20 0.40
30 0.35
40 0.32
60 0.29

3.3.3. Stopping Sight Distance [d]

(Note: If you feel uncomfortable with your understanding of brake reaction time or braking
distance, you might want to review those topics before continuing with Stopping Sight Distance.)

The stopping sight distance is the sum of the braking distance and the distance traversed during the
brake reaction time. In other words, it is the length of roadway that should be visible ahead of you,
in order to ensure that you will be able to stop if there is an object in your path.

For example, let us say that you are negotiating a horizontal curve in a highway when you notice an
object 200 feet ahead of you. If the distance you travel during your brake reaction time is 100 feet
and your braking distance is 130 feet, you will not be able to avoid the collision. If the horizontal
curve were not as tight, you would be able to see the object at a distance of 250 feet, which would
allow you to stop 20 feet short of the object. A properly designed roadway will provide the
minimum stopping sight distance at every point along its length.

In order to calculate the actual sight distance based on the geometry of the roadway, some
assumptions are necessary. The main assumptions are the height of the driver's eyes above the
roadway surface and the height of the object or hazard. In geometric design, these values are 3.5
feet and 0.5 feet, respectively. This represents a reasonable worst-case scenario.

To include the stopping sight distance in your design, calculate the stopping sight distance for a
vehicle traveling on your roadway at the design speed, and then make sure the actual sight distance
that you provide is at least as great as the stopping sight distance.

Trucks and Busses

Trucks and busses require longer braking distances than passenger cars, but their stopping sight
distances are not considered in most designs. This is because the driver's eyes are higher and their
sight distance is consequently increased. The drivers of these vehicles also tend to be more
experienced and more alert. The net effect is that large vehicles can avoid obstacles even though the
road was not specifically designed with them in mind. The engineer must decide when large
vehicles may need extra sight distances and provide these distances where necessary.

3.3.4. Decision Sight Distance

Normally, the stopping sight distance is an adequate sight distance for roadway design. However,
there are cases where it may not be appropriate. In areas where information about navigation or
hazards must be observed by the driver, or where the drivers visual field is cluttered, the stopping
sight distance may not be adequate. In addition, there are avoidance maneuvers that are far safer
than stopping, but require more planning by the driver. These may not be possible if the minimum
stopping sight distance is used for design. In these instances, the proper sight distance to use is the
decision sight distance.

37

The decision sight distance is the distance traversed while recognizing an object or hazard, plotting
an avoidance course, and making the necessary maneuvers. Unlike the stopping sight distance, the
decision sight distance is quite complex. Various design values for the decision sight distance have
been developed from research. The table below gives a few values for the decision sight distance
(AASHTO, 1994).

Design Speed
(km/h)
Decision Sight Distance (meters)
Stop Rural
Road
Stop Urban
Road
Adjustment
Rural Road
Adjustment
Suburban Road
Adjustment
Urban Road
50 75 160 145 160 200
80 155 300 230 275 315
90 185 360 275 320 360
110 265 455 335 390 435

It is up to the engineer to decide when to use the decision sight distance. Providing the extra sight
distance will probably increase the cost of a project, but it will also increase safety. The decision
sight distance should be provided in those areas that need the extra margin of safety, but it isnt
needed continuously in those areas that don't contain potential hazards.

3.3.5. Passing Sight Distance [d]

While passing is not an event that is a major factor in the design of four-lane highways, it is a
critical component of two-lane highway design. The capacity of a two-lane roadway is greatly
increased if a large percentage of the roadway's length can be used for passing. On the other hand,
providing a sufficient passing sight distance over large portions of the roadway can be very
expensive.

Simply put, the passing sight distance is the length of roadway that the driver of the passing vehicle
must be able to see initially, in order to make a passing maneuver safely.

Our real goal is to provide most drivers with a sight distance that gives them a feeling of safety and
that encourages them to pass slower vehicles.

Calculating the passing sight distance required for a given roadway is best accomplished using a
simple model. The model that is normally used incorporates three vehicles, and is based on six
assumptions:

1) The vehicle being passed travels at a constant speed throughout the passing maneuver.
2) The passing vehicle follows the slow vehicle into the passing section.
3) Upon entering the passing section, the passing vehicle requires some time to perceive that the
opposing lane is clear and to begin accelerating.
4) While in the left lane, the passing vehicle travels at an average speed that is 10 mph faster than
the vehicle being passed.
5) An opposing vehicle is coming toward the passing vehicle.
6) There is an adequate clearance distance between the passing vehicle and the opposing vehicle
when the passing vehicle returns to the right lane.

Under these assumptions, the passing sight distance can be divided into four quantifiable portions:

d
1
- The distance the passing vehicle travels while contemplating the passing maneuver, and while
accelerating to the point of encroachment on the left lane.
38

d
2
- The length of roadway that is traversed by the passing vehicle while it occupies the left lane.
d
3
- The clearance distance between the passing vehicle and the opposing vehicle when the passing
vehicle returns to the right lane.
d
4
- The distance that the opposing vehicle travels during the final 2/3 of the period when the
passing vehicle is in the left lane.

Because the purpose of these specific distances might not be obvious at this point, a short
discussion of each of these distances can be found below. In addition, figure 1.0 below gives a
graphical explanation of these distances.


Figure 1.0: Diagram of Passing Sight Distance Components
Source: AASHTO, 1994

d
1
- The perception-reaction-acceleration distance isn't hard to understand or to justify. The only
aspect of this distance that might be confusing is the simultaneous nature of the perception and
acceleration. Some drivers will begin accelerating before they enter the passing section and will
continue to accelerate while they scan the opposing lane for traffic. These drivers tend to accelerate
at a reduced rate. Other drivers will avoid accelerating until they have determined that the opposing
lane is clear, but they will accelerate at a higher rate once they have decided to pass. The net effect
is that the perception-reaction-acceleration distance is identical for both types of drivers. The
distance d
1
and the corresponding time t
1
were measured for several different passing vehicle
speeds. More recent research has confirmed that the accepted values are conservative. See table 1.0.

d
2
- The distance traveled during the occupancy of the left lane is also easy to understand. Since the
speed of the passing vehicle was assumed to be 10 mph faster than the overtaken vehicle, all we
need to know to calculate the distance d
2
is the time that the passing vehicle occupies the left
lane. Values for this time interval were measured for several different passing vehicle speeds. These
measured values were then used to develop design values for d
2
. See table 1.0.

d
3
- The clearance distance might not seem necessary at first, but for now lets take it on faith that an
opposing vehicle is necessary. If this is the case, a maneuver that feels safe will require that a
certain length of roadway is present between the passing vehicle and the opposing vehicle when the
passing vehicle returns to the relative safety of the right lane. The clearance distance that drivers
require depends on their personality. A timid driver might require several hundred feet of clearance
distance, while a more aggressive driver might consider exchanging side mirrors a perfectly
acceptable practice. Studies have shown that the clearance distance is normally between 100 and
300 feet. See table 1.0.

d
4
- The opposing vehicle encroachment distance is the distance that seems to be the most troubling
for students. Let us picture a passing section that is terminated by a sharp reduction in grade, which
prevents the passing driver from seeing any vehicles beyond the end of the passing section. Let us
also assume that the length of the passing section is equal to the sum of the distances d
1
and d
2
. Our
39

passing vehicle driver could pass the slower vehicle before leaving the passing section, but she
might be flirting with destiny in doing so. Her principal problem is that she can't see if there are any
opposing vehicles beyond the passing section that might conflict with her during the maneuver.

The question now is, how much extra sight distance would she need to feel secure that an opposing
vehicle would not conflict with her while she is in the left lane? If we assume that she can abort her
maneuver if an opposing vehicle appears during the interval t
1
or during the first third of the interval
t
2
, we can reduce the sight distance that we need to provide.

Lets say that we make the passing section length equal to the passing sight distance as defined in
reality (d = d
1
+ d
2
+ d
3
+ d
4
). If an opposing vehicle appears just after the first third of the interval
t
2
is over, the passing car can still safely pass the slower car and return to the right lane before the
opposing car becomes a threat. This is because the opposing vehicle is a distance 2/3 d
2
+d
3
+ d
4

away from the passing vehicle. By the time that the passing vehicle has traveled the remaining
2/3 d
2
and returned to the right lane, the opposing car will have traveled d
4
, and the clearance
distance d
3
will separate them. This is why we add the distances d
3
and d
4
to the passing sight
distance. The distance d
4
is calculated by multiplying the speed of the opposing vehicle (normally
assumed to be the speed of the passing vehicle) by 2/3 t
2
.

The table below summarizes the results of field observations directed toward quantifying the
various aspects of the passing sight distance (AASHTO, 1994).

Speed Group (km/h) 50-65 66-80 81-95 96-110
Average Passing Speed (km/h) 56.2 70.0 84.5 99.8
I nitial Maneuver:
Average acceleration (km/h/s) 2.25 2.30 2.37 2.41
Time (s) 3.6 4.0 4.3 4.5
Distance Traveled (m) 45 65 90 110
Occupation of the Left Lane:
Time (s) 9.3 10.0 10.7 11.3
Distance Traveled (m) 145 195 250 315
Clearance Length:
Distance Traveled (m) 30 55 75 90
Opposing Vehicle:
Distance Traveled (m) 95 130 165 210
Total Distance (m) 315 445 580 725

Now that we know how to calculate the required passing sight distance, how do we calculate the
actual passing sight distance that we have provided in our geometric design? To do this, we simply
assume that the driver's eyes are at a height of 3.5 ft from the road surface and the opposing vehicle
is 4.25 ft tall. The actual passing sight distance is the length of roadway ahead over which an object
4.25 ft tall would be visible, if your eyes were at an elevation of 3.5 ft.

40

3.3.6. Horizontal Alignment

Horizontal alignment is a broad term that encompasses several aspects of transportation
engineering. In this discussion, we will focus on the design of horizontal curves.

The key steps in the design of horizontal curves are listed below. This will serve as a guide, as you
explore the remaining topics within the horizontal alignment section of geometric design.

1. Determine a reasonable maximum superelevation rate.
2. Decide upon a maximum side-friction factor.
3. Calculate the minimum radius for your horizontal curve.
4. Iterate and test several different radii until you are satisfied with your design.
5. Make sure that the stopping sight distance is provided throughout the length of your
curve. Adjust your design if necessary.
6. Design the transition segments.

Now that you have a feel for the general design procedure, you can begin to explore the various
steps within the procedure in greater detail. Feel free to return to this guide whenever you wish - it
isn't hard to lose sight of the greater plan while battling the details.

3.3.7. Superelevation and Side-Friction

Most highways will change directions several times over the course of their lengths. These changes
may be in a horizontal plane, in a vertical plane, or in both. The engineer is often charged with
designing curves that accommodate these transitions, and consequently must have a good
understanding of the physics involved.

The superelevation of the highway cross-section and the side-friction factor are two of the most
crucial components in the design of horizontal curves. The superelevation is normally discussed in
terms of the superelevation rate, which is the rise in the roadway surface elevation as you move
from the inside to the outside edge of the road. For example, a superelevation rate of 10% implies
that the roadway surface elevation increases by 1 ft for every 10 ft of roadway width. The side-
friction factor is simply the coefficient of friction between the design vehicle's tires and the
roadway.

Whenever a body changes directions, it does so because of the application of an unbalanced force.
In the special case of a body moving in a circular path, the force required to keep that body
traveling in a circular path is called the centripetal force. When vehicles travel over a horizontal
curve, it is this centripetal force that keeps the vehicles from sliding to the outside edge of the curve.
In the simplest case, where the road is not banked, the entire centripetal force is provided by the
friction between the vehicle's tires and the roadway. If we add some side-slope or superelevation to
the cross-section of the roadway, some of the centripetal force can be provided by the weight of the
car itself.

High rates of superelevation that make cornering more comfortable during the summer by requiring
less frictional force, can make winter driving ponderous by causing slow-moving vehicles to slip
downhill toward the inside of the curve. Because of this, there are practical maximum limits for the
rate of superelevation. In areas where ice and snow are expected, a superelevation rate of 8% seems
to be a conservative maximum value. In areas that are not plagued by ice and snow, a maximum
superelevation rate of 10-12% seems to be a practical limit. Both modern construction techniques
and driver comfort limit the maximum superelevation rate to 12%.

41

The side-friction factor has practical upper limits as well. As was discussed in the braking distance
module, the coefficient of friction is a function of several variables, including the pavement type
and the vehicle speed. In every case, the side-friction factor that is used in design should be well
below the side-friction factor of impending release. In addition to the safety concerns, drivers don't
feel comfortable if the roadway seems to rely heavily on the frictional force. Several studies aimed
at determining the maximum side-friction factors that are comfortable for drivers have been
conducted. Some of the results from these studies are tabulated below. (AASHTO, 1994)

Speed (km/h) Comfortable Side-Friction Factor
40 0.21
50 0.18
55-80 0.15
> 110 < 0.10

The side-friction factors that are employed in the design of horizontal curves should accommodate
the safety and comfort of the intended users.

The module on horizontal curve minimum radii will bring the effects of the superelevation rate and
the side-friction factor together. Both of these concepts contribute to the final alignment of
horizontal curves.

3.3.8. Minimum Radius Calculations [d]

Calculating the minimum radius for a horizontal curve is based on three factors: the design speed,
the superelevation, and the side-friction factor (see superelevation and sidefriction factor modules).
The minimum radius serves not only as a constraint on the geometric design of the roadway, but
also as a starting point from which a more refined curve design can be produced.

For a given speed, the curve with the smallest radius is also the curve that requires the most
centripetal force. The maximum achievable centripetal force is obtained when the superelevation
rate of the road is at its maximum practical value, and the side-friction factor is at its maximum
value as well. Any increase in the radius of the curve beyond this minimum radius will allow you to
reduce the side-friction factor, the superelevation rate, or both.

Using the equations for circular motion, friction, and inclined plane relationships, the following
equation has been derived.

R
mIn
=
v
2
127 [
e
max
1uu
+ f
max



Where:

R
min
= Minimum radius of the curve (m)
V = Design velocity of the vehicles (km/h)
e
max
= Maximum superelevation rate as a percent
f
max
= Maximum side-friction factor

This equation allows the engineer to calculate the minimum radius for a horizontal curve based on
the design speed, the superelevation rate, and the side friction factor.

42

3.3.9. Design I terations

In many ways, horizontal alignment is an art form. The goal is to produce a horizontal curve that is
comfortable and safe to use, and also cost efficient and aesthetically pleasing. The first step is to
calculate the radius of the horizontal curve. We can calculate the radius for any combination of
superelevation and side-friction factors using the equation below.

R =
v
2
127 [
e
1uu
+f


Where:

R = The radius of the curve (m)
V = Design velocity of the vehicles (km/h)
e = Cross-section superelevation rate as a percent
f = Side-friction factor

As long as the radius of your curve is above the minimum radius as described in the minimum
radius module, and as long as you haven't exceeded the practical values for the superelevation or for
the side-friction factor, you know that your design is acceptable.

You will probably need to test several different curve radii before you select your final
design. While iterating, you also need to consider other factors: the cost, environmental impacts,
sight distances, and, of course, the aesthetic consequences of your curve. Most surveying books
contain a complete chapter on the layout of horizontal curves, and consequently, we won't delve
into the surveying issues. Please refer to your surveying texts for this information.

3.3.10. Horizontal Curve Sight Distance [d]

Once you have a radius that seems to connect the two previously disjointed sections of roadway
safely and comfortably, you need to make sure that you have provided an adequate stopping sight
distance throughout your horizontal curve.

Sight distance can be the controlling aspect of horizontal curve design where obstructions are
present near the inside of the curve. To determine the actual sight distance that you have provided,
you need to consider that the driver can only see the portion of the roadway ahead that is not hidden
by the obstruction. In addition, at the instant the driver is in a position to see a hazard in the
roadway ahead, there should be a length of roadway between the vehicle and the hazard that is
greater than or equal to the stopping sight distance. See figure 1.0 below.


Figure 1.0: Sight Distance

43

Because the sight obstructions for each curve will be different, no general method for calculating
the sight distance has been developed. If you do have a specific obstruction in mind, however, there
is an equation that might be helpful. This equation involves the stopping sight distance, the degree
of the curve, and the location of the obstruction.

N =
S7Su
B
_1 - cos
SB
2uu
]

Where:

M = Distance from the center of the inside lane to the obstruction (ft.)
D = Degree of the curve. Where R = 5730 / D
S = Stopping sight distance (ft)
R = Radius of the curve (ft)

Once your rough design has been adjusted to accommodate the sight distance restrictions, and you
are satisfied with the aesthetic and financial consequences of your superelevation scheme, you can
begin to polish your design into its final form.

3.3.11. Transition Segments

Often, horizontal curves are more comfortable and more aesthetically pleasing if the change in
roadway cross-section and curvature is effected over a short transitional segment.

The gradual change in curvature is produced by using a spiral curve. The radius of the spiral curve
starts at infinity and is gradually reduced to the radius of the circular curve that you designed
originally. Adding the spiral curve causes the centripetal acceleration to build up gradually, which is
more comfortable for vehicle occupants. The equation commonly employed to calculate the
minimum length of the spiral transition segment is given below.

L =
S.1S v
3
R C


Where:

L = Minimum length of the spiral curve (ft)
v = Speed (mph)
R = Circular curve radius (ft)
C = Centripetal acceleration development rate (usually between 1 and 3 ft/sec
3
)

The other purpose of the transition segment is to gradually change the cross-section of the roadway
from normal to superelevated. This can be accomplished by rotating the cross-section around any
line parallel to the roadway. The engineer should keep water drainage in mind while considering all
of the available cross-section options. Opinions vary as to how fast the pavement cross-section
should change, but most people agree that the change in curvature and the change in cross-section
should occur in concert with one another. You should look at your local geometric design policy for
more specific information regarding transition segments.

3.3.12. Vertical Alignment

The topics discussed under vertical alignment can be divided into two categories: the design of
highway sections that have ascending or descending grades, and the design of vertical curves that
connect these segments of ascending and descending grades.
44


Grade is a measure of the inclination, or slope, of the roadway. It is defined as the rise over the run.
In other words, a 10% grade simply means that the elevation of the roadway increases by 10 feet for
every 100 feet of horizontal distance. The issues that surround the design of inclined roadway
sections revolve around safety and level of service.

Vertical curves, however, are slightly more complicated. The best feature of the vertical curve, in its
purest form, is that it doesn't require any changes in the roadway cross-section. In this respect,
vertical curves are easier to design than horizontal curves. On the other hand, vertical curves have a
parabolic shape instead of the simple circular shape of the pure horizontal curve. Because this
makes certain calculations more involved, we will spend more time explaining issues that we would
otherwise leave for surveying texts. The general discussion regarding vertical curves covers the
geometry of simple vertical curves. The discussions pertaining to sag and crest vertical curves
include more specific information related to the design of these curves.

3.3.13. Ascending Grades

Efficiency and safety govern the design of ascending grades. Research has shown that the frequency
of collisions increases dramatically when vehicles traveling more than 10 mph below the average
traffic speed are present in the traffic stream. This 10 mph differential is, therefore, a bounding
value in the design of ascending grades.

Research has also shown that most passenger cars are essentially unaffected by grades below 4-5%.
Large commercial vehicles and recreational vehicles, on the other hand, are extremely sensitive to
changes in grade. Design engineers do have some basic guidelines regarding the maximum upgrade
for certain design speeds. Some of these recommended values are tabulated below (AASHTO,
1994). Note that these maximum upgrades are tolerable but not desirable, so they should not be
used as targets for design.

Design Speed (km/h) Maximum Grade (%)
50 7-8
60 intermediate
80 intermediate
110 5

Lets return to our discussion of the speed differential. For each grade, there is a critical length at
which the design vehicles (trucks, RVs) will obtain the 10 mph differential. The figure below can be
used to find the critical length for some common grades (AASHTO, 1984).


45


As long as the length of your ascending grade is below the critical length, you will be able to
maintain a reasonable level of safety, and large vehicles will not aggravate the traffic flow.

The general design process is this: design your roadway so that the ascending grades achieve the
necessary change in elevation while not violating the maximum grade guidelines and not reducing
the speed of trucks to more than 10 mph below the traffic's running speed. This can be done any
number of ways, including a stepped approach with level sections between grades.

While the standards above should be the design goal, it is not always economically or physically
possible to meet them. In these cases, climbing lanes may relieve some of the excess restriction.
Climbing lanes are extra lanes that are reserved for slow vehicles. They allow faster vehicles to
overtake slow vehicles safely and therefore increase the level of service for the highway. According
to AASHTO, a climbing lane can be justified if all three of the criteria below are satisfied.

1. Upgrade traffic flow rate in excess of 200 vehicles per hour.
2. Upgrade truck flow rate in excess of 20 vehicles per hour.
3. One of the following conditions exists:
A 15 km/h or greater speed reduction is expected for a typical heavy truck.
Level-of-service E or F exists on the grade.
A reduction of two or more levels of service is experienced when moving from the approach
segment to the grade.

Climbing lanes are becoming more and more common on two-lane highways. They are rarely used
on multilane and divided-multilane highways, because these roadways currently accommodate the
casual passing of slow vehicles.

3.3.14. Descending Grades

Descending grades pose an entirely new set of problems for the design engineer. Instead of
worrying about reductions in speed, the engineer must be concerned about unbridled increases in
speed. The potential consequences of runaway vehicles are evident when you consider a highly
populated area that is located at the base of a long, steep grade. To avoid catastrophes created by
runaway vehicles, runaway vehicle ramps are often designed and included at critical locations along
the grade.

The location of runaway ramps depends on the geometry of the roadway and the topography of the
surrounding terrain. Logically, a ramp should exist before each turn that cannot be negotiated at
runaway speeds. Ramps should also be placed along straight stretches of roadway, wherever
unreasonable speeds might be obtained. In addition, the ramps should be located on the right side of
the road whenever possible, because opposing traffic or other vehicles may not realize that the truck
is in trouble and be able to yield in time.

Runaway ramps are normally designed so that they can stop a truck moving at a speed of at least 80
mph. Note that this is only a minimum speed and extra ramp length should be provided if the
potential for greater entrance speeds exists. The ramp should also be wide enough to service more
than one vehicle at a time. There are several different ramp types that can be employed to stop
runaway vehicles. Figure 1.0 illustrates the various types of ramps that are commonly used.

46


Figure 1.0: Common Runaway Vehicle Ramps

Along with the runaway ramp considerations, the engineer should give some thought to including a
slow vehicle lane on downgrades. Trucks often use their lower gears and crawl down descending
grades, to minimize the use of their brakes. Including a slow vehicle lane will provide faster traffic
with a safe path for overtaking slow vehicles, and the extra lane may also provide endangered
vehicles with an escape route if they discover a runaway in their rear-view mirror.

3.3.15. Vertical Curves

In highway design, most vertical curves are equal-tangent curves, which mean that the horizontal
distance from the center of the curve to the end of the curve is identical in both directions. Unequal-
tangent vertical curves, which are simply equal-tangent curves that have been attached to one
another, are used only infrequently. Because of its overwhelming popularity, we will limit our
discussion to the geometry of the equal-tangent parabolic curve.

In highway design, the grades of the disjointed segments of roadway are normally known before
any vertical curve calculations are initiated. In addition, the design speed of the roadway, the
stopping sight distance, and the decision sight distance are also well established. The first step in the
design of a vertical curve is the calculation of the curve length, which is the length of the curve as it
would appear when projected on the x-axis. (See figure 1.0 below). Because the stopping sight
distance should always be adequate, the length of the curve is normally dependent upon the
stopping sight distance. Occasionally, as with any other section of a highway, the decision sight
distance is a more appropriate sight distance. In these instances, the decision sight distance governs
the length of the vertical curve. The curve length calculations are slightly different for sag and crest
vertical curves, and are covered separately in those sections of this chapter.

Lets assume that you have already calculated the appropriate length (L) for your curve. At this
point you would probably want to develop the actual shape of the curve for your design
documents. Refer to figure 1.0 throughout the following discussion.

The first step in developing the profile for your curve is to find the center of your curve. The
location of the center-point is where the disjointed segments of the roadway would have intersected,
had they been allowed to do so. In other words, draw lines tangent to your roadway segments and
see where those lines intersect. This intersection is normally called the vertical point of intersection
(VPI).

47


Figure 1.0: Vertical Curve

The vertical point of curvature (VPC) and the vertical point of tangency (VPT) are located a
horizontal distance of L/2 from the VPI. The VPC is generally designated as the origin for the curve
and is located on the approaching roadway segment. The VPT serves as the end of the vertical curve
and is located at the point where the vertical curve connects with the departing roadway segment. In
other words, the VPC and VPT are the points along the roadway where the vertical curve begins
and ends.

One you have located the VPI, VPC, and VPT, you are ready to develop the shape of your
curve. The equation that calculates the elevation at every point along an equal-tangent parabolic
vertical curve is shown below.

Y = vPCy +B - x +
A - x
2
2uu - L


Where:

Y = Elevation of the curve at a distance x from the VPC (ft)
VPCy = Elevation of the VPC (ft)
B = Slope of the approaching roadway, or the roadway that intersects the VPC
A = The change in grade between the disjointed segments (From 2% to -2% would be a change
of -4% or -4)
L = Length of the curve (ft)
x = Horizontal distance from the VPC (ft) (Varied from 0 to L for graphing.)

At this point, you have everything that you need to develop the shape of a simple equal-tangent
vertical curve. The procedure above will work for both sag and crest vertical curves.

3.3.16. Crest Vertical Curves [d]

Crest vertical curves are curves that connect inclined sections of roadway, forming a crest, and they
are relatively easy to design. As you know from the module entitled Vertical Curves, we only
need to find an appropriate length for the curve that will accommodate the correct sight
distance. The stopping sight distance is usually the controlling sight distance, but the decision sight
distance or even the passing sight distance could be used if desired. The passing sight distance is
rarely ever used as the design sight distance, because it demands long, gentle curvatures that are
expensive and difficult to construct.

The sight distance and the length of the curve can be related to each other in one of two ways. The
first possibility is that the sight distance is less than the length of the curve. Alternatively, the length
of the curve could be less than the sight distance. See figure 1.0.

48


Figure 1.0: Sight Distance Possibilities

In any case, there are equations that relate these two parameters to the change in grade for both
possible conditions. The designer must double-check that the equation that is used agrees with its
own assumptions. For example, if the equation that is based on sight distances that are less than the
curve length produces a curve length that is less than the sight distance, you know that the result is
invalid. The equations that are normally used to calculate the lengths of crest vertical curves are
given below.

If S > L then

L = 2S -
2uu(h
1
+ h
2
)
2
A


If S < L then

L =
A S
2
1uu(2h
1
+2h
2
)
2


Where:

L = Length of the crest vertical curve (ft)
S = Sight distance (ft)
A = The change in grades (|G
2
-G
1
| as a percent)
h
1
= Height of the driver's eyes above the ground (ft)
h
2
= Height of the object above the roadway (ft)

The heights in the calculations above should be those that correspond to the sight distance of
interest. For the stopping sight distance, h
1
= 3.5 ft and h
2
= 0.5 ft. For the passing sight distance,
h
1
= 3.5 ft and h
2
= 4.25 ft.

While the sight distance has been portrayed as the only parameter that affects the design of vertical
curves, this isn't entirely true. Vertical curves should also be comfortable for the driver,
aesthetically pleasing, safe, and capable of facilitating proper drainage. In the special case of crest
vertical curves, it just so happens that a curve designed with adequate sight distances in mind is
usually aesthetically pleasing and comfortable for the driver. In addition, drainage is rarely a special
concern for crest vertical curves.

3.3.17. Sag Vertical Curves [d]

Sag vertical curves are curves that connect descending grades, forming a bowl or sag. Designing
them is is very similar to the design of crest vertical curves. Once again, the sight distance is the
49

parameter that is normally employed to find the length of the curve. When designing a sag vertical
curve, however, the engineer must pay special attention to the comfort of the drivers. Sag vertical
curves are characterized by a positive change in grade, which means that vehicles traveling over sag
vertical curves are accelerated upward. Because of the inertia of the driver's body, this upward
acceleration feels like a downward thrust. When this perceived thrust and gravity combine, drivers
can experience discomfort.

The length of sag vertical curves, which is the only parameter that we need for design, is
determined by considering drainage, driver comfort, aesthetics, and sight distance. Once again, the
aesthetics and driver comfort concerns are normally automatically resolved when the curve is
designed with adequate sight distance in mind. Driver comfort, for example, requires a curve length
that is approximately 50% of the curve length required for the sight distance. Drainage may be a
problem if the curve is quite long and flat, or if the sag is within a cut. For more information on
these secondary concerns, see your local design manuals.

The theory behind the sight distance calculations for sag vertical curves is only slightly different
from that for crest vertical curves. Sag vertical curves normally present drivers with a commanding
view of the roadway during the daylight hours, but unfortunately, they truncate the forward spread
of the driver's headlights at night. Because the sight distance is restricted after dark, the headlight
beams are the focus of the sight distance calculations. For sight distance calculations, a 1 upward
divergence of the beam is normally assumed.

In addition, the headlights of the vehicle are assumed to reside 2 ft above the roadway surface. As
with crest vertical curves, these assumptions lead to two possible configurations, one in which the
sight distance is greater than the curve length, and one in which the opposite is true. The figure
below illustrates these possibilities.



As with crest vertical curves, each possibility has a different design equation. All that you need to
do, therefore, is make sure that the results from the equation that you use are consistent with that
equation's assumptions. For example, if you employ the equation that assumes the sight distance is
greater than the curve length, you should make sure that the resulting curve length is less than the
sight distance. The equations for each possibility are given below.

If S > L then

L = 2S -
2uu(B + S tanB)
A


If S < L then

L =
A S
2
2uu(B +S tanB)

50


Where:

L = Curve length (ft)
S = Sight distance (ft) (normally the stopping sight distance)
B = Beam upward divergence () (normally assumed as 1)
H = Height of the headlights (ft) (normally assumed as 2 ft)
A = Change in grade (|G
2
-G
1
| as a percent)

The stopping sight distance is normally the controlling sight distance for sag vertical curves. At
decision points, the roadway should be illuminated by other means so that the sight distance of the
driver is extended. Where possible, increased curve length may also be provided.

Highway overpasses or other obstacles can occasionally reduce the sight distance on sag vertical
curves. In these instances, separate equations should be used to determine the correct curve
length. These equations are readily available in design manuals.

At this point, you have all of the information that you need to develop the precise layout of your
vertical curve. The parabolic curve calculations are identical for sag and crest vertical curves. Just
remember to use the appropriate positive or negative values for the participating grades.

3.4. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable Geometric Design concepts. The
more concerned you are about your understanding of a topic, the more seriously you will want to
approach the example problem for that topic.

3.4.1. Stopping Sight Distance

While descending a -7% grade at a speed of 90 km/h, George notices a large object in the roadway
ahead of him. Without thinking about any alternatives, George stabs his brakes and begins to slow
down. Assuming that George is so paralyzed with fear that he won't engage in an avoidance
maneuver, calculate the minimum distance at which George must have seen the object in order to
avoid colliding with it. You can assume that the roadway surface is concrete and that the surface is
wet. You can also assume that George has a brake reaction time of 0.9 seconds because he is always
alert on this stretch of the road.

Solution

First, we need to calculate the distance that George traveled during his brake reaction time. This is
done using the equation D = V T from physics. Since George's brake reaction time was 0.9
seconds and his velocity was 25 m/sec (90 km/h), the distance he traveled during his brake reaction
time was 22.5 meters. Second, we need to calculate the distance George will travel while
braking. This is done using the equation shown below.

u =
v
2
2g(f + u)


Where:
51


V = George's Velocity, 25 m/sec (90 km/h)
g = Acceleration due to gravity, 9.81 m/sec
2

f = Coefficient of friction, 0.29 (we'll use the value for 96 km/h (60 mph) just to be
conservative)
G = The grade of the road, -0.07 (-7%)

Solving the equation yields a distance of 145 meters. Summing the distance traveled while braking
and the distance traveled during the brake reaction time yields a total stopping sight distance of 168
meters, which is about 15.2 meters short of two football fields. George needed to be about 168
meters away from the object at the instant he first saw it in order to avoid a collision.

3.4.2. Passing Sight Distance

A vehicle moving at a speed of 50 mph is slowing traffic on a two-lane highway. What passing
sight distance is necessary, in order for a passing maneuver to be carried out safely?

Calculate the passing sight distance by hand, and then compare it to the values recommended by
AASHTO. In your calculations, assume that the following variables have the values given:

Passing vehicle driver's perception/reaction time = 2.5 sec
Passing vehicle's acceleration rate = 1.47 mph/sec
Initial speed of passing vehicle = 50 mph
Passing speed of passing vehicle = 60 mph
Speed of slow vehicle = 50 mph
Speed of opposing vehicle = 60 mph
Length of passing vehicle = 22 ft
Length of slow vehicle = 22 ft
Clearance distance between passing and slow vehicles at lane change = 20 ft
Clearance distance between passing and slow vehicles at lane re-entry = 20 ft
Clearance distance between passing and opposing vehicles at lane re-entry = 250 ft

You should also assume that the passing vehicle accelerates to passing speed before moving into the
left lane.

Solution

The first step in calculating the passing sight distance is the calculation of the distance D
1
. This
distance includes the distance traveled during the perception/reaction time and the distance traveled
while accelerating to the passing speed. The distance traveled during the perception reaction time is
computed using D = V T from physics, where V = 73.3 f./sec (50 mph) and T = 2.5
seconds. Solving for D yields a value of 183.3 feet. The distance traveled during the acceleration
portion of D
1
is computed using the equation v
I
2
= v
I
2
+2A B, where V
f
= 88 ft/sec (60 mph),
V
i
= 73.3 f./sec (50 mph), and A = 2.155 ft/sec/sec (1.47 mph/sec). Solving for D yields a value of
550.1 feet. The total distance D
1
is 183.3 + 550.1 = 733.4 feet.

The second portion of the passing sight distance is the distance D
2
, which is defined as the distance
that the passing vehicle travels while in the left lane. This distance can be calculated in the
following way.

While in the left lane, the passing vehicle must traverse the clearance distance between itself and
the slow vehicle, the length of the slow vehicle, the length of itself, and the length of the clearance
52

distance between itself and the slow vehicle at lane re-entry. The time it takes the passing vehicle to
traverse these distances relative to the slow vehicle can be computed from the equation D = V T,
where D = 84 ft (20 ft + 22 ft + 22 ft + 20 ft) and V = 14.67 ft/sec (10 mph = relative speed of
passing vehicle with reference point on the slow vehicle).

Solving for the time T
2
yields a value of 5.7 seconds. The real distance traveled by the passing
vehicle during the time T
2
is calculated using D = V T, where V = 88 ft./sec. (60 mph) and
T = 5.7 seconds. Solving for D yields the distance D
2
or 501.6 ft.

The distance D
3
is the clearance distance between the passing vehicle and the opposing vehicle at
the moment the passing vehicle returns to the right lane. This distance was given as 250 ft. The
distance D
4
is the final component of the passing sight distance and is defined as the distance the
opposing vehicle travels during 66% of the time that the passing vehicle is in the left lane. This
distance is computed using D = V T, where V = 88 ft./sec. (60 mph) and T = 3.7 seconds
(5.7 66%). Solving for D yields a value of 325.6 ft. for D
4
.

The total passing sight distance is, therefore, D
1
+ D
2
+ D
3
+ D
4
or 1811 ft. The passing sight
distance recommended by AASHTO for speeds within the 50 mph - 60 mph range is 1900 ft. Our
approximation came within 100 ft. of the values recommended by AASHTO.

3.4.3. Horizontal Curve Radius Calculations

A new transportation engineer is charged with the design of a horizontal curve for the Queen's
Highway in Canada. His final design calls for a curve with a radius of 520 meters. Would you sign
your name to his plans?

Assume that the design speed for the Queen's highway is 110 km/h. You can also assume that snow
and ice will be present on the roadway from time to time (it's Canada).

Solution

The first step in a review of his plans would be to make sure that the curve radius as designed is
greater than the minimum curve radius. For a design speed of 110 km/h, the comfortable side-
friction factor is 0.10. In addition, since the roadway will be covered with snow and ice from time
to time, the maximum superelevation rate is 8%. With this information we can go ahead and
calculate the minimum curve radius using the equation below.

R
mIn
=
v
2
127 [
e
max
1uu
+ f
max



Where:

R
min
= Minimum radius (m)
V = Design speed, 110 km/hr
e
max
= Maximum superelevation rate, 8%
f
max
= Maximum side-friction factor, 0.10

Substituting and solving yields a minimum radius of 530 meters. The 520 meter radius that is called
for in the plans would probably work, but it might be uncomfortable for the vehicle occupants. A
larger radius would be more appropriate.

53

3.4.4. Horizontal Curve Sight Distance

A large grain elevator is located 40 feet from the centerline of a two-lane highway, which has
12-foot wide lanes. The elevator is situated on the inside of a horizontal curve with a radius of 500
feet. Assuming that the elevator is the only sight restriction on the curve, what is the minimum sight
distance along the curve?

Solution

The first thing that we need to do is calculate the distance from the edge of the grain elevator to the
center of the nearest lane. This turns out to be 40 - 6 = 34 ft. Next, we need to calculate the degree
of the curve using the equation R = 5730/D. The degree of the curve turns out to be 11.46. The last
step involves solving for the sight distance using the equation below.

N =
S7Su
B
_1 -cos
SB
2uu
]

Where:

M = Distance from the center of the inside lane to the obstruction, 34 ft
D = Degree of the curve, 11.46.

Where R = 5730 / D

S = Sight distance (ft)
R = Radius of the curve, 500 ft

Substituting the values for the variables and solving for the sight distance yields a sight distance of
371 feet. You might want to change the position of the elevator and see how it affects the sight
distance.

3.4.5. Transition Segments

After designing a horizontal curve with a radius of 1910 feet for a highway with a design speed of
70 mph, your final task is to design the transition segments. Your local design code requires that
any superelevation within the curve be run-off over a distance equal to or greater than the distance a
vehicle would travel in two seconds at your design speed. In addition, the spiral curves must have
the minimum length given by the equation below.

L =
S.1S v
3
R C


Where:

L = Minimum length of the spiral curve (ft.)
v = Speed (mph)
R = Circular curve radius (ft.)
C = Centripetal acceleration development rate (usually between 1 and 3 ft/sec
3
)

If you use a centripetal acceleration development rate of 2 ft/sec
3
, what is the minimum length of
your transition segments?


54

Solution

We'll calculate the required length of your transition segments based on the superelevation
restrictions first. At 70 mph (102.6 ft/s) you would travel a distance of 205.2 feet in two
seconds. Your transition segments should, therefore, slowly change the cross-section of the
road over the course of 205.2 feet. The minimum length of the spiral curve is investigated by
substituting the correct values into the equation below.

L =
S.1S v
3
R C


Where:

L = Minimum length of the spiral curve (ft)
v = Design speed, 70 mph
R = Circular curve radius, 1910 ft
C = Centripetal acceleration development rate, 2 ft/sec
3


Solving for the curve length yields a minimum spiral curve length of 282.8 ft.

Since you are required to have a 282.8 foot-long spiral curve, you should gradually change the road
cross-section from its normal state to the superelevated state over 282.8 ft. Your transition segments
should be 282.8 feet long.

3.4.6. Ascending Grades

If a highway with traffic normally running at 65 mph has an inclined section with a 3% grade, how
much can the elevation of the roadway increase before the speed of the larger vehicles is reduced to
55 mph?

Solution

Looking at the table in the Ascending Grades module, we can see that a 3% grade causes a
reduction in speed of 10 mph after 1400 feet. To find the exact increase in the elevation of the
highway we would need to employ some simple trigonometry. But, since the angle of a 3% grade is
small, we can just estimate the elevation increase by multiplying the length of the grade by the
grade itself. This yields 1400 0.03 = 42 ft. The elevation of the roadway can only be increased by
about 42 feet before heavy vehicles are reduced to a speed of 55 mph.

3.4.7. Crest Vertical Curves

You have been instructed to design a crest vertical curve that will connect a highway segment with
a 3% grade to an adjoining segment with a -1% grade. Assume that the minimum stopping sight
distance for the highway is 540 feet. If the elevation of the VPC is 1500 ft, what will the elevation
of the curve be at L/2?

Solution

The first step in the analysis is to find the length of the crest vertical curve. The grade changes from
3% to -1%, which is a change of -4% or A = |-4%|. In addition, for the stopping sight distance
h
1
= 3.5 ft and h
2
= 0.5 ft. Since we know S = 540 ft, we can go ahead and solve for the length of
the crest vertical curve.
55


If S > L then (invalid because L > S)

L = 2S -
2uu(h
1
+ h
2
)
2
A


If S < L then

L =
A S
2
1uu(2h
1
+2h
2
)
2


Where:

L = Length of the crest vertical curve (ft)
S = Sight distance, 540 ft
A = The change in grades, |-4%|
h
1
= Height of the driver's eyes above the ground, 3.5 ft
h
2
= Height of the object above the roadway, 0.5 ft

The curve length calculated from the 'S < L' equation was 877.5 feet, which is greater than the sight
distance of 540 feet. To find the elevation of the curve at a horizontal distance of L/2 from the VPC,
we need to use the equation below.

Y = vPCy +B - x +
A - x
2
2uu - L


Where:

Y = Elevation of the curve at a distance x from the VPC (ft)
VPCy = Elevation of the VPC, 1500 ft
B = Slope of the approaching roadway, or the roadway that intersects the VPC, 0.03
A = The change in grade between the disjointed segments, -4 (From 3% to -1% would be a
change of -4%)
x = L/2 = 877.5 ft / 2 = 438.75 ft

The equation above yields a curve elevation of 1508.8 feet at a distance L/2 from the VPC.

3.4.8. Sag Vertical Curves

If a stopping sight distance of 400 ft. is to be maintained on a sag vertical curve with tangent grades
of -3% and 0%, what should the length of the curve be? Assume a headlight beam upward
divergence angle of 1.

Solution

Since we know everything that we need to know to solve this problem, we'll jump straight into the
equations.

If S > L then

L = 2S -
2uu(B + S tanB)
A

56


If S < L then (invalid because L < S)

L =
A S
2
2uu(B +S tanB)


Where:

L = Curve length (ft)
S = Sight distance, 400 f.
B = Beam upward divergence, 1
H = Height of the headlights, 2 ft (assumed as 2 ft)
A = Change in grade, 3% (|G
2
-G
1
| as a percent)

Solving the equations above results in a curve length of 201 feet. You can find the elevation of any
point along the curve once you have the curve length. See the crest vertical curve example problem.

3.5. Glossary

Acceleration Development Rate: The rate at which the centripetal acceleration necessary to
negotiate a horizontal curve is developed on the transition segment leading up to the curve.

Actual Sight Distance: The sight distance provided by the highway as designed.

Brake Reaction Time: The elapsed time between recognition of an object in roadway ahead and
application of the brakes.

Braking Distance: The distance traveled while braking to a complete stop.

Centripetal Force: The force required to keep an object moving in a circular path. The centripetal
force is always normal to the direction of the object.

Coefficient of Friction: A dimensionless parameter that quantifies the resistance to sliding at the
interface of two surfaces.

Crest (vertical) Curve: A curve that connects a segment of roadway with a segment of roadway
that has a more negative grade. (uphill to level, uphill to downhill...)

Decision Sight Distance: The sight distance that should be provided wherever drivers are forced to
make decisions or are forced to cope with large amounts of information. (Also see 'pre-maneuver
time')

Design Speed: The speed at which a vehicle should be able to traverse a roadway safely under
favorable environmental conditions.

Grade (roadway): The slope of the roadway surface. Grade is expressed as the change in elevation
per 100 feet of horizontal distance.

Horizontal Alignment: The part of geometric design involved with designing the shape of the
roadway within the horizontal plane.

57

Length (vertical curve): The horizontal distance from one end of the vertical curve to the other, or
the horizontal distance between the VPC and the VPT.

Passing Sight Distance: The sight distance required for drivers to feel comfortable about making a
passing maneuver.

Pre-Maneuver Time: The time required for a driver to process information relative to a hazard,
plot an avoidance course, and initiate the required avoidance maneuver.

Sag (vertical) Curve: A vertical curve that connects a segment of roadway with a segment of
roadway that has a more positive grade. (downhill to level, downhill to uphill...)

Side-Friction Factor: The dimensionless factor used to describe the frictional resistance to
slippage normal to the direction of travel.

Sight Distance: The length of roadway ahead over which an object of a specific height is
continuously visible to the driver.

Stopping Sight Distance, (Minimum): The distance required for a driver to react to a hazard in the
roadway ahead and bring his/her vehicle to a complete stop. The sum of the distance required to
stop the car and the distance traveled during the break reaction time.

Superelevation: Inclined roadway cross-section that employs the weight of a vehicle in the
generation of the necessary centripetal force for curve negotiation.

Superelevation Rate: The slope of the roadway cross-section. For example, a 10 ft wide roadway
with a superelevation rate of 10% would be 1 ft higher on one side than it is on the other.

Tangent Grade: A grade that shares a common slope with the end of a vertical curve.

Vertical Alignment: The portion of geometric design that deals with the shape of the roadway in
the vertical plane.

Vertical Curve: A parabolic curve used to provide a gradual change in grade between roadway
segments with differing grades.

















58

4. Parking Lot Design

4.1. I ntroduction

Off-street parking is an important part of the transportation system. It is an efficient means of
storing vehicles while they arent in use, and it causes little disruption to the neighboring roadways.
Additionally, since parking is the terminal or destination for a trip, the availability of off-street
parking can affect the attractiveness of destinations as well as transportation modes. The
attractiveness of a destination is reduced if there is a delay or difficulty in parking. The use of
transit systems is increased in areas where parking is scarce. To be efficient, the transportation
system must include adequate parking facilities at all places that attract trips.

Off-street parking plays an important role in the efficiency of the overall transportation system. This
chapter is designed to help the undergraduate engineering student understand the fundamentals of
planning and designing off-street parking.

4.2. Lab Exercises

This exercise will help increase your understanding of Parking Lot Design, by presenting a more
complicated problem that requires more thorough analysis.

4.2.1. Lab Exercise One: Facility Analysis and Design

A local off-street parking facility might be inadequate. You have been asked to evaluate the
adequacy of the parking facility. If your results show that the facility is inadequate, you are to
propose suitable modifications and/or additions that will correct the problem. You should prepare a
brief report that summarizes your analysis and any proposed modifications to the facility.

Your instructor will designate the off-street parking facility and provide any additional information
that might be needed.

Tasks to be Completed

Task 1: Develop an inventory of the parking facility including:

the location, condition, type, and number of parking spaces,
any time limits, hours of availability and other restrictions, and
the geometry of the spaces and other features.

Task 2: Estimate the peak parking period for the facility and complete an accumulation count study
for the facility during that period. Using your results, develop an accumulation graph.

Task 3: Perform a simplified license plate survey and estimate the average length of time that
vehicles are parked at the facility during the peak parking period.

Task 4: Using the information from the first three tasks, calculate the probability that an incoming
car will not find a parking space during the peak parking period. Determine whether or not the
probability of rejection is acceptable.

Task 5: If the current parking facility is inadequate, design modifications and/or additions that will
correct the problem. Your goal is to suggest the most economical solution. Report any assumptions
that you make.
59


4.3. Theory and Concepts

Many facets of parking are interrelated. Parking programs usually evolve from parking studies that
determine the supply of parking and the current demand, as well as estimates of future parking
demand. Parking facility planning and design requires a determination of the number of spaces
needed, the proper location for these spaces, and a workable layout with acceptable operating
controls. In addition, other factors need to be considered, such as characteristics of the
transportation system and users, transportation attractions and generators, and transportation
operations. To start learning more about Parking Lot Design, select a topic from the list below.

4.3.1. Parking Studies

Studies must be conducted to collect the required information about the capacity and use of existing
parking facilities. In addition, information about the demand for parking is needed. Parking studies
may be restricted to a particular traffic producer or attractor, such as a store, or they may encompass
an entire region, such as a central business district.

Before parking studies can be initiated, the study area must be defined. A cordon line is drawn to
delineate the study area. It should include traffic generators and a periphery, including all points
within an appropriate walking distance. The survey area should also include any area that might be
impacted by the parking modifications. The boundary should be drawn to facilitate cordon counts
by minimizing the number of entrance and exit points.

Once the study area has been defined, there are several different types of parking studies that may
be required. These study types are listed below and discussed in detail in the remaining paragraphs.

Inventory of Parking Facilities
Accumulation Counts
Duration and Turnover Surveys
User Information Surveys
Land Use Method of Determining Demand

Inventory of Parking Facilities:

Information is collected on the current condition of parking facilities. This includes:

the location, condition, type, and number of parking spaces.
parking rates if appropriate. These are often related to trip generation or other land use
considerations.
time limits, hours of availability and any other restrictions.
layout of spaces: geometry and other features such as crosswalks and city services.
ownership of the off-street facilities.

Accumulation Counts:

These are conducted to obtain data on the number of vehicles parked in a study area during a
specific period of time. First, the number of vehicles already in that area are counted or estimated.
Then the number of vehicles entering and exiting during that specified period are noted, and added
or subtracted from the accumulated number of vehicles. Accumulation data are normally
summarized by time period for the entire study area. The occupancy can be calculated by taking
accumulation/total spaces. Peaking characteristics can be determined by graphing the accumulation
60

data by time of day. The accumulation graph usually includes cumulative arrival and cumulative
departure graphs as well.




Above Figures taken from:
Khisty, C. and M. Kyte, Lab and Field Manual for Transportation
Engineering, Prentice Hall, Englewood Cliffs, NJ, 1991.

Duration and Turnover Surveys:

The accumulation study does not provide information on parking duration, turnover or parking
violations. This information requires a license plate survey, which is often very expensive. Instead,
modifications are often made to the field data collection protocols. Note that there is usually a
tradeoff between data collection costs and study accuracy. Spending more time and money may
increase accuracy, but at what point does the incremental change in accuracy become too
expensive?
61


In planning a license plate survey, assume that each patrolling observer can check about four spaces
per minute. The first observer will be slower, because all the license plate numbers will have to be
recorded, but subsequent observers will be able to work much faster. The form shown below can be
used for a license plate survey.

Parking turnover is the rate of use of a facility. It is determined by dividing the number of available
parking spaces into the number of vehicles parked in those spaces in a stated time period.

Table1. Typical License Plate Survey Field Form for Curbside Survey

Street_______Side____Study Date_________________
Data Collector_________From_____________________
To ________________ Direction of Travel ___________

Space
No
Space
Desc.
Time at beginning of Patrol
8:00 8:30 9:00 9:30 10:00 10:30 11:00 11:30

User Information Surveys:

Individual users can provide valuable information that is not attainable with license plate surveys.
The two major methods for collecting these data are parking interviews and postcard studies. For
the parking interviews, drivers are interviewed right in the parking lot. The interviews can gather
information about origin and destination, trip purpose, and trip frequency. The postage paid
postcard surveys requests the same information as in the parking interview. Return rates average
about 35%, and may include bias. The bias can take two forms. Drivers will sometimes
overestimate their parking needs in order to encourage the surveyors to recommend additional
parking. Or, they may file false reports that they feel are more socially acceptable.

Land Use Method of Determining Demand:

Parking generation rates can be used to estimate the demand for parking.

Tabulate the type and intensity of land uses throughout the study area.
Based on reported parking generation rates, estimate the number of parking spaces needed for
each unit of land use.
Determine the demand for parking from questionnaires. A rule of thumb is to overestimate the
demand for parking by about 10 %. If the analysis suggests that the parking demand for a
particular facility will be 500 spaces, then the design should be for 550 spaces.



62

4.3.2. Adequacy Analysis

The adequacy of a parking facility can be measured by calculating the probability that an entering
vehicle will not be able to find a parking space. A high probability of rejection (not finding a space)
may indicate that expansion of the parking facility is warranted.

The probability of rejection can be calculated by comparing the traffic load to the number of
parking stalls as shown below.

First, the traffic load is estimated using:

A = Q T

Where:

A = traffic load,
Q = incoming vehicle flow rate, and
T = the average parking duration.

Make sure that your units of time cancel each other. If you give Q in vehicles per hour, then use T
in units of hours.

Next, calculate the probability of rejection using the following formula:

P =
A
M
N!
1 +A +
A
2
2
+ +
A
M
N!


Where:

P = the probability of rejection,
A = the traffic load, and
M = the number of parking stalls.

If the probability of rejection is high, you may want to consider adding more parking stalls to the
parking facility.

4.3.3. Parking Facility Design Process

The goal in designing off-street parking facilities is to maximize the number of spaces provided,
while allowing vehicles to park with only one distinct maneuver. It would be nice to present a
step-by-step procedure for reaching this goal, but it isnt that simple. Parking lot design requires
balancing a variety of concerns. For example, you might decide on a nice layout for your parking
lot, only to realize that you havent provided any spaces for persons with disabilities. The next
iteration would correct this error, but might very well create another problem. You simply have to
hammer out all of the kinks, until you end up with a design that satisfies all of your criteria.

One way to start is to imagine that you are parking your own car in a lot. What maneuvers would
you need to make? Knowing that, what needs to be included in the design to make sure all those
maneuvers are possible? Use the following list of maneuvers to guide your thinking.

1. Vehicle enters from street (space provided by entry driveway).
63

2. Vehicle searches for a parking stall (space provided by circulation and /or access aisles).
3. Vehicle enters the stall (space provided by the access aisle).
4. Vehicle is parked (stall designed to accommodate the vehicles length and width plus space to
open vehicle doors).
5. Pedestrians access the building or destination (usually via the aisles).
6. Vehicle exits the parking stall (space provided by the access aisle).
7. Vehicle searches for an exit (space provided by the access and circulation aisles).
8. Vehicle enters the street network (space provided by the exit driveways).

4.3.4. Entrance Considerations

The first maneuver that a parking vehicle will make involves leaving the street and entering the
off-street parking lot. This maneuver, while simple, requires some careful thought by the parking lot
designer.

Analysis of the demand for parking may indicate that there are periods during the day in which a
large number of vehicles want to enter the parking facility at roughly the same time. The entrance to
the parking lot must be able to handle the entering traffic without forcing vehicles to wait in the
street, because stagnant vehicles will reduce the capacity of the adjacent street. To avoid conflicts
with other traffic, entrances should be located as far from intersections and conflict points as
possible. Multiple entrances may ease access and reduce restriction on the adjacent roadways.

Years of experience have produced general entrance dimensions that seem to work. The basic
nominal design width for a two-way driveway serving commercial land use is 30 ft., with 15 feet
radii. With greater volumes such as at a shopping center, a 36-ft driveway may be appropriate. It
should be marked with two exit lanes (each 10 or 11 ft wide) and a single entry lane (14-16 ft wide)
to accommodate the off-tracking path of entering vehicles. Larger commercial facilities such as
regional shopping centers may require twin entry and exit lanes separated by a 4-12 foot median.

Many areas of the country have specific regulations or guidelines for the design of access facilities.
It is important that the local and state regulations concerning access management are followed when
designing access to off-street parking facilities.

4.3.5. I nternal Considerations

There are two major internal maneuvers that the parking lot designer must consider, vehicles
searching for an open stall and vehicles searching for an exit. These internal maneuvers require
space, which is space that cannot be used for parking.

Off-street parking facilities normally operate in one of two ways. The first and most common
operation is self-parking, in which the driver maneuvers the vehicle through the parking lot. The
second operation is attendant parking, in which parking attendants maneuver the vehicle through
the parking lot. Parking facilities that use self-parking must normally include larger aisles, as
individuals unfamiliar with the parking facility may require extra room to maneuver. Attendant
parking is normally more expensive to operate than self-parking.

Tollbooths and other restrictions at entrances or exits also affect the internal operation of parking
facilities. Tollbooths require reservoir space within the parking facility for vehicles that are waiting.
In general, 2-3 spaces per lane are required at entrances to self-parking lots where a ticket needs to
be acquired. At exits, a much larger reservoir should be provided, because toll collection requires
extra time.

64

Finally, in areas where winter snowfall is common, consideration of the snowfall removal operation
should be included in the design process. Adequate space must be provided for snow removal
equipment to maneuver.

4.3.6. Parking Stall Layout Considerations

The objective of the layout design is to maximize the number of stalls, while following the
guidelines below.

The layout of the parking facility must be flexible enough to adapt to future changes in vehicle
dimensions.
The stall and aisle dimensions must be compatible with the type of operation planned for the
facility.

The critical dimensions are the width and length of stalls, the width of aisles, the angle of parking,
and the radius of turns. All of these dimensions are related to the vehicle dimensions and
performance characteristics. In recent years there have been a number of changes in vehicle
dimensions. The popularity of minivans and sport utility vehicles has had an impact on the design
of parking facilities. For the near future, a wide mix of vehicle sizes should be anticipated. There
are three approaches for handling the layout:

1. Design all spaces for large-size vehicles (about 6 feet wide and 17-18 ft long).
2. Design some of the spaces for large vehicles and some for small vehicles (these are about 5 ft
wide and 14-15 ft long).
3. Provide a layout with intermediate dimensions (too small for large vehicles and too big for
small vehicles).

For design, it is customary to work with stalls and aisles in combinations called "modules". A
complete module is one access aisle servicing a row of parking on each side of the aisle. The width
of an aisle is usually 12 to 26 feet depending on the angle at which the parking stalls are oriented.

Stall Width

For simplicity, the stall width is measured perpendicular to the vehicle, not parallel to the aisle. If
the stall is placed at an angle of less than 90, then the width parallel to the aisle will increase while
the width perpendicular to the vehicle will remain the same.

Stall Length

The length of the stall should be large enough to accommodate most of the vehicles. The length of
the stall refers to the longitudinal dimension of the stall. When the stall is rotated an angle of less
than 90, the stall depth perpendicular to the aisle increases up to 1 foot or more. It should be noted
that the effective stall depth depends on the boundary conditions of the module, which could
include walls on each side of the module, curbs with or without overhang, or drive-in versus back-in
operations. For parking at angles of less than 90, front bumper overhangs beyond the curbing are
generally reduced with decreasing angle and, for example, drop to about 2 feet at 45 angles. The
Table 8-3 below gives the standard dimensions for several different layouts as defined by
Figure 8-4.

65





66

Table 8-3 and Figure 8-4 where taken from: Weant, R.A. and Levinson, H.S., Parking, Eno
Foundation, 1990, page 161.

Interlock Module

A special type of module, the interlock, is possible at angles below 90. There are two types of
interlock. The most common, and preferable, type is the bumper-to-bumper arrangement. The
second type, the "herringbone" interlock, can be used at 45 and is produced by adjacent sides
having one way movements in the same direction. This arrangement requires the bumper of one car
to face the fender of another car. Figure 8-3 shows several different module layouts that are
commonly used.



Comparing Angle Efficiencies

The relative efficiencies of various parking angles can be compared by looking at the number of
square feet required per car space (including the prorated area of the access aisle and entrances).
Where the size and shape of the tract is appropriate, both the 90 and the 60 parking layouts tend to
require the smallest area per car space. In typical lot layouts for large size vehicles, the average
overall area required (including cross aisles and entrances) ranges between 310 and 330 square
feet/car. A very flat angle layout is significantly less efficient than other angles.


67

One-Way Aisles

There are many conditions where one-way aisles are desirable. With parking angles less than 90,
drivers can be restricted to certain directions. However, the angle should usually be no greater than
75. Drivers may be tempted to enter the parking aisles and stalls from the wrong direction when
the stall angle is too large. Adjacent aisles generally have opposite driving directions.

4.3.7. Exit Considerations

The last maneuver that a vehicle will make in a parking facility involves leaving the facility and
entering the adjacent street network. Inefficiencies in this part of the parking process can lead to
reduced capacity in both the parking facility and the adjacent street network.

As was discussed in the section entitled Entrance Considerations, access and egress points should
be located as far as possible from any conflicting points on the adjacent street network. This
normally means that entrances and exits are placed mid-block. Multiple exit lanes may be
required, so that right-turning vehicles can avoid waiting for left-turning vehicles at the exit. If the
flow rate of departing vehicles is low, or if the adjacent street is one-way, a single lane may be
sufficient.

The type of parking facility also impacts the exit design. Facilities that have tollbooths near the
exits will require multiple exit lanes. They may also require that a large portion of the parking lot be
devoted to lanes for vehicles waiting to pay at the tollbooths.

4.3.8. ADAAG Requirements

The Americans with Disabilities Act Accessibility Guidelines for Buildings and Facilities
(ADAAG) specifies the number and dimensions of accessible parking spaces. Where possible, the
accessible parking spaces should be provided on the accessible path to the facility entrance and also
minimize the distance traveled.

Total Parking in Lot Required Minimum Number Of Accessible Spaces
1 to 25 1
26 to 50 2
51 to 75 3
76 to 100 4
100 to 150 5
151 to 200 6
201 to 300 7
301 to 400 8
401 to 500 9
501 to 1000 2 percent of total
1001 and over 20 plus 1 for each 100 over 1000

One in eight accessible spaces, but not less than one, should be "van accessible". These spaces
should be 96 in (2440mm) wide. Parking access aisles need to be part of the accessible path to the
building. Two adjacent accessible spaces may share a common access aisle. The access aisle should
be 5 feet wide. Parking spaces and access aisles should be level, with surface slopes not greater than
1:50 (2%) in all directions.

68

4.4. Professional Practice

In order to supplement your knowledge about the various concepts within Parking Lot Design, and
in order to give you a glimpse of how these various topics are discussed in the professional
environment, we have included selected excerpts from professional design manuals.

4.4.1. Parking Studies


The following excerpts were taken from the1992 edition of the Transportation Planning Handbook,
published by the Institute of Transportation Engineers (p. 199-400).

Parking Studies (p. 199)

Parking studies are used to evaluate the current supply of parking or to plan for future parking
needs. Some parking studies are only concerned with the adequacy of parking for a particular need,
such as a shopping mall, office building, or a sports facility. Other studies are designed to evaluate
the parking conditions in an area to establish time limits, parking rates, and the need for additional
parking. Some studies are used to aid operational analyses in relation to removal or modification of
curb parking. Still others are required to evaluate residential parking impacted by encroachment
from outside parkers. There are a wide variety of other specialized studies to meet specific needs.

Supply and Demand (p. 400)

Parking supply is merely the number and location of all parking spaces in the study area. The
supply is defined by the parking inventory described earlier in this chapter (under inventories).
Supply is much easier to quantify than is demand because it is a physical count. Demand, on the
other hand, is an estimate of the number of drivers who wish to park in the study area at any given
time. Supply is generally constant, although there can be some changes during the day (e.g., tow
away zones during peak hours, part-time loading zones, etc.). Demand varies by time. In fact, one
of the elements to be defined in the study is the time of peak demand. In some areas there may be
multiple peaks because of the differing uses within the study area. A simple example is an office
complex. The peak employee accumulation may be by 9:00 A.M., while the peak client or visitor
accumulation may be 11:30 A.M. or 2:30 P.M. Deliveries or service personnel may peak at still
different times.

Current demand may be estimated in those study areas where supply greatly exceeds demand by
merely counting the accumulated vehicles at various times of the day. However, when the demand
reaches 85 percent or more of the supply, it may not represent the true demand because there may
be additional demand that is not present because of the lack of adequate parking.

User Characteristics

User characteristics analyses are made to assist in parking management in an area. Such studies are
used in establishing time-limit parking, employee parking, loading zones, etc. Information is
obtained on the magnitude of the various segments of the parking demand. In other words, the study
is used to project the demand for short-term parking (15 to 20 minutes); for errands at banks,
pharmacies, dry cleaners, etc.; for limited parking (1 to 2 hours) covering short-term shopping or
business appointments; for longer term parking (8 hours or more) for employees in the area.


69

4.4.2. Types of Facilities

The following excerpt was taken from the1992 edition of the Transportation Planning Handbook,
published by the Institute of Transportation Engineers (p. 175).

Off-street facilities range from the parking pad, carport, or garage of a single-family home to lots or
garages serving large parking generators such as shopping centers, airports, and sporting events.
Most off-street parking is accommodated in free facilities (technically this is a misnomer, since all
parking carries a cost which is reflected in the price of a home, the rent of a retail or commercial
building, the price of a product, etc). However, this term is used to distinguish from those types of
commercial facilities which charge a specific fee to the driver. . . .

Unfortunately, a large supply of parking also is provided at the curb, along streets. The use of
streets for curb parking exacts a heavy toll in accidents and congestion (along the more heavily
traveled routes). While most curb parking is free and open to the general public, the use of parking
meters in business areas converts the street curb to a charge, or revenue producing, operation.
Additionally, some curb spaces are limited to specific users such as bus stops and loading zones for
trucks, taxis, or passenger pickup/drop-off.

4.4.3. Types of Operation

The following excerpt was taken from the1992 edition of the Traffic Engineering Handbook,
published by the Institute of Transportation Engineers (pp. 204-205).

Most of the so-called public parking facilities represent a more centralized and general-purpose use.
Parkers usually have a choice of several destinations. Much of the time they also have a choice of
alternate places to park. A facility open to the general public usually needs to attract parkers if
investment in its construction is to be justified. When the facility is revenue-financed, the need is
obvious. Even when the parking is free, justification is needed for the expenditure of benefit district
assessment funds, parking meter revenue, or other public funds used to acquire land and to
construct and operate the facility.

The design of a general-purpose parking facility must take into consideration the type of proposed
operation with attendants, by self parking, or with a combination of the two. The most economical
operation occurs where patrons park their own cars. In heavily used facilities where patrons pay for
parking, it is sometimes feasible to utilize attendants to park the cars after the patron pulls into the
lot. This also frequently occurs in older parking garages. Some facilities can operate on both
systems, with certain areas reserved for self-parkers and other areas served by attendants. . . .

A basic factor in design of a parking facility is the expected use by type of parker or type of
generator being served. Parking duration can be either short-term or long term, or a combination of
both. Design dimensions are often larger in facilities for short-term parkers because of the high
turnover rate and the need to provide easy access and circulation. . . .

4.4.4. Operational Design Elements

The following excerpt was taken from the1992 edition of the Traffic Engineering Handbook,
published by the Institute of Transportation Engineers (pp. 204-205).

The design of a parking facility is very strongly influenced by its intended operation. The basic
design elements and their associated operational features may be identified in successive steps as
follows:
70


1. Vehicular access from the street system (entry driveway);
2. Search for a parking stall (circulation and/or access aisles);
3. Maneuver space to enter the stall (access aisles);
4. Sufficient stall size to accommodate the vehicles length and width plus space to open car doors
wide enough to enter and leave vehicle (stall dimensions);
5. Pedestrian access to and from the facility boundary (usually via the aisles) and vertically by
stairs, escalators, or elevators in multilevel facilities;
6. Maneuver space to exit form the parking stall (access aisles);
7. Routing to leave facility (access and circulation aisles);
8. Vehicular egress to the street system (exit driveway); and
9. Any revenue-control system (may involve elements of entry, exit, or both).

The simplest form of off-street parking is a single stall at home. Assuming a straight driveway,
steps 1 and 8 above use the same lane and curb cut, and step 9 does not apply. Steps 2 and 7 are
rudimentary. Thus, a driveway serving a one-car parking stall or garage cannot be considered as
representing a second parking space, if such parking would block continuous access to the basic
stall. Step 6 usually involves backing out into the public street or alley, as part of steps 7 and 8.
Herein lies the essential difference between low-volume parking and what generally should be
practiced in facilities designed to handle more than a few cars. Except along alleys, the larger lots
should have all parking and unparking maneuvers contained off-street. Frequent backing of cars
across sidewalks and into public streets increases congestion and creates hazards.

For the large facilities, and particularly garages, an operational concept necessarily precedes
structural, architectural, and other design elements. The concept begins with the question, "What do
we plan to serve?" From answers to this question, design features emerge such as user ease of
access, security, vehicle circulation and walk patterns, signing, lighting, and equipment needs.

4.4.5. Change of Mode Parking

The following excerpts were taken from the1992 edition of the Transportation Planning Handbook,
published by the Institute of Transportation Engineers (pp. 183-184).

. . . There are two general types of park and ride lots: (1) a change from the private vehicle to some
form of public transportation such as bus, rapid transit, or suburban rail and (2) carpooling.

Transit Stations

The interfaces are the following:

Pedestrian or walk-in-traffic
Private automobile park-and-ride or pickup/drop-off
Transit transfers such as bus to bus, or bus to rail
Taxi

For most locations, there are two elements to the park-and-ride function: (1) long-term, all day
commuter parking, which represents the major consideration, and (2) short-term spaces that are
desirable to encourage midday shopper, sporting event, or other personal trip usage of the transit
facility during off peak periods. These spaces would typically be used from 4 to 6 hours. Efficiency
of land use is enhanced by combining the P/D operation with the short-term space needs. Thus, a
very short time limit during the A.M. and P.M. peak transit activity, such as 5 to 10 minutes, is
imposed on a limited number of spaces adjacent to the station. These spaces are then available for
71

intermediate-term parking during the balance of the day. The major P/D problem involves the
pickup element in the evening, when motorists arrive and temporarily park while awaiting arrival of
the commuter train or express bus. . . .Additional planning elements of the transit terminal include
loading/unloading spaces for buses and waiting areas for taxis.

The important parking characteristics of a transit station include the number of P/D spaces needed
versus the number of long-term spaces. Reliable estimators for the number of P/D spaces, on a per-
originating-daily-passenger basis, are needed but have not been identified. Three studies in the
Chicago area found a range of 0.05 to 0.07 (average 0.06) spaces per originating passenger;
however, additional research on this parking demand is needed in other cities. . . .

The parking space demand per originating passenger at various types of terminals is given in Table
6.9 (not included here) and suggests a need of about one space per three passengers. The Chicago
area developed a method of estimating current and future parking demand at each station, using data
from ticket sales by mail:

1. Determine the natural service area (NSA) by geographic zones having boundaries defined by
existing patterns of user origins and expected future development.
2. Calculate the current NSA.
3. Calculate the current NSA parking demand.
4. Project ridership growth.
5. Determine the future NSA.

Fringe Parking

Change-of-mode parking facilities can be located at the outer edges of the CBD, or at more remote
distances. Those located near CBDs are served by local or special shuttle buses. Those located
farther away are typically served by express buses, rapid transit, and/or suburban rail. An ITE
committee found that bus-serviced lots have the greatest usage close to the CBD, with a smaller
peak at the 11- to 13- mile range. Rail lots have the greatest usage in the range from 5 to 15 miles
from the CBD. Most bus-serviced lots have transit times greater than automobile travel times,
whereas those with rail typically have shorter travel times. Most change-of-mode lots have transit
service for 14 or more hours per day, and peak-hour transit service headways of 25 minutes or less.
In the Cleveland fringe, buses were reported operating with 5-minute headways during the peak
hours.

Locational factors for parking facilities were identified by Ellis, Bennett, and Rassam:

1. Fringe parking facilities should be located in transportation corridors so that they intercept home
to work trips destined to the CBD at a point where there is sufficient density of traffic demand
that high-quality transit service may be offered.
2. To the extent feasible, facilities should be located on land that is already used for parking or in
low-grade nonresidential use.
3. Such facilities should be located on sites compatible with land uses and activities in the
immediately adjacent area.
4. Potential joint-use aspects of a facility should be considered.
5. Trade-offs in the scale of the facility (such as the level of transit service as opposed to its
neighborhood impacts and the ease of access) should be considered.

72

4.4.6. Downtown Areas

The following excerpt was taken from the1992 edition of the Transportation Planning Handbook,
published by the Institute of Transportation Engineers (p. 177).

Looking at CBDs in general, the following relationships are typical:

1. On-street (curb) parking is related to city population, typically decreasing to about 10 percent of
total supply in cities over 250,000 population.
2. The demand for long-term, work purpose parking increases with population size, ranging from
20 percent of total parkers for cities under 100,000 to over 30 percent for cities approaching 1
million or more population.
3. The average walking distance increases with city size.
4. The parking duration varies by trip purpose and population size. Work trip and residential
parking exhibit the longest durations, while durations for all trip purposes increase with city
size.
5. The parking turnover at the curb is usually three to four times greater than for off-street spaces.
At all facilities, turnover is influenced by the type of parker, the rate structure, local regulations,
and enforcement levels. Furthermore, larger cities generally experience lower parking turnover
than smaller cities.
6. Parking accumulation peaks between 11:00 A.M. and 2:00 P.M. in the average CBD; however,
different trip purposes exhibit unique accumulation patterns. The peak accumulation seldom
exceeds 85 percent of the total parking supply, even though parts of the area are severely
deficient in parking supply (location is the key factor). Peak accumulation tends to increase with
population size, but at a diminishing rate.

4.4.7. Location

The following excerpt was taken from the1992 edition of the Transportation Planning Handbook,
published by the Institute of Transportation Engineers (p. 199).

Key Factors

To assure optimum use, general-purpose parking facilities must be properly located. Whether free
or commercial, the public purpose of a parking lot or garage is to enhance local economic values,
and/or reduce street congestion. Factors that determine appropriate locations for individual facilities
include amount and type of parking shortages, type of nearby generators, facility-user
considerations (whether long-term or short-term), development costs, and street system elements,
such as capacity, directional flows, and turn restrictions. The total parking system of an area should
be considered as it relates to balancing of supply to needs and the street access network.

Most work on parking facility location involves CBD areas. However, other needs exist in outlying
business areas, in older apartment areas with severe parking shortages, and at universities.

4.4.8. Off-Street Zoning

The following excerpt was taken from the1992 edition of the Transportation Planning Handbook,
published by the Institute of Transportation Engineers (p. 403).

Many jurisdictions have zoning regulations that specify the level of parking that must be provided
for various land uses. Data may be obtained from the Institute of Transportation Engineers
publication "Parking Generation" or by making observations of peak parking demand at similar
73

land uses to that under consideration. Peak periods vary for differing land uses. Peak residential
demand occurs in the early morning hours. Some restaurants peak at lunchtime while others peak in
the early evening. Office buildings peak in midmorning while shopping centers peak around noon
on Saturdays.

4.4.9. Design of Off-Street Facilities

The following excerpts were taken from the1992 edition of the Traffic Engineering Handbook,
published by the Institute of Transportation Engineers (pp. 205-215).

Elements of Good Design (pp. 205-206)

In designing any off-street parking facility, the elements of customer service, convenience, and
safety with minimum interference to street traffic flow must receive high priority. Drivers desire to
park their vehicles as close to their destination as possible. The accessibility, ease of entering,
circulating, parking, unparking, and exiting are important factors. Good dimensions and internal
circulation are more important than a few additional spaces. Better sight distances, maneuverability,
traffic flow, parking ease, and circulation are the results of well-organized, adequately designed lot
or garage.

Site Characteristics

Factors such as site dimensions, topography, and adjacent street profiles affect the design of off-
street parking facilities. The relation of the site to the surrounding street system will affect the
location of entry and exit points and the internal circulation pattern.

Access Location

External factors such as traffic controls and volumes on adjacent streets must be considered ---
particularly the location of driveways or garage ramps. It is desirable to avoid locating access or
egress points where vehicles entering or leaving the site would conflict with large numbers of
pedestrians. Similarly, street traffic volumes, turning restrictions, and one-way postings may limit
points at which entrances and exits can logically be placed. It is important to investigate these
factors at the beginning of design.

Driveways should be located to provide maximum storage space and distance form controlled
intersections. . . .

General Elements and Layout Alternatives (p. 212)

Because of their lack of walls or cover, parking lots have no ventilation problems, and lighting is
sometimes provided by relatively tall poles, thus affording high efficiencies and minimizing the
number of poles. Generally, lots have clear sight lines and offer a feeling of greater security than in
a more confined space. Lots are not restricted on vehicle heights and thus afford access to both
commercial and emergency vehicles. . . .

Generally, the layout of a parking lot seeks to strike a balance among maximizing capacity,
maneuverability, and circulation. . . .

The general advantages of 90 parking, as compared with lesser angles, are:

1. Most common and understandable;
74

2. Can sometimes be better fitted into buildings;
3. Generally most efficient if site is sufficiently large;
4. Uses two-way movement (can allow short, dead-end aisles);
5. Allows unparking in either direction. Thus it can minimize travel distances and internal conflict;
6. Does not require any aisle directional signs or markings;
7. Wide aisles often provide room to pass vehicles stopped and waiting for an unparking vehicle;
8. Wide aisles increase separation for pedestrians walking in the aisle and between moving
vehicles;
9. Wide aisles increase clearance from other traffic in the aisle, during unparking maneuvers;
10. Fewer total aisles (hence easier to locate parked vehicle).

Several advantages and disadvantages of angle parking (usually 45 to 75 ), are:

1. Easiest in which to park
2. Can be adapted to almost any width of site by varying the angle;
3. Requires slightly deeper stalls but much narrower aisles and modules;
4. Drivers must unpark and proceed in original direction; hence producing greater out-of-way
travel and conflict;
5. Unused triangles at end of parking aisles reduce overall efficiency;
6. To avoid long travel, additional cross aisles for one-way travel are required, which adds to gross
area used per car parked;
7. Difficult to sign one-way aisles.

Wheel Stops and Speed Bumps (p. 215)

In general, the ends of parking stalls within lots can be marked in a satisfactory fashion by only a
paint line. Wheel stop blocks in the interior of a lot have disadvantages, for they may interfere with
and present a hazard to people walking between cars, provide traps for blowing debris, and interfere
with snow plowing in northern climates. . . .

Wheel stops are often used along the side boundaries of a lot, where large landscaped areas extend
beyond the edge of pavement and an occasional override would present no significant hazard.

4.4.10. Supplemental Specifications and I mplementation

The following excerpts were taken from the1992 edition of the Transportation Planning Handbook,
published by the Institute of Transportation Engineers (pp. 190-191).

To be effective, a zoning code must specify the number of required spaces and must contain
sufficient controls to ensure that all the parking is convenient and usable.

Relation to Site and J oint Use

Zoning can aid sound community development if it causes all owners to provide adequate off-street
parking and loading facilities for their property. Each building may have its own parking lot or
garage, or the development of consolidated, common-use parking facilities may be more practical
and desirable in a business area. However, zoning should apply to business districts (including the
CBD) to the extent that each developer is required to contribute their fair share of the acquisition
and development cost for the parking needed to serve their property. This can be done by cash
contributions to an area parking fund in an amount equal to the estimated cost of providing the
specified number of spaces.

75

Stall Sizes and Access

Good driveway design is particularly important for the higher volume commercial driveways. In
areas with high pedestrian activity, it is good practice to restrict driveway widths and radii and to
meet sidewalk grades and a short distance in form the curb, thus creating a short hump. Such
measures ensure vehicular entry and exit at low speeds. In all other areas, use of greater widths,
large radii, and flat driveway slopes frequently requiring step-down curbs is desirable to speed up
the entry and exit of vehicles and thus increase ease and capacity of access. The recommended stall
and access dimensions for zoning or local administrative regulations are covered in Chapter 7 of the
Traffic Engineering Handbook, and in the ITE Committee 5D-8 Recommended Practice.

4.5. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable concepts. The more concerned you
are about your understanding of a topic, the more seriously you will want to approach the example
problem for that topic.

4.5.1. Adequacy Analysis

Over the course of an 8-hour day, 96 vehicles enter a local electronics stores parking lot. The
parking lot has 5 spaces and the average customer stays in the grocery store for 15 minutes.
Calculate the probability that an incoming car will be rejected.

Solution

First, we need to calculate the incoming flow rate. This is done as follows:

Q = 96 vehicles / 8 hours
Q = 12 vehicles/hour

Since we know the average vehicle is parked for 15 minutes, or 0.25 hours, we can calculate the
traffic load as follows.

A = Q T
A = 12 vehicles/hour 0.25 hours
A = 3 vehicle

Now that we have the traffic load, we can find the probability of rejection using the equation below.

P =
A
M
N!
1 +A +
A
2
2
+ +
A
M
N!


Where:

P = the probability of rejection,
A = the traffic load, and
M = the number of parking stalls.

76

P =
S
5
12u
1 +S +
S
2
2
+
S
3
6
+
S
4
24
+
S
5
12u
= u.11

Each entering vehicle has an 11% chance of being rejected. As a result, the electronics store loses
one out of each 10 customers entering their lot.

4.5.2. Space Requirements

A new sandwich shop is nearing completion and a parking lot needs to be designed. The
storeowners anticipate that, on the average 12-hour day, 360 vehicles will visit the sandwich shop.
The owners also anticipate that the average vehicle will remain parked for 10 minutes. How many
parking spaces need to be provided in order to guarantee that no more than 1 vehicle in 50 will be
unable to find a parking space?

Solution

First, we need to determine the traffic load. The incoming flow rate is calculated as shown below.

Q = 360 vehicles / 12 hours
Q = 30 vehicles/hour

The average parking duration is 10 minutes or 0.167 hours. The traffic load is calculated as shown
below.

A = 30 vehicles/hour 0.167 hours
A = 5

The maximum probability of rejection is 1 in 50, or 0.02. Using the probability of rejection
equation, we can solve for the number of spaces required.

P =
A
M
N!
1 +A +
A
2
2
+ +
A
M
N!


Where:

P = the probability of rejection (0.02),
A = the traffic load (5), and
M = the number of parking stalls.

Solving the equation for M yields a value of 10. The parking lot at the sandwich shop must have at
least 10 spaces, in order to meet the owners expectations.

Note that we have used average parking rates in this analysis. The sandwich shops particular
situation could dictate that more spaces are required. For example, say that the shop serves 80% of
its customers between 11 A.M and 2 P.M. The majority of the customers are arriving during a much
shorter time frame than the 12 hours that we used to find the incoming flow rate. In this case, more
parking spaces would be required.

77

4.6. Glossary

Accessible Path: a barrier-free path that persons with mobility or sensory impairments can safely
follow without obstacles or obstructions. Accessible paths are at least 5 feet wide and level.

Aisle: the portion of the parking lot devoted to providing immediate access to the parking stalls.
The recommended aisle width is dependent on the parking angle. A parking angle of 45
o
requires
an aisle width of 12 feet for a 9.0-foot stall, and a 90
o
parking angle requires an aisle width of 26
feet for a 9.0-foot stall. These dimensions lead to wall to wall distances of 47 feet for 45
o
and 63
feet for 90
o
.

CBD: Central Business District, typically ranging from an average size of 27 blocks (10,000-25,000
population cities) to over 200 blocks (cities over 1,000,000 population).

CBD Core: the heart of business, commercial, financial and administrative activity. Typically
ranges in size from an average of 7 blocks (10,000-25,000 population cities) to over 60 blocks
(cities over 1,000,000 population).

CBD Fringe: the area immediately surrounding the CBD, usually within 2-3 blocks.

Change of Mode: the transfer from one form of transportation to another. A park and ride lot is an
example of a change of mode, where an auto driver parks the vehicle and rides public transportation
for the remainder of the trip.

Cordon Count: the simultaneous counting of all traffic entering and leaving a given area such as a
CBD. It is generally a manual vehicle classification count, supplemented with automatic traffic
recorder counts.

Duration: the length of time a vehicle remains in one parking space.

Long Term Parking: parking with a duration of three hours or more.

Module: a complete module is one access aisle, servicing a row of parking on each side of the aisle.
Both the access aisle and the parking stalls serviced by that aisle are part of the module.

Outlying Business District: commercial area generally removed by a mile or more from a central
CBD.

Parking Accumulation: the total number of vehicles parked in a specific area (usually segregated
by type of parking facility) at a specific time.

Parking Demand: the number of vehicles with drivers desiring to park at a specific location or in a
general area. It is usually expressed as the number of vehicles during the peak-parking hour.

Parking Space or Stall: an area large enough to accommodate one parked vehicle with unrestricted
access (no blockage by another parked vehicle).

Parking Supply: the number of spaces available for use, usually classified by on-street curb
(metered and unmetered), lot and garage. Further differentiation of the types of parking is useful,
such as those available to the general public, and private spaces earmarked for a specific purpose
such as loading.

78

Parking Volume: the total number of vehicles that park in a study area during a specific length of
time.

Partial Module: one access aisle combined with a single one-side row of parking.

Short Term Parking: parking with a duration of three hours or less.

Stall Length: The longitudinal dimension of the stall, normally 18.5 feet.

Stall Width: The width of each parking space as measured crosswise to the vehicle. The most
common width is 8.5 to 9.0 feet.

Study Period: the time during which the parking study is conducted, usually between 10:00 A.M.
and 6:00 P.M. Increasing emphasis, however, is being placed on inclusion of the morning and
evening periods within the length of the study. Certain uses, such a theatres, may peak in the
evening hours, while residential parking demand peaks around 3:00 A.M.

Trip Purpose: the primary reason for the individuals journey to the study area. Typical purposes
include shopping, working, business, and recreation.

Turning Radii: The radius of the circle that is traveled by the design vehicle when completing a
turn. Large turning radii should be provided. These are a function of the parking angle and end
island design, but in general the turning radii should be at least 18 feet.

Turnover: the number of different vehicles parked at a specific parking space or facility during the
study period. Parking turnover measures utilization.

Van Accessible: a parking space that is at least 8 feet wide, with a minimum access aisle of 5 feet
along the right side of the parking space.























79

5. Roadway Design

5.1. I ntroduction

Roadway Design consists of many steps, beginning with route selection and ending with highway
construction. These steps include (but are not necessarily limited to):

Route selection
Surveys and preparation of maps
Determining design controls and criteria
Calculating horizontal and vertical curve parameters (addressed in the Geometric Design
chapter)
Selecting appropriate cross-section elements and their parameters (lane and shoulder widths,
slopes, etc.)
Drawing roadway profiles and cross sections
Developing traffic signal plans (addressed in the Signal Timing Design chapter)
Specifying earthwork quantities (excavation and/or fill requirements)

This chapter addresses all of the elements listed above. There are three additional elements of
roadway design not discussed in this chapter that are equally important:

Evaluating potential environmental impacts
Determining pavement characteristics
Preparing preliminary, intermediate and final cost estimates

As you can see, Roadway Design can be a long and complicated process. This chapter is designed
to help the undergraduate engineering student understand the major aspects of this complex topic.

5.2. Lab Exercises

This exercise will help increase your understanding of Roadway Design, by presenting a more
complicated problem that requires more thorough analysis.

5.2.1. Lab Exercise One: Roadway Design

In this lab exercise, you will be required to define the best alignment for a highway connecting two
or more separate points on a topographical map. The topographical map may be supplied by your
instructor, or you may be asked to find an appropriate map on your own. Your instructor will select
the points that must be connected by your roadway.

Your design should:

avoid damaging areas that are environmentally sensitive.
include plots of the vertical profile of your roadway and existing points on the ground.
obey AASHTO or local geometric design guidelines for setting grades, vertical curves, sight
distances, and drainage requirements.
include end area profiles (cut and fill) and complete earthwork calculations, using an
appropriate shrinkage factor.

Your final design should be accompanied by a brief report that summarizes the details and features
of your selected design.
80


Tasks to be Completed

Task 1. Acquire a topographic map of a suitable area. Discuss possible terminal points with your
instructor and define any areas that are environmentally sensitive or especially hazardous.

Task 2. Develop an optimum roadway alignment. Review AASHTO and/or local guidelines to
check for grades, vertical curves, sight distances, or drainage requirements that are not met by the
selected alignment. Modify as necessary.

Task 3. Prepare a vertical profile diagram of the ground and centerline for the selected route by
following the rules listed below.

State distances in feet from the end of the project.
Round horizontal distances to nearest 50 ft.
Calculate elevations along the alignment at every 100-ft station.

Task 4. At 100-ft stations, prepare cross-section diagrams of the ground and superimpose a highway
cross-section template. Calculate cut and fill areas. Section lengths and average end areas may be
used to calculate the cut and fill volumes.

Task 5. Complete the earthwork calculations using an appropriate shrinkage factor. Prepare a mass-
haul diagram and determine whether you will have to borrow or waste any material in order to
complete the project. Attempt to balance cut and fills.

5.3. Theory and Concepts

The topics shown below are some of the nuts and bolts of basic roadway design. Additionally,
because bike lanes have become a fundamental piece of most roadway construction or
reconstruction projects, we've included a brief discussion on their basic design. Topics followed by
the characters '[d]' include an Excel demonstration.

5.3.1. Route Selection and Alignment

Two of the most important considerations in selecting the route for a proposed highway are

1. the physical features of the area and
2. how these features relate to the geometric design controls.

Physical features that affect route selection include topography, ground (soil) conditions, and
surrounding land use. Any possible environmental impacts posed by construction of a new highway
must also be considered. First, the highway designer reviews topographic, geologic and soil maps as
well as available aerial photographs of the area.

The designer looks for conditions that will require sudden changes in alignment. For example, areas
that would necessitate connecting long straight sections with sharp curves should be avoided. Areas
that are subject to floods or avalanches make highway construction difficult, expensive and/or
unsafe.

Highway alignment is influenced by terrain. In general, the terrain or topography of an area is
classified as level, rolling or mountainous. In level terrain, selection of an alignment is influenced
by factors such as the cost of right-of-way, land use, waterways that may require expensive
81

bridging, existing roads, railroads, and subgrade conditions. In rolling terrain, a number of factors
need to be considered, including: grade and curvature, depths of cut and heights of fill, drainage
structures, and number of bridges. Grades are the greatest challenge in mountainous country.


Flat Rolling Mountainous

Typically, several preliminary maps are drawn showing various alignments. Selection of an
alignment is a trial and error process, as the proposed alignments are checked for compliance with
the horizontal and vertical control criteria. The selection of the final alignment is based on a
comparison of costs and environmental and social impacts.

5.3.2. Surveys and Maps

Acquisition of land for highway right of way requires a cadastral survey to establish existing
property lines and to establish and mark (monument) new boundaries. Cadastral land surveyors
identify and establish monuments that document the legal boundaries between public and private
lands. A topographic survey is made to establish the configuration of the ground and the location of
natural and man-made objects. A located centerline survey is generally made after the topographic
survey is completed and alternative alignments have been evaluated. The final alignment is
determined and then a survey of the centerline of the planned highway is conducted.

Many different types of maps are produced in the course of designing a highway. The most
common include:

Location or Vicinity Maps present the highway location in relation to surrounding physical
features.
Topographic Maps illustrate elevation with the use of contour lines and spot elevations.
Planimetric Maps show features such as roads, buildings, water, fences, vegetation, bridges,
railroads.
Detail Base Maps, generally produced at scales ranging from 1:200 to 1:1000, combine features
of the topographic and planimetric maps, and illustrate the following:

utilities (above and below ground)
recorded survey monuments
exposed geologic features
section corners, property corners, right of way monuments and other pertinent
boundaries or corners
proposed highway alignment features such as stations, bearings, and curve data

Keep in mind that each jurisdiction probably has its own map requirements and map terminology.

Examples of planimetric and cadastral survey maps are shown below.


82


Planimetric Map Cadastral Survey

5.3.3. Design Controls and Criteria

The physical design of a new highway is controlled by many factors. This module addresses factors
common to most of the functional classifications.

Design Speed

"Design speed is the maximum safe speed that can be maintained over a specified section of a
highway when conditions are so favorable that the design features of the highway govern."
(AASHTO, 1990). The selection of a suitable design speed will depend on the terrain and functional
class of the highway. Typical design speeds for freeways range from 50 mph to 70 mph depending
on the terrain type (level, rolling or mountainous).

Traffic Volume

The traffic engineers measure or indicator of traffic volume is the average daily traffic (ADT). The
ADT is the volume that results from dividing a traffic count obtained during a given time period by
the number of days in that time period. For example, given a traffic count of 52,800 vehicles that
was taken over a continuous period of 30 days, the ADT for this count equals 1,760 vehicles
(52,800 divided by 30). Another commonly used measure of traffic volume is the annual average
daily traffic (AADT), which is determined by dividing a count of the total yearly traffic volume by
365. The ADT and the AADT are not the same and its important to be aware of the time period
when calculating the ADT.

Design Hour Volume

The DHV is a two-way traffic volume that is determined by multiplying the ADT by a percentage
called the K-factor. Values for K typically range from 8 to 12% for urban facilities and 12 to 18%
for rural facilities. Neither the AADT nor the ADT indicate the variations in traffic volumes that
occur on an hourly basis during the day, specifically high traffic volumes that occur during the peak
hour of travel. The traffic engineer has to balance the desire to provide an adequate level of service
(LOS) for the peak hour traffic volume with proposing a design in which the highway capacity
would only be utilized for a few hours of the year. This is where the design hour volume (DHV)
comes in.

83

Directional Design Hour Volume

The directional design hour volume (DDHV) is the one-way volume in the predominant direction of
travel in the design hour, expressed as a percentage of the two-way DHV. For rural and suburban
roads, the directional distribution factor (D) ranges from 55 to 80 percent. A factor of
approximately 50 percent is used for urban highways. Keep in mind that the directional distribution
can change during the day. For example, traffic volume heading into the central business district is
usually higher than outbound traffic in the morning, but the reverse is true during the afternoon
peak hour. In summary, DDHV = ADT (or AADT) K D.

Vehicle Characteristics

Traffic engineers design highways that will accommodate all classes of vehicles. Width and height,
overhangs and minimum turning paths at intersections are important parameters to have at hand
during the design process. AASHTO states that the vehicle which should be used in designing for
normal operations is the largest one that represents a significant percentage of the traffic for the
design year.

Geometric Design Elements

Major elements of the highway design include stopping sight distance, passing sight distance, and
horizontal and vertical alignment. These elements are all addressed in the chapter "Geometric
Design".

5.3.4. Vertical Profile [d]

The vertical profile of a highway is made up of straight lines (grade lines) and curves, as shown in
the following figure.



The curves joining the grade lines are called vertical curves, and their function is to make a smooth
transition from one grade to another. Details of designing sag and crest vertical curves are presented
in the chapter "Geometric Design".

During or after completion of the detail base map (see Surveys and Maps), the traffic engineer
prepares a vertical profile of the alignment. Information needed to create a vertical profile includes
the vertical curve data and the elevations of the existing ground surface along the chosen route. The
first step is to draw the existing ground level along the centerline of the proposed alignment, with
elevation data on the vertical scale. Then draw the centerline of the alignment on the profile, as
shown below:

84



At a minimum, data included with the centerline are the elevation and stations of all points of
intersection and the lengths of vertical curves. The Excel demonstration provided with this concept
uses station notation and elevation information to develop a vertical crest curve and the
accompanying information.

5.3.5. Cross Section Elements [d]

Roadway cross sections include the elements shown below.



Travel Lanes

Historically, 10-foot lanes were standard for "first-class" paved highways. Today, public agencies
prefer lane widths of 12 feet for designing freeways and major traffic arterials. For two-lane
highways, a 24-foot wide roadway is necessary for buses and commercial vehicles to have
sufficient clearance. As demonstrated in the module on Capacity and Level of Service, lane width
affects highway capacity. Anything less than 12 feet tends to reduce speed. However, there are
instances when existing rights of way and development will control lane widths. These situations
must be carefully evaluated in order to develop the safest design. The number of lanes is determined
by estimates of traffic volumes and lane capacity, as discussed in the Capacity and Level of Service
module.

85

Slopes usually fall in both directions from the centerline of two-lane highways. Each half of a
divided roadway is sloped individually and may be crowned separately as well. Drivers barely
perceive cross slopes up to 2 percent; 1.5 to 2 percent are common cross slope values. Values
greater than 2 percent can be unsafe.

Shoulders

The shoulder is the portion of the roadway between the outer edge of the traveled lane and the
inside edge of the ditch, gutter, curb, slope or median (in divided roadways). As drivers, we all
know the benefits of having adequate shoulder widths when our cars break down. Shoulders also
provide lateral support for pavement subbase, base and surface courses. Shoulder widths are usually
determined by the traffic volume and the percent of heavy vehicles. Shoulders vary in width from 2
feet to 6 feet on non-freeway roadways and from 4 feet to 10 feet on freeways or other major roads.
Shoulders are sloped so that fluids drain away from the traveled roadway. In general, asphalt or
concrete-paved shoulders are sloped from 2 to 6 percent, gravel shoulders from 4 to 6 percent and
turf shoulders at about 8 percent.

Sideslopes

The purpose of sideslopes is to provide a transition from the roadway shoulder to the original
ground surface. Foreslopes extend from the shoulder edge to a drainage ditch or directly to the
ground surface, depending on the terrain. Backslopes extend from the outside edge of the drainage
ditch to ground surface or to the "cut" surface of a roadside. AASHTO states that foreslopes steeper
than 3:1 (33%) are recommended only where conditions do not permit the use of flatter slopes.
Backslopes steeper than 3:1 may be difficult to maintain and need to be evaluated with regard to
slope stability.

5.3.6. Cut and Fill Sections

Cut Sections

A detailed engineering soils analysis of a proposed highway alignment is a crucial part of the
highway design process. The results of the soils analysis are used to develop the design details of
cut sections such as depth and slope of the cut. The engineer has to keep in mind that the volume of
excavation increases significantly as the depth of the cut increases, and therefore usually tries to
avoid excessive cut depths. Cut slopes are rarely steeper than 2:1 (2 units horizontal to 1 unit
vertical or about 27 degrees from horizontal) except in very competent materials such as solid rock.
AASHTO recommends that cut slopes steeper than 3:1 be evaluated with regard to soil stability and
traffic safety.

Fill Sections

The greatest amount of roadway construction in rural areas occurs on fill. In flat terrain, the
highway pavement should be elevated several feet above the original ground surface to aid
drainage. Slopes for fill should be determined in accordance with the guidelines discussed under
Cross Section Elements. It is desirable to keep the height of the fill section to 30 feet or less, with
20 feet being a preferred maximum. With fill heights greater than 20 feet, it may be more
economical to build a bridge, depending on the topography.



86

5.3.7. Earthwork [d]

One important aspect of roadway design is determining the amount of earthwork necessary on a
project. Earthwork includes the excavation of existing earth material and any placement of fill
material required for constructing the embankment. The manual method for determining earth
excavation and embankment amounts involves three steps: cross sections of the proposed highway
are placed on the original ground cross sections, the areas in cut and the areas in fill are calculated,
and the volumes between the sections are computed. Cut and fill are the terms that are usually used
for the areas of the section; the terms excavation and embankment generally refer to volumes. The
methods used to manually calculate cut and fill areas are presented in most surveying textbooks.

Mass diagrams (or mass-haul diagrams) are plots of the cumulative volumes of cut and fill along an
alignment. Typically, the mass diagram is plotted below a profile of the route, with the ordinate at
any station representing the sum of the volumes of cut and fill up to that station. An example of a
mass diagram is shown below, with its associated profile. Steps used to create a mass diagram are
presented in the Excel demonstration included with this page. The most economical way to handle
the distribution of earthwork volumes can be determined from the diagram.




87


The rising curve on the mass diagram indicates excavation and a descending curve indicates
embankment. If a horizontal line is drawn to intersect the diagram at two points, excavation and
embankment (adjusted for shrinkage) will be equal between the two stations represented by the
points of intersection. Such a horizontal line is called a balance line, because the excavation
balances the embankment between the two points at its ends.

Since the ordinates represent the cumulative volume of excavation and embankment, the total
volumes of excavation and embankment will be equal where the final ordinate equals the initial
ordinate. If the final ordinate is greater than the initial ordinate, there is an excess of excavation (as
shown in this mass diagram); if it is less than the initial ordinate, the volume of embankment is the
greater and additional material must be obtained to complete the embankment. Highway engineers
strive to balance the amount of cut and fill during a highway construction project to avoid costly
hauling of materials.

5.3.8. Designing Bike Lanes

The Intermodal Surface Transportation Efficiency Act (ISTEA) places increased importance on the
use of the bicycle as a viable transportation mode, and calls on each state Department of
Transportation to encourage its use. AASHTO's Guide for the Development of Bicycle Facilities is
the basic reference for bicycle facility designers. It has been adopted, in part or in its entirety, by
many state and local governments. The AASHTO bicycle guidelines state "all new highways,
except those where bicyclists will be legally prohibited, should be designed and constructed under
the assumption that they will be used by bicyclists."

On existing multi-lane arterials and collectors with relatively high motor vehicle volumes and/or
significant truck/bus traffic, a right (curb) lane wider than 12 feet is desirable to better
accommodate both bicyclists and motor vehicles in the same travel lane. AASHTO and the National
Advisory Committee on Uniform Traffic Control Devices suggest reducing the inside vehicle lanes
from 12 feet to 11 feet for the purpose of widening the right-hand lane for bicycle use. The
AASHTO bicycle guidelines recommend a "usable" curb lane width of 14 feet on road segments
where parking is not permitted in the curb lane. Usable width generally cannot be measured from
curb face to lane stripe, because adjustments must be made for drainage grates (even the "bicycle
safe" ones) and longitudinal joints between pavement and gutter sections. For instance, on those
road segments where no parking is allowed but drainage grates and the longitudinal joints are
located 18 inches from the curb face, the travel lane (from joint line to lane stripe) should be 14 feet
in width, reflecting the unsuitability of bicycle riding on the outside 18 inches of the roadway.

If parking is permitted in the curb lane, then the minimum width of the curb lane, from curb face to
through travel lane is 14 feet, with 15 feet being the desirable width. In this design situation, the
lane width is measured from the curb face, since parked motor vehicles can occupy the curb flag
(gutter section). Conversely, when bicycles travel directly adjacent to a curb, they cannot safely
operate in the gutter section. Wide curb lanes are not striped or generally promoted as "bicycle
routes", but are often all that is needed to accommodate bicycle travel. An example of a 151/2-foot
curb lane is shown below.

88



Bicycle lanes are constructed when it is desirable to delineate available road space for preferential
use by bicyclists or motorists and to provide for more predictable movements by each. Bicycle lane
markings can increase a bicyclist's confidence that motorists will not stray into his/her path of
travel. Likewise, passing motorists are less likely to swerve to the left out of their lane to avoid
bicyclists on their right. Bike lanes are generally established on urban arterials and sometimes on
urban collector streets. Bicycle lanes are delineated by painted lane markings. They should always
be one-way facilities and carry traffic in the same direction as adjacent motor vehicle traffic. Two-
way bicycle lanes on one side of the roadway are unacceptable because they promote riding against
the flow of motor vehicle traffic. Wrong-way riding is a major cause of bicycle accidents. Bicycle
lanes on one-way streets should be on the right of the street, except in areas where a bicycle lane on
the left will decrease the number of conflicts (e.g., those caused by heavy bus traffic).

5.4. Professional Practice

In order to supplement your knowledge about the various concepts within roadway design, and in
order to give you a glimpse of how these various topics are discussed in the professional
environment, we have included selected excerpts from professional design manuals.

5.4.1. Route Selection

The following excerpt was taken from page 226 of the 1990 edition of AASHTO's A Policy on
Geometric Design of Highways and Streets.

The topography of the land traversed has an influence on the alignment of roads and streets.
Topography does affect horizontal alignment, but it is more evident in the effect on vertical
alignment. To characterize variations, engineers generally separate topography into three
classifications according to terrain. Level terrain is that condition where highway sight distances, as
governed by both horizontal and vertical restrictions, are generally long or could be made to be so
without construction difficulty or major expense. Rolling terrain is that condition where the natural
slopes consistently rise above and fall below the road or street grade and where occasional steep
slopes offer some restriction to normal horizontal and vertical roadway alignment.

89

Mountainous terrain is that condition where longitudinal and transverse changes in the elevation of
the ground with respect to the road or street are abrupt and where benching and side hill excavation
are frequently required to obtain horizontal and vertical alignment.

Terrain classification pertains to the general character of a specific route corridor. Routes in valleys
or passes or mountainous areas that have all the characteristics of roads or streets traversing level or
rolling terrain should be classified as level or rolling. In general, rolling terrain generates steeper
grades, causing trucks to reduce speeds below those of passenger cars, and mountainous terrain
aggravates the situation, resulting in some trucks operating at crawl speeds.

5.4.2. Surveys and Maps

The following excerpt was taken from pages 1-2 of the 1996 Oregon Department of Transportation
Highway Design Manual.

Before a control traverse is run in the field, careful planning and research must be done. The county
surveyor's office contains a wealth of knowledge about existing monumentation along the course of
the project i.e., public land survey corners, property corners, GPS monuments, etc. Roadway
Engineering in Salem, Oregon has on file all of the right of way maps, transit notes and as-
constructed plans. For help with general orientation of a project, a 7.5 minute or 15 minute quad
sheet is helpful. The quad sheet will also show section corners that have been found or reset, to
assist with your field search.

With the above information in hand, some careful planning about the project is the next step.
Identify what will be used for the control of this project i.e., NGS or NOAA control monuments,
new GPS control points, existing right of way monuments or random traverse points with solar
observation for basis of bearing.

Identify which public land survey corners will be tied on the project. No less than 2 corners per
township will be tied unless it is determined that there is insufficient monumentation along the
project to meet this requirement. Public land survey corners will be tied on each end of the project.

Cadastral Survey

Acquisition of land for highway right of way requires a cadastral survey to establish existing
property lines and to establish and monument new boundaries. This work must be done in
compliance with the laws of the State of Oregon....

Topographical Survey

The topographic survey is made to establish the configuration of the ground and the location of
natural and man-made objects. A planimetric map made by Photogrammetry will be of value, but
some field survey work is usually necessary to complete the topographic map.

5.4.3. Design Controls and Criteria

Driver performance is one of the roadway design criteria that is not quantifiable, but is important.
The following excerpt on driver performance was taken from page 42 of the 1990 edition of
AASHTO's A Policy on Geometric Design of Highways and Streets.

An appreciation of driver performance is essential to proper highway design and operation. Design
suitability rests as much on the ability of the highway to be used safely and efficiently as on any
90

other criterion. When drivers use a highway designed to be compatible with their capabilities and
limitations, their performance is aided. When a design is incompatible with the attributes of drivers,
the chances for driver errors increase, and accidents and inefficient operation often result.

At the start of the 20th century, approximately 4 percent of American's population was 65 years of
age or older. Persons 65 years of age or older accounted for 15 percent of the driving age population
in 1986, and will increase to 22 percent by the year 2030.

Elderly drivers and pedestrians are a significant and rapidly growing segment of the traffic stream
with a variety of age-related sensory-motor impairments. As a group, they have the potential to
adversely affect the highway system's safety and efficiency... Thus, designers and engineers should
be aware of the problems and requirements of the elderly, and consider applying applicable
measures to aid their performance.

5.4.4. Horizontal and Vertical Alignment

The following excerpt about the combination of horizontal and vertical alignments in roadway
design was taken from page 297 of the 1990 edition of AASHTO's A Policy on Geometric Design of
Highways and Streets.

Coordination of horizontal alignment and profile should not be left to chance but should begin with
preliminary design, during which stage adjustments can readily be made. Although a specific order
of study for all highways cannot be stated, a general procedure applicable to most facilities can be
outlined.

The designer should use working drawings of a size, scale and arrangement so that he can study
long, continuous stretches of highway in both plan and profile and visualize the whole in three
dimensions. Working drawings should be of sufficiently small scale, generally 1 inch = 100 feet or
1 inch = 200 feet with the profile plotted jointly with the plan. A continuous roll of plan-profile
paper usually is suitable for this purpose.

After study of the horizontal alignment and profile in preliminary form, adjustments in each, or
both, can be made jointly to obtain the desired coordination. At this stage the designer should not be
concerned with line calculations other than known major controls. The study should be made
largely on the basis of a graphical analysis...

The coordination of horizontal alignment and profile from the viewpoint of appearance usually can
be accomplished visually on the preliminary working drawings. Generally, this visual method
results in a satisfactory product when done by an experienced designer. This means of analysis may
be supplemented by models or perspective sketches at locations where the effect of certain
combinations of line and grade are questionable.

5.4.5. Cross Sections

The following excerpt was taken from page 328 of the 1990 edition of AASHTO's A Policy on
Geometric Design of Highways and Streets.

Two-lane and wider undivided pavements on tangents or on flat curves have a crown or high point
in the middle and slope downward toward both edges. The downward cross slope may be a plane or
curved section or a combination of the two. With plane cross slopes, there is a cross slope break at
the crown line and a uniform slope on each side. Curved cross sections usually are parabolic, with a
slightly rounded surface at the crown line and increasing cross slope toward the pavement edge.
91

Because the rate of crown slope is variable, the parabolic section is described by the crown height,
i.e., the vertical drop from the center crown line to the pavement edge. The advantage of the curved
section lies in the fact that the cross slope steepens toward the pavement edge, thereby facilitating
drainage. The disadvantages are that curved sections are more difficult to construct, the cross slope
of the outer lanes may be excessive, and warping of pavement areas at intersections may be
awkward or difficult to construct.

On divided highways, each one-way pavement may be crowned separately, as on two-lane
highways, or it may have a unidirectional slope across the entire width of pavement, which is
almost always downward to the outer edge. Where freeze-thaw conditions are a problem, each
pavement of a divided highway should be crowned separately.

5.5. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable roadway design concepts. The
more concerned you are about your understanding of a topic, the more seriously you will want to
approach the example problem for that topic.

5.5.1. Traffic Volume

Consider a rural highway with a projected 20-year AADT of 40,000 vpd. For the type of highway
and region in question, it is known that peak-hour traffic currently is approximately 20% of the
AADT, and that the peak direction generally carries 65% of the peak-hour traffic. What is the
DDHV?

Solution

An approximate DDHV could be estimated as DDHV = AADTKD. K is the percentage of the
AADT that occurs in the peak hour and D is the directional distribution percentage.

Therefore, DDHV = 40,000 0.20 0.65 = 4000 vph

5.5.2. Vertical Alignment

A highway design includes the intersection of a +5.8% grade with a 2.9% grade at station 1052+75
at elevation 100.50 feet above sea level. Calculate the center line elevation along this highway for
every 100-ft station on a parabolic vertical curve of 600-ft length. Youll need to use concepts
introduced in the Geometric Design Chapter.

Solution

The VPC is at station 1052+75 3+00 = 1049+75

The elevation of the VPC equals the elevation of the VPI minus of the curve length times the
initial grade = 100.50 - (300 0.058) = 83.10 feet.

The VPT is at station 1052+75 + 3+00 = 1055+75

The elevation of the VPT equals the elevation of the VPI minus of the curve length times the
second grade = 100.50 - (300 0.029) = 91.80 feet.
92


The equation for a parabolic vertical curve is y = (r / 2) * x
2
+ g
1
* x + (elevation of VPC) where
y = station elevation, r = rate of change of the grade of the curve [(g
2
- g
1
) / (length of curve in
stations)] and x = stations beyond the VPC.

First, r = (-2.9 - 5.8) / 6 = -1.45% per station.

The computations are shown in the following table.

Station x x
2
(r/2) * x
2
g
1
* x
Elevation
VPC
Elevation
Curve
1049+75 0 0 0 0.00 83.10 83.10
1050+75 1 1 -0.73 5.80 83.10 88.18
1051+75 2 4 -2.90 11.60 83.10 91.80
1052+75 3 9 -6.53 17.40 83.10 93.98
1053+75 4 16 -11.60 23.20 83.10 94.70
1054+75 5 25 -18.13 29.00 83.10 93.98
1055+75 6 36 -26.10 34.80 83.10 91.80

5.5.3. Cross Sections

Right-of-way width is the sum of the cross section elements such as the number of lanes, shoulders,
ditches, and sideslopes. Determine the minimum width of the right of way for a four lane rural
highway (two lanes in each direction) with the following dimensions:

12foot lanes at a slope of 2%
6-foot shoulders at a slope of 6:1
Ditch width is 2 feet
10-foot foreslopes at a slope of 3:1
15-foot backslopes at a slope of 3:1.

The dimensions are measured along the surface of the element.

Solution

Using the pythagorean theorem, the width required for 4 lanes is 4 (12 / (12 + 0.022) 1/2) = 48
feet. Using the same procedure for the remaining elements:

Shoulder width = 2 (6 / (12 + (1/6) 2) 1/2) = 11.8 feet

Ditch width = 2 2 = 4 feet

Foreslope width = 2 (10 / (12 + (1/3) 2) 1/2) = 19 feet

Backslope width = 2 (15 / (12 + (1/3) 2) 1/2) = 28.5 feet

Total width = 48 + 11.8 + 4 + 19 + 28.5 = 111.3 feet



93

5.6. Glossary

AADT: Annual average daily traffic determined by dividing a count of the total yearly traffic
volume by 365. Units are vehicles per day.

ADT: Average daily traffic that results from dividing a traffic count obtained during a given time
period by the number of days in that time period.

Cadastral Survey: A survey that identifies the legal boundaries between public and private lands.

Design Speed: The maximum safe speed that can be maintained when conditions permit the design
features of the highway to govern.

D: Directional distribution. The percentage of the two-way traffic volume traveling in the direction
of interest, expressed as a decimal.

DHV: Design hour volume. The two-way traffic volume that is determined by multiplying the ADT
by a percentage factor called the K-factor (see below).

DDHV: Directional design hour volume expressed as vehicles per hour.

Functional Class: The classification of roadways by operational status such as freeway, arterial,
collectors and local.

K-factor: The proportion of daily traffic occurring during the peak hour, expressed as a decimal.
For design purposes, K represents the proportion of AADT occurring during the thirtieth highest
peak hour of the year.

Located Centerline Survey: A survey of the centerline of a planned or existing highway.

Planimetric Map: Illustrates features such as roads, buildings, water, fences, vegetation, bridges,
railroads.

Stations: Points along a line (usually a survey line) of equal distance designated either in feet or
meters. Points at multiples of 100 meters or feet are typically called full stations.

Topographic Map: Illustrates elevation with the use of contour lines and spot elevations.

Topographic Survey: A survey that establishes the configuration of the ground and the location of
natural and man-made objects.












94

6. Signal Timing Design

6.1. I ntroduction

There are innumerable locations around the country where roads intersect. Conflicting traffic
movements cannot share the same space at the same time. Because of their ability to separate traffic
movements in time, traffic signals are one of the most common regulatory fixtures found at
intersections.

Since traffic signals are so common, professional engineers are often expected to know the basics of
Signal Timing Design. Signal Timing Design, at its simplest level, involves finding the appropriate
duration for all of the various signal indications. The design process involves assembling the results
of several independent calculations.

This chapter introduces the fundamentals of Signal Timing Design by discussing each step of the
process in detail.

6.2. Lab Exercises

This exercise will help increase your understanding of Signal Timing Design, by presenting a more
complicated problem that requires more thorough analysis.

6.2.1. Lab Exercise One: Signal Timing and LOS

Your assignment is to determine the level of service (LOS) at which a local intersection is
functioning during its peak period. If the measured LOS is unacceptable, you are to propose
changes to the signal-timing plan that would correct the problem or show why an easy solution does
not exist.

Your instructor may select an intersection in your area for this analysis, or you may be allowed to
select an intersection. In addition, your instructor may indicate specific movements that are of
particular interest.

Tasks to be Completed

Task 1. Measure and record the dimensions of the study intersection. Prepare a dimensioned
drawing of the facility.

Task 2. During the peak period, record the average lengths of the various signal indications. You
should record enough information to accurately describe the current signal-timing plan for the
intersection.

Task 3. During the peak period, record the number of vehicles passing through the intersection
within each of the various movements. Make sure you record enough information to complete the
required LOS computations.

Task 4. Determine the LOS of the various intersection movements using the information you have
collected. Record any interesting observations.

Task 5. If the LOS analysis reveals a problem, propose a new signal-timing plan that would correct
the problem or show why an easy solution does not exist.

95

Task 6. Prepare a brief report that outlines your work and summarizes your conclusions and
proposals.

6.3. Theory and Concepts

Signal timing design incorporates several calculations that seem, at first, to be completely
independent. The results of these calculations, however, are all pieces of a larger puzzle, which is
assembled at a later stage. The concepts that are the most crucial to signal timing design are listed
below. Simply click on the links to explore these concepts.

6.3.1. Basic Timing Elements

Signalized intersections permit conflicting traffic movements to proceed efficiently and safely
through space that is common to those movements. This is accomplished by separating the
individual movements in time rather than in space.

The various movements are collected and allowed to move in turn, or in phases. Each phase of a
signal cycle is devoted to only one collection of movements. These movements are those that can
proceed concurrently without any major conflict. For example, the straight-through and right-turn
movements of a street can be permitted to use an intersection simultaneously without any danger to
the motorists involved. This might be one phase of a multi-phase cycle.

Some movements are allowed to proceed during a phase even though they cause conflicts.
Pedestrians are commonly allowed to proceed across intersections even though right-turn
movements are occurring. These movements are called permitted, while protected movements are
those without any conflicts.

In any case, the movements at an intersection can be grouped, and then these groups can be served
during separate phases.

The basic timing elements within each phase include the green interval, the effective green time, the
yellow or amber interval, the all-red interval, the intergreen interval, the pedestrian WALK interval,
and the pedestrian crossing interval. Each of these elements is described below.

The green interval is the period of the phase during which the green signal is illuminated.
The yellow or amber interval is the portion of the phase during which the yellow light is
illuminated.
The effective green time is contained within the green interval and the amber interval. The
effective green time, for a phase, is the time during which vehicles are actually discharging
through the intersection.
The all-red interval is the period following the yellow interval in which all of the intersection's
signals are red.
The intergreen interval is simply the interval between the end of green for one phase and the
beginning of green for another phase. It is the sum of the yellow and all-red intervals.
The pedestrian WALK interval is the portion of time during which the pedestrian signal says
WALK. This period usually lasts around 4-7 seconds and is completely encompassed within the
green interval for vehicular traffic. Some pedestrian movements in large cities are separate
phases unto themselves.
Finally, the pedestrian crossing time is the time required for a pedestrian to cross the
intersection. This is used to calculate the intergreen interval and the minimum green time for
each phase.
96

This brief look at the basic signal timing elements should help you navigate through the rest of
the signal timing design concepts. Please remember to visit the glossary if you aren't sure about
a definition.

6.3.2. Queuing Theory [d]

Queuing theory provides the design engineer with a traffic flow model that can be used in the
design of signalized intersections. Consider a simple situation in which traffic is arriving at an
intersection approach in a uniform manner, with equal and constant headways between each
vehicle. This constant flow rate is shown in the figure below.


Figure 1: Constant Arrival Flow

During the red interval for the approach, vehicles cannot depart from the intersection and
consequently, a queue of vehicles is formed. When the signal changes to green, the vehicles depart
at the saturation flow rate until the standing queue is cleared. Once the queue is cleared, the
departure flow rate is equal to the arrival flow rate. Figure 2 illustrates this behavior..


Figure 2: Departure Flow or Service Flow versus Time

The combined effect of the arrival and departure flow rates is illustrated by graphing queue length
versus time. During the red interval, the line of vehicles waiting at the intersection begins to
increase. The queue reaches its maximum length at the end of the red interval. When the signal
changes to green, the queue begins to clear as vehicles depart from the intersection at the saturation
flow rate. See the figure below.


Figure 3: Queue Length versus Time

97

There is another graph that allows us to glean even more information from our model. Imagine a
plot where the x-axis is time and the y-axis contains the vehicle numbers according to the order of
their arrival. Vehicle one would be the first vehicle to arrive during the red interval and would be
the lowest vehicle on the y-axis. If you were to plot the arrival and departure (service) times for
each vehicle, you would get a triangle as shown in figure 4 below.


Figure 4: Vehicles versus Time

While this graph may not seem informative at first, a second look reveals its insights. For a given
time, the difference between the arrival pattern and the service pattern is the queue length. For a
given vehicle, the difference between the service pattern and the arrival pattern is the vehicle delay.
In addition, the area of the triangle is equivalent to the total delay for all of the vehicles. See figure
5 below.


Figure 5: Graph Properties

As you would expect, the first vehicle to be stopped by the red signal experiences the most delay. In
addition, the queue is longest just before the green interval begins. Queuing theory provides a
foundation for the optimization of signal timing.

6.3.3. Design Process Outline

This page is meant to guide you through the design process. Knowing a little more about how these
various concepts are used together will make the individual concepts easier to understand.

Pre-Design Data Collection:

The design of signal timing schemes is a fairly simple, though multi-step process. First, you need to
know most of the roadway conditions surrounding the intersection you are working on. This
includes the number of lanes, the width of the lanes, the width of the intersection, the width of the
shoulders, and more. Second, you need to have information regarding the composition of the traffic,
such as the percentage of busses and the percentage of trucks within the traffic stream. You also
need to know the peak hour volumes and peak fifteen-minute volumes for all of the various
movements.

98

The Design Process:

The basic steps in the design process (assuming you are using Webster's method see Cycle Length
Determination module) are listed below. While this particular listing is oriented toward Webster's
method, most of the other methods incorporate the same concepts, but in a slightly different way.

1. Decide on a phasing plan.
2. Calculate the length of the intergreen period for each phase of your cycle.
3. Calculate the minimum green time for each phase based on the pedestrian crossing time.
4. Calculate or measure the saturation flow rate for each approach or lane.
5. Calculate the design flow rate for each approach or lane using the peak hour volume and peak
hour factor.
6. Find the critical movements or lanes, and calculate the critical flow ratios.
7. Calculate the optimum cycle length.
8. Allocate the available green time using the critical flow ratios from step six.
9. Calculate the capacity of the intersection approaches or lanes.
10. Check the capacities/design flow rates and green intervals/minimum green intervals. Adjust
your cycle timing scheme if necessary.

Even though this outline is tailored for Webster's method, you'll find that most of the other design
methods involve many of the same calculations. Refer to this roadmap frequently as you proceed
through this chapter, so that you can see how each calculation is related to the design process.

6.3.4. I ntergreen Time [d]

The intergreen period of a phase consists of both the yellow (amber) indication and the all-red
indication (if applicable). This phase is governed by three separate concepts: stopping distance,
intersection clearance time, and pedestrian crossing time, if there are no pedestrian signals.

The yellow signal indication serves as a warning to drivers that another phase will soon be receiving
the right-of-way. The intergreen interval, therefore, should be long enough to allow cars that are
greater than the stopping distance away from the stop-bar to brake easily to a stop. The intergreen
interval should also allow vehicles that are already beyond the point-of-no-return to continue
through the intersection safely.

This issue is called the"dilemma zone" concept. If the intergreen time is too short, only those
vehicles that are close to the intersection will be able to continue through the intersection safely. In
addition, only vehicles that are reasonably distant will have adequate time to react to the signal and
stop. Those who are in between will be caught in the "dilemma zone," and wont have enough time
to stop or safely cross the intersection. Figure 1 shows this situation graphically.


Figure 1: Dilemma Zone

The only responsible thing to do, it seems, is to eliminate the dilemma zone. This would allow any
vehicle, regardless of its location, to be able to safely stop or, alternatively, safely proceed during
the intergreen period. This is done by making sure that any vehicle closer to the intersection than its
minimum braking distance can safely proceed through the intersection without accelerating or
speeding.

99

First, we calculate the minimum safe stopping distance. The equation for this distance is given
below and a more detailed discussion of this distance can be found in the geometric design portion
of this website.

Minimum Safe Stopping Distance:

SB = 1.47 v
0
t
r
+
(1.47 v
o
)
2
(Su (f _u))


Where:

SD = Min. safe stopping dist. (ft)
V
o
= Initial velocity (mph)
t
r
= Perception/Reaction time (sec)
f = Coefficient of friction
G = Grade, as a percentage

Next, we calculate the time required for a vehicle to travel the minimum safe stopping distance and
to clear the intersection. This is simple kinematics as well.

Intersection Clearance Time:

T =
(SB + L + W)
(1.47 v
o
)


Where:

T = Intersection clearance time (sec)
V
o
= Initial velocity (mph)
L = Length of the vehicle (ft)
SD = Min. safe stopping dist. (ft)
W = Width of the intersection (ft)

Now that youve determined the first two elements of the intergreen period lengthstopping
distance and intersection clearance timeyou need to consider the pedestrians. The intergreen time
for intersections that have signalized pedestrian movements is the same as the intersection clearance
time.

If you have an intersection where the pedestrian movements are not regulated by a separate
pedestrian signal, you need to protect these movements by providing enough intergreen time for a
pedestrian to cross the intersection. In other words, if a pedestrian begins to cross the street just as
the signal turns yellow for the vehicular traffic, he/she must be able to cross the street safely before
the next phase of the cycle begins. The formula for this calculation is shown below.

Pedestrian Crossing Time:

PCT =
W
v


Where:

PCT = Pedestrian crossing time (sec)
100

W = Width of the intersection (feet)
V = Velocity of the pedestrian (usually 4 ft/sec)

Once you have considered the safety of both the vehicular traffic and the pedestrian traffic for the
given phase, you can choose the intergreen time. The intergreen time is equal to whichever is larger,
the pedestrian crossing time or the intersection clearance time.

As you know, the intergreen period is composed of the yellow interval and the all-red interval. The
allocation of the intergreen time to these separate intervals is a question that is answered best by
referring you to your local codes. In some areas, the yellow time has been standardized for several
speeds. This would make the all-red time the difference between the standard yellow time and the
intergreen time. One other option is to allocate all of the intergreen period as calculated to the
yellow interval. You could then tack on an all-red period as a little extra safety. This, however,
might increase delay at your intersection.

6.3.5. Pedestrian Crossing Time, Minimum Green I nterval [d]

The pedestrian crossing time serves as a constraint on the green time allocated to each phase of a
cycle. Pedestrians can safely cross an intersection as long as there are not any conflicting
movements occurring at the same time. (Permitted left and right turns are common exceptions to
this rule.) This allows pedestrians to cross the intersection in both the green interval and the
intergreen interval. Thus, the sum of the green interval and the intergreen interval lengths, for each
phase, must be large enough to accommodate the pedestrian movements that occur during that
phase.

At this point, two separate conditions arise. If you have an intersection in which the pedestrian
movements are not assisted by a pedestrian signal, you need to make sure that the green interval that
you provide for vehicles will service the pedestrians as well. In this case, the minimum green
interval length is somewhere between 4 and 7 seconds. You already took care of the pedestrian
crossing time considerations when you calculated the intergreen period length. (See the module on
the intergreen period.)

If, on the other hand, you plan to provide a pedestrian signal, you need to calculate the pedestrian
crossing time as described below. This will not only give you the information you need to program
the pedestrian signal, but it will also allow you to find the minimum green interval for your
vehicular movements as well.

We only need a few assumptions to calculate the pedestrian crossing time.

Assumptions:

The WALK signal will be illuminated for approximately 7 seconds.
A pedestrian will begin to cross the street just as the DON'T WALK signal begins to flash.
Pedestrians walk at an average pace of 4 feet/second.
The WALK interval must be completely encompassed by the green interval of the
accompanying vehicle movements.

Calculations:

The total time required for the pedestrian movements (T) is the sum of the WALK allowance (Z)
and the time required for a person to traverse the crosswalk (R).

101

R = (width of intersection, in feet) / (4 ft/sec)

T = Z + R

The pedestrian crossing time governs the minimum green time for the accompanying phase in the
following way. If the time it takes the pedestrian to traverse the crosswalk (R) is greater than the
intergreen time (I), the remainder of the time (Z+R-I) must be provided by the green
interval. Therefore, the minimum green interval length (g
min
) for each phase can be calculated using
the equation below.

g
min
= T - intergreen time (I) or g
min
= Z + R I

If the above equation results in a minimum green interval that is less than the WALK time (Z), the
minimum green interval length is equal to the WALK time (Z).

g
min
= Z

You now have the minimum length of the green interval for the vehicular movements, as governed
by the pedestrian movements. The WALK interval for the pedestrians is whatever you assumed, and
the DON'T WALK flashes for the remainder of the green and intergreen intervals.

Many design manuals suggest that the distance the pedestrian is assumed to travel can be reduced to
the distance between the curb and the center of the farthest lane. On another note, if the vehicular
traffic requires an extended green period, feel free to let the pedestrains partake of the extra time as
well.

6.3.6. Saturation Flow Rate and Capacity [d]

Saturation Flow Rate

Saturation Flow Rate can be defined with the following scenario: Assume that an intersections
approach signal was to stay green for an entire hour, and the traffic was as dense as could
reasonably be expected. The number of vehicles that would pass through the intersection during that
hour is the saturation flow rate.

Obviously, certain aspects of the traffic and the roadway will effect the saturation flow rate of your
approach. If your approach has very narrow lanes, traffic will naturally provide longer gaps between
vehicles, which will reduce your saturation flow rate. If you have large numbers of turning
movements, or large numbers of trucks and busses, your saturation flow rate will be reduced. Put
another way, the saturation flow rate (s) for a lane group is the maximum number of vehicles from
that lane group that can pass through the intersection during one hour of continuous green under the
prevailing traffic and roadway conditions. The saturation flow rate is normally given in terms of
straight-through passenger cars per hour of green. Most design manuals and textbooks provide
tables that give common values for trucks and turning movements in terms of passenger car units
(pcu).

Determining the saturation flow rate can be a somewhat complicated matter. The saturation flow
rate depends on roadway and traffic conditions, which can vary substantially from one region to
another. Its possible that someone in the area has already completed a measurement of the
saturation flow rate for an approach similar to yours. If not, you'll need to measure it in the field.
One other possibility, which is used quite frequently, is to assume an ideal value for the saturation
flow rate and adjust it for the prevailing conditions using adjustment factors. A saturation flow rate
102

of 1900 vehicles/hour/lane, which corresponds to a saturation headway of about 1.9 seconds, is a
fairly common nominal value. Design manuals usually provide adjustment factors that take
parameters such as lane-width, pedestrian traffic, and traffic composition into account.

Capacity

Capacity is an adjustment of the saturation flow rate that takes the real signal timing into account,
since most signals are not allowed to permit the continuous movement of one phase for an hour. If
your approach has 30 minutes of green per hour, you could deduce that the actual capacity of your
approach is about half of the saturation flow rate. The capacity, therefore, is the maximum hourly
flow of vehicles that can be discharged through the intersection from the lane group in question
under the prevailing traffic, roadway, and signalization conditions. The formula for calculating
capacity (c) is given below.

c =
g
C
s

Where:

c = capacity (pcu/hour)
g = Effective green time for the phase in question (sec)
C = Cycle length (sec)
s = Saturation flow rate (pcu/hour)

Capacity can be calculated on several levels, depending on the amount of information you want to
obtain. You could calculate the capacity for each individual lane, or you could lump the lanes
together and find the capacity of an entire approach. You need to decide what makes sense for your
situation.

Capacity can be used as a reference to gauge the current operation of the intersection. For example,
let us assume that you know the current flow rate for a lane group and you also know the capacity
of that lane group. If the current flow rate is 10% of the capacity, you would be inclined to think
that too much green time has been allocated to that particular lane group. You'll see other uses for
capacity as you explore the remaining signal timing design concepts.

6.3.7. Peak Hour Volume, Peak Hour Factor, Design Flow Rate

Peak Hour Volume

The peak hour volume is the volume of traffic that uses the approach, lane, or lane group in
question during the hour of the day that observes the highest traffic volumes for that
intersection. For example, rush hour might be the peak hour for certain interstate acceleration
ramps. The peak hour volume would be the volume of passenger car units that used the ramps
during rush hour. Notice the conversion to passenger car units. The peak hour volume is normally
given in terms of passenger car units, since changing turning all vehicles into passenger car units
makes these volume calculations more representative of what is actually going on.

The peak hour flow rate is also given in passenger car units/hour. Sometimes these two terms are
used interchangeably because they are identical numerically.




103

Peak Hour Factor

The peak hour factor (PHF) is derived from the peak hour volume. It is simply the ratio of the peak
hour volume to four times the peak fifteen-minute volume. For example, during the peak hour, there
will probably be a fifteen-minute period in which the traffic volume is more dense than during the
remainder of the hour. That is the peak fifteen minutes, and the volume of traffic that uses the
approach, lane, or lane group during those fifteen minutes is the peak fifteen-minute volume. The
peak hour factor is given below.

Peak Hour Factor:

PBF =
Peak houi volume
4 Peak fifteen minute volume


Design Flow Rate

The design flow rate or the actual flow rate, for an approach, lane, or lane group is the peak hour
volume (flow rate) for that entity divided by the peak hour factor. A simpler way to arrive at the
design flow rate is to multiply the peak fifteen-minute volume by 4. However you derive the figure,
most calculations, such as those that measure the current use of intersection capacity, require the
actual flow rate (design flow rate).

6.3.8. Critical Movement or Lane [d]

While each phase of a cycle can service several movements or lanes, some of these lanes will
inevitably require more time than others to discharge their queue. For example, the right-turn
movement of an approach may service two cars while the straight-through movement is required to
service 30 cars. The net effect is that the right-turn movement will be finished long before the
straight-through movement. What might seem to be an added complexity is really an opening for
simplicity. If each phase is long enough to discharge the vehicles in the most demanding lane or
movement, then all of the vehicles in the movements or lanes with lower time requirements will be
discharged as well. This allows the engineer to focus on one movement per phase instead of all the
movements in each phase.

The movement or lane for a given phase that requires the most green time is known as the critical
movement or critical lane. The critical movement or lane for each phase can be determined using
flow ratios. The flow ratio is the design (or actual) flow rate divided by the saturation flow rate. The
movement or lane with the highest flow ratio is the critical movement or critical lane. You will see
how this concept is applied in the cycle length and green split discussions.

6.3.9. Cycle Length Determination [d]

Once you know the total cycle length, you can subtract the length of the amber and all-red periods
from the total cycle length and end up with the total time available for green signal indications.
Efficiency dictates that the cycle length should be long enough to serve all of the critical
movements, but no longer. If the cycle is too short, there will be so many phase changes during an
hour that the time lost due to these changes will be high compared to the usable green time. But if
the cycle is too long, delays will be lengthened, as vehicles wait for their turn to discharge through
the intersection. Figure 1 provides a graphical portrayal of this phenomenon.

104


Figure 1: Cycle Length versus Delay

Several methods for solving this optimization problem have already been developed, but Websters
equation is the most prevalent. Webster's equation, which minimizes intersection delay, gives the
optimum cycle length as a function of the lost times and the critical flow ratios. Many design
manuals use Webster's equation as the basis for their design and only make minor adjustments to
suit their purposes. Webster's equation is shown below.

C
o
=
1.SL + S
1 -
v
s


Where:

C
o
= Optimum cycle length (sec)
L = Sum of the lost time for all phases, usually taken as the sum of the intergreen periods (sec)
V/s = Ratio of the design flow rate to the saturation flow rate for the critical approach or lane in
each phase

After you have calculated the optimum cycle length, you should increase it to the nearest multiple
of 5. For example, if you calculate a cycle length of 62 seconds, bump it up to 65 seconds. Once
you have done this, you are ready to go. If you know the intergreen times for all of the phases, you
can then calculate the total available green time and allocate it to the various phases based on their
critical movements. (See the module entitled green split determination.)

6.3.10. Green Split Calculations [d]

Once you have the total cycle length, you can determine the length of time that is available for
green signal indications by subtracting the intergreen periods from the total cycle length. But, the
result is useless unless you know how to allocate it to all of the phases of the cycle.

As explained in the module about critical movement analysis, the critical movements or lanes are
used to distribute the available green time among all of the phases. The flow ratio for a movement
or lane is the actual (design) flow rate, for that entity, divided by the saturation flow rate. The
critical flow ratio, which is the one that is important for this calculation, is the flow ratio for the
critical movement or lane.

Green time is allocated using a ratio equation. Each phase is given a portion of the available green
time that is consistent with the ratio of its critical flow ratio to the sum of all the critical flow
ratios. This calculation is simple to do and hard to say, which makes it refreshingly different from
most of the other calculations we encounter in engineering.

105

The proportion of the available green time that should be allocated to phase "i" can be found using
the following equation:

g
I
=
[
v
s

v
s
u
T


Where:

g
i
= The length of the green interval for phase "i" (sec)
(V/s)
i
= The critical flow ratio for phase "i"
G
T
= The available green time for the cycle (sec)

You now have the length of the green interval for each phase of your cycle. At this point, you might
want to look at the timing adjustments module.

6.3.11. Timing Adjustments

Once you have calculated the lengths of the minimum green intervals, green intervals, and
intergreen intervals, as well as the design flow rates and capacities for each of your phases; it is
time to ask yourself whether or not your results actually work.

The first and most obvious check involves the green intervals. Check the length of the green
interval for each phase. If it is not greater than the length of the phase's minimum green interval,
you need to bump up the cycle length and add green time to that phase until the green interval is
equal to or greater than the minimum.

The second check involves capacity. If the capacity of a particular phase is below the design flow
rate for that phase, you should back-calculate the effective green time that would allow the phase to
run at the design flow rate. Once again, simply increasing the cycle length and allocating more time
to the green interval of the troubled phase will solve the problem.

Webster noted that the cycle length can vary between 0.75C
o
and 1.5C
o
without adding much delay,
so don't worry too much about adding a second or two to the nominal cycle length.

6.3.12. Computing Delay and LOS [d]

One way to check an existing or planned signal timing scheme is to calculate the delay experienced
by those who are using, or who will use, the intersection. The delay experienced by the average
vehicle can be directly related to a level of service (LOS). The LOS categories, which are listed
below, contain information about the progression of traffic under the delay conditions that they
represent. This allows you, as a designer or evaluator, to visualize and understand the traffic flow
conditions surrounding an intersection, even though the intersection might still be on the drawing
board.

The first step in the LOS analysis is to calculate the average delay per vehicle for various portions
of the intersection. You might be interested in the LOS of an entire approach, or alternatively, you
might be interested in the LOS of each individual lane. The equation for the average vehicle delay is
given below.



106

Average Stopped Delay Per Vehicle:

u =
_u.S8C[1 -
g
C

2
_
j1 -
g
C
X[
+ 17SX
2
_(X - 1) + _(X -1)
2
+ _16
X
C
]_
1
2
_

Where:

d = Average stopped delay per vehicle for the lane or lane group of interest (sec)
C = cycle length (sec)
g/C = green ratio for the lane or lane group
g = The effective green time for the lane or lane group (sec)
X = V/c ratio for the lane group
V = The actual or design flow rate for the lane or lane group (pcu/hour)
c = Capacity of the lane group (pcu/hour)

This equation predicts the average stopped delay per vehicle by assuming a random arrival pattern
for approaching vehicles. The first term of the equation accounts for uniform delay, or the delay that
occurs if arrival demand in the lane group is uniformly distributed over time. The second term of
the equation accounts for the incremental delay of random arrivals over uniform arrivals, and for the
additional delay due to cycle failures.

As was mentioned before, the level of service for signalized intersections is defined in terms of
average stopped delay per vehicle. This delay is directly related to the driver's level of discomfort,
frustration, fuel consumption, and loss of travel time. The following paragraphs describe the various
LOS categories.

Level of Service A: Operations with low delay, or delays of less than 5.0 seconds per vehicle. This
LOS is reached when most of the oncoming vehicles enter the signal during the green phase, and
the driving conditions are ideal in all other respects as well.

Level of Service B: Operations with delays between 5.1 and 15.0 seconds per vehicle. This LOS
implies good progression, with some vehicles arriving during the red phase.

Level of Service C: Operations with delays between 15.1 and 25.0 seconds per vehicle. This LOS
witnesses longer cycle lengths and fair progression.

Level of Service D: Operations with delays between 25.1 and 40.0 seconds per vehicle. At this
LOS, congestion is noticeable and longer delays may result from a combination of unfavorable
progression, long cycle lengths, and high V/c ratios.

Level of Service E: Operations with delay between 40.1 and 60.0 seconds per vehicle. This LOS is
considered unacceptable by most drivers. This occurs under over-saturated intersection conditions
(V/c ratios over 1.0), and can also be attributed to long cycle lengths and poor progression.

As you can see by now, the LOS illuminates the qualitative aspects of signal operation.

6.4. Professional Practice

Engineering can occasionally seem like a double major. You are expected to learn the theories and
concepts while in school, and then how things are actually done while on the job. In an effort to
bring these two aspects of engineering together for you, we have included excerpts from real design
107

manuals and other professional references as used by professional engineers. This allows you to
learn about the theory, but also to see how that theory is really applied.

The professional practice materials were taken from several different design manuals and
references. The code in your area may differ somewhat from the excerpts presented here, and
consequently, you should not reference these aids for any legitimate design work.

6.4.1. Design Process Outline

The following excerpts were taken from the 1991 Manual of Traffic Signal Design, 2nd Edition,
published by the Institute of Transportation Engineers (pp. 139-140)
General Considerations

The functional objective of signal timing is to alternate the right-of-way among the various phases
in such a way as to:

provide for the orderly movement of traffic.
minimize average delay to vehicles and pedestrians.
reduce the potential for accident-producing conflicts.
maximize the capacity of each intersection approach.

Unfortunately, these desirable attributes are not compatible. For example, delay may be minimized
by using as few phases as possible and the shortest practical cycle length. To reduce accident
potential requires fewer conflicts. Therefore, multiple phases and longer cycles are indicated.
Maximizing approach capacity requires the minimum number of phases to service the demand.
Accordingly, it is necessary to exercise engineering judgment to achieve the best possible
compromise among these objectives.

Timing for Pre-timed Control

There are several fundamental aspects of developing timing settings for pretimed signal control.
Some of there fundamentals are also applicable to actuated signal timing. The essential elements
include:

Number of timing plans.
Phase change intervals (yellow change plus all-red clearance)
Pedestrian timing requirements (including decision whether or not to use pedestrian indications)
Cycle length calculations
Split calculations
Flashing operation

To function effectively, pre-timed signal operations must take into account a number of local
intersection variables and hardware characteristics. It is therefore difficult to set forth
comprehensive guidelines to fit all possible situations. In many situations, it is desirable to monitor
the initial operations and adjust the timing settings to reflect the unique character of the intersection
and traffic flow.

6.4.2. I ntergreen Time

The following excerpt was taken from section 12-306.2 of the 1995 Idaho Transportation
Department Traffic Manual.

108

Vehicle Signal Change Interval

A vehicle signal change interval is that period of time in a traffic signal cycle between conflicting
green intervals. It is the time required to terminate one green indication before initiating a
conflicting green indication characterized by either a yellow interval or a yellow and all-red
interval. At the present time, there is considerable discussion of proper timing for change interval
with no recommended national practice adopted at this time.

The Idaho Motor Vehicle Code permits vehicles to enter the intersection on a yellow indication -
termed as a permissive yellow rule. These vehicles have lawfully entered the intersection and
accordingly are permitted to clear the intersection on the remaining yellow interval, an all-red
interval, or subsequent green indication. It should also be noted that Idaho Code permits vehicles to
enter the intersection on a green indication only after yielding the right-of-way to vehicles lawfully
within the intersection.

However, drivers are not always that observant of vehicles entering the intersection, particularly at
the far side of an intersection, which can lead to a conflict between the two vehicles.

The recommended formula for determining an appropriate change interval is:

Y + R = t +
v
2 a _ 2 u g
+
W + L
v


Where:

Y = length of the yellow interval
R = length of the all-red interval
t = driver perception/reaction time, recommended at 1.0 seconds.
V = velocity of approaching vehicle in feet/second, recommended that the 85 percentile signal
approach speed or the posted speed limit, converted to feet/second, be used.
a = vehicle deceleration rate, recommended as 10 feet per second
2
.
g = acceleration due to gravity at 32 feet per second
2
.
G = grade of the signal approach in percent divided by 100 or 2 percent is 0.02. A downhill
grade results in a negative term, i.e., -2 Gg.
W = width of intersection measured in feet from the near side stop line to the far edge of the
conflicting traffic lane along the vehicle path.
L = length of vehicle on clearance, recommended as 20 feet for passenger cars.

The above formula will determine the total change interval composed of a yellow interval and all-
red interval. The recommended minimum yellow intervals for traffic signals on the state highway
system in Idaho are as follows:

Approach Speed Standard Yellow I nterval All-Red Clearance I nterval
25 mph 3.2 sec Optional
30 mph 3.2 sec Optional
35 mph 3.2 sec Optional
40 mph 4.0 sec Required
45 mph 4.0 sec Required
50 mph 4.0 sec Required
55 mph 4.0 sec Required
> 55 mph 5.0 sec Required
109


The all-red clearance interval is determined by computing the change interval, "Y+R," noted above
and subtracting the standard yellow interval. The yellow interval has been standardized to present
the drivers the same yellow interval at comparable intersections. Additional clearance time is then
provided by adding an all-red interval for a longer change interval.

It should be recognized that longer change intervals detract from the available intersection green
time and are only needed if there are potential vehicle or vehicle-pedestrian conflicts between signal
phases. Note that the term (W+L)/V provides additional clearance time for a vehicle to clear the
intersection conflict zone. However, it is desirable to set a minimum yellow interval based on
engineering judgment and then adjust the change interval using an all-red interval if needed.

An all-red interval may be desirable at an intersection to provide additional time for a vehicle to
clear the intersection before there are conflicts with pedestrians or other vehicles. The need for an
all-red interval must consider a number of factors as follows:

Sight distance between vehicles or vehicle/pedestrian conflicts.
Phasing of signal indications resulting in location of clearing vehicle versus conflicting vehicle
or pedestrian movements.
Width of intersection or length of turning path of vehicle.
Start up delay of a conflicting pedestrian or vehicle movement plus the time to reach a point of
conflict with the clearing vehicle.
Speed of the approaching vehicle.
Required intersection clearance for a protected left-turn movement relative to position in
intersection versus conflicting pedestrians or vehicles.
Field observation of intersection operations relative to vehicle conflicts with only a yellow
interval and intersection accidents attributable to vehicle change interval.

The all-red intervals should not be less than 0.5 seconds and would normally be limited to 2.0
seconds. The determination of the all-red interval should be based on the factors noted above,
calculated values, intersection observations, vehicle clearance practices at comparable intersections,
and engineer judgment.

6.4.3. Pedestrian Crossing Time, Minimum Green Time

The following excerpt was taken from the 1990 Manual of Traffic Signal Design, Second Edition,
published by the Institute of Transportation Engineers (pp. 144-145).

Pedestrian Timing Requirements

Pedestrian movements across signalized intersections are typically accommodated by one of the
following operational options:

Pedestrians cross the street with the parallel vehicular green indication (no pedestrian signal
display).
Pedestrian movements are controlled by a concurrent separate pedestrian signal display.
Pedestrians move on an exclusive phase while all vehicular traffic is stopped.

The essential factor in any of these options is to provide adequate time for the pedestrian to enter
the intersection (walk interval) and to safely cross the street (pedestrian clearance interval). In cases
where there are no separate pedestrian displays and the pedestrian moves concurrently with
110

vehicular traffic on the parallel street, the time allocated to vehicular traffic must consider the time
required for pedestrians to react to the vehicular green indication and move across the street.

When separate pedestrian displays (WALK, DONT WALK) are used, the minimum WALK
interval generally ranges from 4 to 7 seconds (as recommended by the MUTCD 4D-7). This allows
the pedestrian ample opportunity to leave the curb before the pedestrian clearance interval
commences. Various research studies have indicated that when there are fewer than 10 pedestrians
per cycle, the lower 4 second WALK interval is usually adequate.

The MUTCD mandates that a pedestrian clearance interval always be provided where pedestrian
indications are used. During this interval, a flashing DONT WALK indication is displayed long
enough to allow the pedestrian to travel from the curb to the center of the farthest travel lane before
opposing vehicles receive a green indication. Some agencies terminate the flashing DONT WALK
and display a steady DONT WALK at the onset of the yellow vehicular change interval. This
encourages those pedestrians still in the crosswalk to complete the crossing without delay. The
calculation of the pedestrian clearance time therefore includes the yellow change interval. That is,
the pedestrian clearance time equals the flashing DONT WALK plus the yellow change interval. . . .

The typical walking speed of 4 ft/s, as cited in the MUTCD, is assumed to represent the "normal"
pedestrian. There are, however, various categories within the general population that walk at a
slower rate. For example, some female pedestrians walk slower than some male pedestrians; very
young children, the elderly, and the handicapped also walk at a slower rate. Research on pedestrian
characteristics verify that over 60% of all pedestrians move slower than 4 ft/s and 15% walk at or
below 3.5 ft/s.

Although this may imply that the lower walking speed (3.5 ft/s) should be used in calculating the
pedestrian timing, many engineers argue that the slower rate creates longer cycle lengths, ultimately
resulting in longer vehicular delays.

6.4.4. Capacity/Saturation Flow Rate

The following excerpts were taken from the 1994 Highway Capacity Manual, published by the
Transportation Research Board.

Capacity at signalized intersections is based upon the concept of saturation flow and saturation flow
rate. Saturation flow rate is given the symbol s and is expressed in units of vehicles per hour of
effective green time (vphg) for a given lane group.

The flow ratio for a given lane group is defined as the ratio of the actual or projected demand flow
rate for the lane group (v
i
) to the saturation flow rate (s
i
). The flow ratio is given the symbol (v/s)
i

(for lane group i).

The capacity of a given lane group may be stated as

c
I
= s
I
[
g
I
C


where;

c
i
= capacity of lane group i, vph,
s
i
= saturation flow rate for lane group i, vphg,
g
i
/C = effective green ratio for lane group i.

111

Saturation flow rate is defined as the flow rate per lane at which vehicles can pass through a
signalized intersection in such a stable moving queue. By definition, it is computed as

s =
S6uu
h


where;

s = saturation flow rate (vphgpl),
h = saturation headway (sec),
3,600 = number of seconds per hour.

6.4.5. Peak Hour Volume, Design Flow Rate, PHF

The following excerpts were taken from the 1994 Highway Capacity Manual, published by the
Transportation Research Board.

Peak Hour and Design Hour

Capacity and other traffic analyses focus on the peak hour of traffic volume, because it represents
the most critical period for operations and has the highest capacity requirements. The peak hour
volume, however, is not a constant value from day to day or from season to season.

If the highest hourly volumes for a given location were listed in descending order, a large variation
in the data would be observed, depending on the type of route and facility under study.

Rural and recreational routes often show a wide variation in peak-hour volumes. Several extremely
high volumes occur on a few selected weekends or other peak periods, and traffic during the rest of
the year is at much lower volumes, even during the peak hour. This occurs because the traffic
stream consists of few daily or frequent users; the major component of traffic is generated by
seasonal recreational activities and special events.

Urban routes, on the other hand, show little variation in peak-hour. . . .

The relationship between the 15-min flow rate and the full hourly volume is given by the peak hour
factor, defined in Part A of this chapter (see below).

Whether the design hour was measured, established from the analysis of peaking patterns, or based
on modeled demand, the peak-hour factor (PHF) is applied to determine design hour flow rates.

Peak-hour factors in urban areas generally range between 0.80 and 0.98. Lower values signify
greater variability of flow within the subject hour, and higher values signify little flow variation.
Peak-hour factors over 0.95 are often indicative of high traffic volumes, sometimes with capacity
constraints on flow during the peak hour.

(Description of PHF from Part A, as referred to above.)

Peak rates of flow are related to hourly volumes through the use of the peak-hour factor. This factor
is defined as the ratio of total hourly volume to the peak rate of flow within the hour:

PBF =
Bouily volume
Peak iate of flow (within the houi)

112


If 15-min periods are used, the PHF may be computed as

PBF =
v
4 v
15


Where

PHF = peak-hour factor,
V = hourly volume (vph), and
V
15
= volume during the peak 15 min of the peak hour (veh/15 min).

Where the peak-hour factor is known, it may be used to convert a peak-hour volume to a peak rate
of flow, as follows (equation 2-3):

v =
v
PBF


Where

v = rate of flow for a peak 15-min period (vph),
V = peak-hour volume (vph), and
PHF = peak-hour factor.

Equation 2-3 need not be used to estimate peak flow rates where traffic counts are available. The
chosen count interval must allow the identification of the maximum 15-min flow period. The rate
may then be directly computed as 4 times the maximum 15-min count.

Many of the procedures use this conversion to allow computations to focus on the peak flow period
within the peak hour.

6.4.6. Critical Movement or Lane

The following excerpt was taken from the 1995 Canadian Capacity Guide for Signalized
Intersections, Second Edition, published by the Institute for Transportation Engineers (District 7 -
Canada), (p. 46.)

Critical Lanes

The analysis and evaluation of signalized intersections, including most planning tasks, proceed on a
lane-by-lane basis. Not all the lanes, however, are equally important. Normally, in every phase there
is only one lane for which the relationship between the arrival flow and saturation flow results in
the longest green interval requirement. Such lanes are called critical lanes. The number of critical
lanes equals the number of phases in a cycle and, together, they have a decisive influence on the
cycle time.

A critical lane can be recognized by the highest flow ratio in a given phase:

ycrit
j
= max (y
ij
) = max (q
ij
/S
ij
)

Where:
113


ycrit
j
= flow ratio for the critical lane in phase j
y
ij
= flow ratio for lane i in phase j
q
ij
= arrival flow in lane i discharging in phase j (pcu/h)
S
ij
= saturation flow in lane i discharging in phase j (pcu/h). . . .

6.4.7. Cycle Length Determination

The following excerpt was taken from section 12-306.3 of the 1995 Idaho Transportation
Department Traffic Manual.

Cycle Length

Cycle length is composed of the total signal time to serve all of the signal phases including the
green time plus any change interval. Longer cycles will accommodate more vehicles per hour but
that will also produce higher average delays.

The best way is to use the shortest practical cycle length that will serve the traffic demand. Vehicles
at a signal installation do not instantaneously enter the intersection. Early studies by Greenshields
found that the first vehicle had a starting delay of 3.7 seconds to enter the intersection with
subsequent vehicles requiring an average of 2.1 seconds each. Generally, vehicles will pass over an
approach detector with a headway of 2 to 2.5 seconds. For general calculation purposes, an average
time of 2.5 seconds per vehicle to enter the intersection is a conservative value. This value can be
used to estimate signal timing for planning purposes.

The cycle length includes the green time plus the vehicle signal change interval for each phase
totaled to include all signal phases. A number of methods have been used to determine cycle lengths
as outlined in the Highway Capacity Manual, ITE Manual on Traffic Signal Design, and ITE
Transportation and Traffic Engineering Handbook. Webster provided the basic empirical formula
that would minimize intersection delay as follows:

C =
1.S L + S
1.u -Y
I


Where:

C = optimum cycle length in seconds adjusted usually to the next highest 5 second interval.
Cycle lengths in the range of 0.75C to 1.5C do not significantly increase delay.
L = Unusable time per cycle in seconds usually taken as a sum of the vehicle signal change
intervals.
Y
i
= critical lane volume each phase/saturation flow

The saturation flow will be between 1500 and 1800 vehicles per hour. Refer to Highway Capacity
Manual. The "Y" value should be computed for each phase and totaled to arrive at Y
i
for all
phases.

Note: The traffic volumes used should be the predicted volumes at time of signal turn-on. The
volumes should also be the peak hour or peak fifteen-minute period for the cycle determination.

When the cycle length has been determined the vehicle signal changes are deducted giving the total
cycle green time which can be proportioned to each signal phase on the basis of critical lane
114

volumes. The individual signal phase times are then the proportioned time plus the vehicle change
interval on each phase.

To ensure that critical lane volumes are adequately served, a capacity check should be computed for
each green interval.

6.4.8. Green Split Calculations

This excerpt was taken from the1995 Canadian Capacity Guide for Signalized Intersections, Second
Edition, published by the Institute for Transportation Engineers (District 7 - Canada), (p. 58).

Green Intervals by Balancing Flow Ratios

This procedure uses flow ratios for the critical lanes. First, the total time available in the cycle for
the allocation of green intervals is determined as:

g
j
= c - I
j


Where:

g
j
= total green time available in the cycle (s)
c = selected cycle time (s) (See the explanation below.)
I
j
= intergreen period following phase j (s).

This total available green time is allocated in proportion to the flow ratio of the critical lane for the
corresponding phase and the intersection flow ratio:

g
j
=g
j
_
y
j
Y
]

Where:

g
j
= green interval for phase j (s)
y
j
= flow ratio for the critical lane in phase j
g
j
= total green time available in the cycle (s)
Y = intersection flow ratio (sum of the critical flow ratios for all phases).

(Explanation for selected cycle time)

The Canadians list several different methods for calculating the total cycle time. The engineer is
supposed to select the method that suits his/her purposes best.

6.4.9. Timing Adjustments

The following excerpt was taken from section 12-306.3 of the 1995 Idaho Transportation
Department Traffic Manual.

(This is a continuation of the CYCLE LENGTH discussion that was visited in the "cycle length
determination" professional practice page.)

To ensure that critical lane volumes are adequately served, a capacity check should be computed for
each green interval. This can be done by making the following computations for each phase:
115


1. For each signal phase, determine the critical lane.
2. Then for each signal phase, determine in that critical lane the vehicles served per cycle.
3. That phase minimum green time would be as follows:

Phase Minimum
Green Interval = Vehicles per cycle 1.1 2.1 sec + 3.7 sec
1.1 sec provides a 10% increase for capacity traffic fluctuations
2.1 sec is the average headway per vehicle
3.7 sec is the time delay to start a traffic queue

4. The total cycle length equals the sum of the phase minimum green intervals determined in item
no. 3.

The minimum green interval should be less than green intervals determined above, under the
Webster method. If not, the cycle length should be increased with additional time allocated to those
phases not meeting the capacity criteria.

(The Webster method portion, as referred to, was not included on this page.)

6.4.10. Computing Delay and LOS, Operational Analysis Outline

The theory and concepts module on this topic covered the LOS grades and their corresponding
traffic flow descriptions, and also presented a simple formula for the calculation of the delay. This
module discusses the longer and more complex process for calculating the delay and LOS of an
existing intersection. The excerpt below gives an overview of the process used in operational
analysis. Your text probably contains a detailed description of this process, and you would be wise
to look it over.

The following excerpt was taken from the 1994 edition of the Highway Capacity Manual, published
by the Transportation Research Board.

Operational Analysis

Operational analysis results in the determination of capacity and level of service for each lane group
as well as the level of service for the intersection as a whole. It requires that detailed information be
provided concerning geometric, traffic, and signalization conditions at the intersection. These may
be known for existing cases or projected for future situations. Because the operational analysis of
signalized intersections is complex, it is divided into five distinct modules, as follows:

1. Input Module: All required information upon which subsequent computations are based is
defined. The module includes all necessary data on intersection geometry, traffic volumes and
conditions, and signalization. It is used to provide a convenient summary for the remainder of
the analysis.
2. Volume Adjustment Module: Demand volumes are generally stated in terms of vehicles per
hour for a peak hour. The volume adjustment module converts these to flow rates for a peak 15-
min analysis period and accounts for the effects of lane distribution. The definition of lane
groups for analysis also takes place in this module.
3. Saturation Flow Rate Module: The saturation flow rate is computed for each of the lane groups
established for analysis. The flow rate is based upon adjustment of an "ideal" saturation flow
rate to reflect a variety of prevailing conditions.
116

4. Capacity Analysis Module: Volumes and saturation flow rates are manipulated to compute the
capacity and v/c ratios for each lane group and the critical v/c ratio for the intersection.
5. LOS Module: Delay is estimated for each lane group established for analysis. Delay measures
are aggregated for approaches and for the intersection as a whole, and levels of service are
determined.

6.5. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable concepts. The more concerned you
are about your understanding of a topic, the more seriously you will want to approach the example
problem for that topic.

6.5.1. I ntergreen Time

On your way home from work a light turns yellow ahead of you. You are too close to the
intersection to stop without a heroic effort, so you proceed toward the intersection, assuming that
you'll get through it before the opposing phase is unleashed. To your surprise, the intersection signal
turns red before you have made it to the stop-bar. Luckily you clear the intersection, but
unfortunately, the local sheriff witnessed your maneuver. As part of your plea for mercy, you
mention that the signal seems to have an inadequate intergreen period that produces a dilemma
zone. Your plea doesn't work and you resolve to dispute the matter in court.

You return to the intersection and measure the intergreen time, using a stopwatch. It turns out to be
6 seconds. You also note the speed limit (50 mph in this case), the width of the intersection (around
60 feet), and your car's length (18 feet). The approach to the intersection is level, and you assume
that the coefficient of friction is around 0.5. Because you were only paying casual attention to the
road when the incident occurred, you decide to use 1 second as your perception reaction time. Can
you successfully argue that a dilemma zone exists? If one exists, what should be done to the
intergreen time to fix the problem?

Solution

The first step in this analysis is to calculate the minimum stopping distance you had under the given
circumstances. The minimum safe stopping distance can be calculated using the formula below.

SB = 1.47 v
o
t
r
+
(1.47 v
o
)
2
Su (f _ u)


Placing the given information into the equation yields:

SB = 1.47 Su 1 +
(1.47 Su)
2
Su (u.S _ u)


Solving this equation gives us a stopping distance of 434 feet.

Next you must calculate the time required to traverse the sum of the stopping distance, intersection
width, and one car length. This will give you the intergreen time that is necessary for a car to safely
pass through the intersection from the point-of-no-return. The intersection clearance time is given
by the equation below.

117

T =
SB + L + W
1.47 v
o


Placing the given information into the equation yields:

T =
4S4 + 18 + 6u
1.47 Su


Solving this equation gives us an intersection clearance time of 7 seconds. Since the intersection
clearance time provided was only 6 seconds, and a full 7 seconds is required for a car to safely pass
through the intersection from the point-of-no-return, your claim that a dilemma zone exists is well
founded. To fix the problem, the city should increase the intergreen time by 1 second.

6.5.2. Pedestrian Crossing Time, Minimum Green I nterval

A senior citizen using a crosswalk at a local intersection was struck by a vehicle. Following the
incident, a number of other citizens complained that the allocated pedestrian crossing time was
insufficient at the intersection. You have been asked to evaluate the situation.

You estimate the width of the intersection as 60 ft and the average pedestrians pace as 4 ft/sec. You
also record the WALK time (10 sec), concurrent green interval length (14 sec), and the intergreen
time (6 sec). Can you prove that the green interval given to the concurrent vehicular movement was
insufficient based on the pedestrian crossing time? If this pedestrian movement has an extremely
low flow rate, which is why this incident didn't happened before, how would you correct the safety
problem without increasing the delay observed by the vehicular movements?

Solution

The first step in this solution is to calculate the time required for a person to cross the intersection.
This time can be calculated from the equation below.

R =
wiuth of inteisection
walking speeu of peison


Substituting the given information into this equation, we get a crossing time (R) of 15 seconds.
Next, the total time that should be devoted to pedestrians is calculated by adding the WALK time
(Z) to the crossing time (R). This gives us a required pedestrian total time of 25 seconds. Adding the
vehicular green interval length (14 sec) and intergreen times (6 sec) that were provided gives us a
value of 20 seconds. Since the pedestrian phase requires 25 seconds and is currently only given a
total time of 20 seconds, conflicting traffic will begin moving a full 5 seconds before the last
pedestrian has made it to safety. No wonder this incident occurred. This problem could be remedied
by increasing the length of the vehicular movement's green interval by 5 seconds. You might want
to prove this using the equation g = Z + R - I.

One way to alleviate the problem without increasing the cycle length is to reduce the WALK time
that is given to the pedestrians by 5 seconds. This would bring the required pedestrian total time to
the 20 seconds that is currently provided. This option is only available because the pedestrian flow
rate is very low for this intersection.



118

6.5.3. Capacity/Saturation Flow Rate

Your first assignment as a transportation engineer is to design a small signalized intersection. One
step in this design process requires that you find the saturation flow rate for the eastern approach.
You decide to try and calculate the saturation flow rate from field observations of an approach that
is part of an intersection very similar to yours. After recording the departure headways of the first
few discharging vehicles for several different cycles, you calculate the average headway and get a
value of 2.1 seconds.

Calculate the saturation flow rate. If the eastern approach to your intersection has an estimated
green time of 20 seconds and the total cycle length will be around 45 seconds, what is the
approximate capacity for the eastern approach?

Solution

Since we know, from queuing theory, that the vehicles in a queue will discharge at the saturation
flow rate, we can take the average headway of those vehicles and convert it into the saturation flow
rate. Since one vehicle entered the intersection every 2.1 seconds when the queue was discharging,
3600/2.1 vehicles would enter the intersection in an hour if the queue were long enough and the
approach was given a green signal for an entire hour. Therefore, our saturation flow rate is 1714
veh/hr. Since we weren't given any information to the contrary, we will assume that these vehicles
were all passenger cars and call our saturation flow rate 1714 pcu/hr.

To find the capacity, we first need to calculate the green ratio (g/C). If the length of the green
interval (g) is 20 seconds and the cycle length (C) is about 45 seconds, the green ratio will be about
0.44. Capacity is the product of the green ratio and the saturation flow rate. In this case, the capacity
of the eastern approach would be about 760 pcu/hr.

6.5.4. Peak Hour Volume, Design Flow Rate, PHF

It is commonly known in your area that the heaviest traffic flow rates occur between 4:00 PM and
6:30 PM. Your assignment for the day is to find the peak hour volume, peak hour factor (PHF), and
the actual or design flow rate for an existing one-lane approach. To do this, you obtain a click-
counter and position yourself at the intersection. For each fifteen-minute interval, you record the
numbers of right-turns, left-turns, straight-through trucks, and straight-through passenger cars. Your
tabulated values are as shown below.

Time I nterval Left Turns Right Turns ST Trucks ST Cars
4:00-4:15 5 10 6 30
4:15-4:30 6 15 8 26
4:30-4:45 4 7 10 35
4:45-5:00 7 16 8 40
5:00-5:15 10 13 6 49
5:15-5:30 9 12 12 55
5:30-5:45 14 15 8 65
5:45-6:00 12 12 10 50
6:00-6:15 10 9 8 39
6:15-6:30 9 12 4 30

119

If a truck is equal to 1.5 passenger cars and a right-turn is as well, and if a left-turn is equal to 2.5
passenger cars, then calculate the peak hour volume, peak hour factor (PHF), and the actual (design)
flow rate for this approach.

Solution

The first step in this solution is to find the total traffic volume for each 15 minute period in terms of
passenger car units. This is done by multiplying the number of trucks by 1.5, the number of right
turns by 1.5, and the number of left turns by 2.5. We then add these three numbers and the volume
of straight-through cars together to get the total volume of traffic serviced in each interval. Once we
have this, we can locate the hour with the highest volume and the 15 minute interval with the
highest volume. The peak hour is shown in blue below with the peak 15 minute period shown in a
darker shade of blue.

Time I nterval I nterval Volume (pcu)
4:00-4:15 67
4:15-4:30 76
4:30-4:45 71
4:45-5:00 94
5:00-5:15 103
5:15-5:30 114
5:30-5:45 135
5:45-6:00 113
6:00-6:15 90
6:15-6:30 77

The peak hour volume is just the sum of the volumes of the four 15 minute intervals within the peak
hour (464 pcu). The peak 15 minute volume is 135 pcu in this case. The peak hour factor (PHF) is
found by dividing the peak hour volume by four times the peak 15 minute volume.

PBF =
464
4 1SS
= u.86

The actual (design) flow rate can be calculated by dividing the peak hour volume by the PHF,
464/0.86 = 540 pcu/hr, or by multiplying the peak 15 minute volume by four, 4135 = 540 pcu/hr.

6.5.5. Critical Movement or Lane

As a transportation engineer about to embark on the cycle length and green split calculations, you
need to find the critical lane for each phase of a two-phase signal cycle. In this example problem we
will only focus on one phase. The approaches that are serviced in this phase will have two
lanes, one servicing left-turns and straight-through traffic, and the other servicing right-turns and
straight-through traffic. The design flow rates and saturation flow rates for each lane are given
below.

Lane Description Design Flow Rate Saturation Flow Rate
North-bound L,S 600 pcu/hr 1200 pcu/hr
North-bound R,S 500 pcu/hr 1700 pcu/hr
South-bound L,S 450 pcu/hr 1330 pcu/hr
South-bound R,S 720 pcu/hr 1600 pcu/hr
120


Which lane is the critical lane for this phase, and what is the critical flow ratio for this phase?

Solution

The critical lane is the lane that requires the most time to service its queue. It can be found by
locating the lane with the highest flow ratio (V/s). Simply calculate the flow ratio for each lane by
dividing the design flow rate by the saturation flow rate. Then find the lane with the largest flow
ratio.

Lane Description Flow Ratio
North-bound L,S 0.5
North-bound R,S 0.294
South-bound L,S 0.338
South-bound R,S 0.45

It looks like the north-bound left-turn and straight-through lane is the critical lane for this
phase. The critical flow ratio is just the flow ratio for the critical lane (0.5).

6.5.6. Cycle Length Determination

As part of a signal design team, you have been assigned to find the optimum cycle length for a
three-phase cycle. Field observations and calculations by yourself and others are the basis for your
work. So far, you know the critical flow ratio for each phase and the intergreen time for each
phase. Calculate the optimum cycle length for your signal, given the critical flow ratios and
intergreen times below. What would the optimum cycle length be if all of the critical flow ratios
were near zero?

Phase Number Critical Flow Ratio I ntergreen Time
1 0.233 6 sec
2 0.13 4 sec
3 0.256 7 sec

Solution

Webster's optimum cycle length equation, which is shown below, has two variables on the right-
hand side. These are the total cycle lost time (L) which is usually taken as the sum of all the
intergreen times, and the sum of all the critical flow ratios ((V/s)).

C
o
=
1.SL + S
1 - [
v
s



In this case, the sum of the intergreen period lengths is 17 seconds (L=17). The sum of the critical
flow ratios is 0.619. When we substitute these values into the equation above we obtain an optimum
cycle length of 80 seconds. If all of the critical flow ratios were very near zero, the denominator in
Webster's equation would approach unity and the optimum cycle length would be 31 seconds.



121

6.5.7. Green Split Calculations

Assuming that both of the critical movements in a two-phase cycle have the same saturation flow
rate, what percentage of the available green time would each phase receive, given the design flow
rates for the critical movements listed below?

Situation Number Phase 1 Flow Rate (pcu/hr) Phase 2 Flow Rate (pcu/hr)
1 500 250
2 400 100
3 90 30
4 100 80

Solution

The available green time is allocated based on the ratio of the critical flow ratios to the sum of the
critical flow ratios. However, in this case we can simplify the calculations because the saturation
flow rate is assumed to be identical for both of the critical movements. This means that the green
time is allocated according to the ratios of the design flow rates to the sum of the design flow rates.
This simplification is shown below.
g
I
=
u [
v
s

v
s


simplifies to
g
I
=
u v
I
v


Since we weren't given the available green time, we'll forget about it and focus on the ratios. For
situation number one, the design flow rate for the critical movement in phase one was 500 pce/hr
while the critical design flow rate for phase two was 250 pcu/hr. The sum of these flow rates is 750
pcu/hr. Hence, phase one will receive 67% (500/750) of the available green time, while phase two
will receive 33% (250/750). The results are tabulated below.

Situation
Number
Phase 1 Flow Rate
(pcu/hr)
Phase 2 Flow Rate
(pcu/hr)
Phase 1 % of G Phase 2 % of G
1 500 250 67 33
2 400 100 80 20
3 90 30 75 25
4 100 80 56 44

6.5.8. Timing Adjustments

You have just finished allocating the available green time to a two-phase cycle. The actual (design)
flow rates for the two critical lanes are 350 pcu/hr and 700 pcu/hr respectively. The optimum cycle
length was 55 seconds and of the available green time, 14 seconds were allocated to phase one (350
pcu/hour) and 27 seconds were allocated to phase two (700 pcu/hr). Both phases have intergreen
intervals of 6 seconds. The lengths of the pedestrian WALK intervals for phases one and two are 10
seconds and 16 seconds respectively. The width of the intersection for phases one and two is 48 ft
and 68 ft respectively. Assume the saturation flow rate is 1900 pcu/hr for both lanes. Does this
timing scheme require any adjustments? If so, what should the final signal timing plan be?

122

Solution

The first thing we will check is the capacities of the two critical lanes. We'll calculate the capacities
by multiplying the green ratio (g/C) by the saturation flow rate (s). For phase one, the green ratio is
14/55 and the saturation flow rate is 1900 pcu/hr. This gives a capacity of 484 pcu/hr, which is more
than adequate to handle the 350 pcu/hr design flow rate. Phase two has a capacity of 933 pcu/hr,
which is also more than adequate to handle its design flow rate.

Next, we will check the minimum length of the green interval based on pedestrian movements. In
phase one, the WALK interval is 10 seconds long and the crossing time is 48 ft/(4 ft/s), or 12
seconds. The total time required for pedestrians is 22 seconds. The vehicular movement provides
only 14 seconds of green and 6 seconds of intergreen. Thus, the total time before the next phase
begins is only 20 seconds. To remedy this, two seconds should be added to both the total cycle time
and the green interval for phase number one. Pedestrians in phase number two receive 16 seconds
of WALK time and require 68 ft/(4 ft/sec) = 17 seconds of crossing time. The total time required for
the pedestrians in phase number two is, therefore, 33 seconds. The green interval and intergreen
interval for phase number two add up to 33 seconds, which perfectly matches the pedestrian
crossing time.

As it turns out, phase one's green interval needed to be increased by two seconds in order to serve
the pedestrian movements. The total cycle length and phase one's green interval were both increased
by two seconds, while all the other signal timing variables were left untouched.

6.5.9. Computing Delay and LOS

Because several complaints have been received from local drivers, you have been assigned to
determine the level of service for a one-lane approach at a local signalized intersection. The cycle
length is 80 seconds, and 30 seconds of effective green time are enjoyed by the approach in
question. The actual flow rate of traffic through the approach is 400 pcu/hr and the saturation flow
rate for the approach is 1750 pcu/hr. What is the LOS for this approach?

Solution

Before we can calculate the delay for the approach, we need to know the green ratio (g/C), the
capacity (c), and the ratio V/c (X). The green ratio for this approach is 30/80 or 0.375. The capacity
is (g/C) s which equals 0.375 1750 or 656 pce/hr. The ratio V/c is 400/656 or 0.609. The
average vehicle delay is given by the equation below.

u =
_u.S8C[1 -
g
C

2
_
j1 -
g
C
X[
+ 17SX
2
_(X - 1) + _(X -1)
2
+ _16
X
C
]_
1
2
_

By placing the values calculated above into the equation, we obtain an average vehicle delay of 23
seconds. This corresponds to the level of service grade "C".

6.6. Glossary

Actual Flow Rate: The design flow rate, or the maximum flow that is expected to use the
intersection. See the theory and concepts modules on peak hour volume, design flow rate, and PHF.

All-red interval: Any portion of a signal cycle in which a red indication is observed by all
approaches.
123


Approach: The portion of an intersection leg that is used by traffic approaching the intersection.

Capacity: The maximum number of vehicles that can reasonably be expected to pass over a given
roadway or section of roadway, in one direction, during a given time period and under the
prevailing roadway, traffic, and signalization conditions.

Change interval: Identical to the intergreen interval.

Clearance interval: Identical to the all-red interval.

Critical Flow Ratio: The flow ratio of the critical lane group within a phase. The actual or design
flow rate for the critical movement divided by the saturation flow rate for that movement.

Critical Movement or Lane: The lane or movement for each phase, depending on how you choose
to subdivide you intersection, that requires the most green time.

Critical Volume: A volume, or combination of volumes, which produces the greatest utilization of
capacity for the street or lane in question, given in terms of passenger car units per hour per lane or
mixed vehicles per hour per lane.

Cycle: A complete sequence of signal indications. Each phase has been serviced and the cycle is
beginning again.

Cycle Length: The time required for one full cycle of signal indications, given in seconds.

Delay: The stopped time per vehicle (in seconds per vehicle), usually calculated separately for each
lane group.

Design Flow Rate: Identical to the actual flow rate.

Effective Green Time: The green time that is actually used by traffic. Some lost time occurs
initially while traffic responds to the green signal and begins to accelerate. Some time is also lost
during the intergreen period as vehicles stop in anticipation of the next phase.

Flow Rate: The rate, in vehicles per hour or passenger car units per hour, at which traffic is
entering an intersection.

Flow Ratio: The ratio of the actual flow rate to the saturation flow rate.

Green I nterval: The portion of a signal phase in which the green signal is illuminated.

Green Ratio: The ratio of the effective green time to the cycle length.

Green Time: The length of the green interval and its change interval, given in seconds.

Hourly Volume: The number of mixed vehicles that traverse a given section of lane or roadway
during an hour.

I ntergreen: The time interval between the end of a green indication for one phase and the
beginning of green for the next phase.

124

I ntersection Flow Ratio: The sum of all the critical flow ratios--one from each phase.

Lane Group: Any group of lanes. Lanes can be combined during the signal timing design process
in order to simplify the calculations.

Legs (intersection): The portions of the intersecting streets or roadways that are within close
proximity to the actual intersection.

Level of Service (LOS): A measure of the operating conditions of an intersection. See the theory
and concepts modules for more detail.

Lost Time: The time during a given phase in which traffic could be discharging through the
intersection, but is not. This is the period during the green interval and change intervals that is not
used by discharging traffic.

Passenger Car Units: A unit of measure whereby large trucks and turning movements are
converted to passenger cars using multiplication factors. This allows you to deal with mixed traffic
streams more accurately than if you had assumed all vehicles were created equal.

Peak-Hour: The hour of the day that observes the largest utilization of capacity, or the hour of the
day in which the largest number of vehicles use the intersection approach or lane of interest.

Peak-Hour Factor: The ratio of the number of vehicles entering an approach during the peak hour
to four times the number of vehicles entering during the peak 15 minute period. In the absence of
field information, a value of 0.85 is normally used.

Pedestrian Crossing Time: The time that is required for a pedestrian to cross the intersection.

Phase: The portion of the cycle that is devoted to servicing a given traffic movement.

Phase Sequence: The predetermined order in which the phases of a cycle occur.

Queue: A closely spaced collection of vehicles.

Roadway Conditions: The physical aspects of the roadway, such as lane-width, number of lanes,
easements, bike lanes, shoulder width, and any other aspect of the roadway.

Saturation Flow Rate: The maximum number of vehicles from a lane group that would pass
through the intersection in one hour under the prevailing traffic and roadway conditions if the lane
group was given a continuous green signal for that hour. This assumes that there is a continuous
queue of vehicles with minimal headways.

Signalization Conditions: All the various aspects of the signal system, including timing, phasing,
actuation, and so on.

Split: A percentage of a cycle length allocated to each of the various phases in a signal cycle.

Traffic Conditions: The qualities of traffic, such as traffic speed, density, vehicle types, and traffic
flow rate.



125

7. Traffic Flow Theory

7.1. I ntroduction

Traffic Flow Theory is a tool that helps transportation engineers understand and express the
properties of traffic flow. At any given time, there are millions of vehicles on our roadways. These
vehicles interact with each other and impact the overall movement of traffic, or the traffic flow.

Whether the task is evaluating the capacity of existing roadways or designing new roadways, most
transportation engineering projects begin with an evaluation of the traffic flow. Therefore, the
transportation engineer needs to have a firm understanding of the theories behind Traffic Flow
Analysis.

This chapter is designed to help the undergraduate engineering student understand the fundamentals
of Traffic Flow Theory.

7.2. Lab Exercises

These exercises will help increase your understanding of Traffic Flow Theory, by presenting more
complicated problems that require more thorough analysis.

7.2.1. Lab Exercise One: Flow Models [d]

The attached data sets reflect field data taken at two sites in Huskytown. You are an engineering
intern and you have been asked by your supervisor to analyze the data and prepare a brief report
documenting any conclusions and recommendations that you have.

Using 7.2.1. Data Set One and 7.2.1. Data Set Two data sets, analyze the data. Assume
Greenshield's model for space mean speed as a function of density.

Tasks to be Completed

Task 1. Using regression to determine the theoretical equation that describes the speed as a function
of density:

Plot the actual data and theoretical data and describe any correlation.
Report all the important statistical results of the regression analysis.

Task 2. Using histograms, calculate and graph the following:

flow
speed
occupancy

Task 3. Using any analysis tools, plot:

flow versus occupancy
flow versus density
speed versus flow
speed versus density

126

Task 4. Write up a report stating the information specified above, as well as your own conclusions
about the data provided.

7.2.2. Lab Exercise Two: Shock Waves/Queue Formation

A freeway in Thrillville with two lanes in one direction has a capacity of 2000 vphpl under normal
stable flow conditions. On a particular morning, one of these lanes becomes blocked by a small
accident for 15 minutes, beginning at 7 a.m. The arrival pattern of vehicles is as follows:

Time/Flow (vph)
7-8 a.m./4000
8-9 a.m./3900
9-10 a.m./3500


After 10 a.m / 2800

This assignment is divided into two cases:

(a) the capacity of this section reduces to 1800 vphpl under unstable or forced flow conditions,

(b) the capacity of the section remains 2000 vphpl under forced flow conditions.

Please determine the following for each case above:

1. How long a queue will be established due to this blockage?
2. When will the maximum queue occur?
3. How long will it take to dissipate the queue from the time of the breakdown?

Tasks to be Completed

Task 1. Identify the type of problem and use the appropriate analytical tools to answer the questions.
State your analysis approach.

Task 2. Clearly lay out your solution and write up your solution in a brief report. Make sure that you
clearly label any and all graphs that you use as part of your solution.

7.3. Theory and Concepts

A course in transportation engineering wouldn't be complete without discussing some elements of
Traffic Flow Theory. Most junior level courses cover several aspects of Traffic Flow Theory,
including the topics listed below. To begin learning about Traffic Flow Theory, just click on the
link of your choice. Topics followed by the characters '[d]' include an Excel demonstration.

7.3.1. Types of Traffic Flow

Traffic flow can be divided into two primary types. Understanding what type of flow is occurring in
a given situation will help you decide which analysis methods and descriptions are the most
relevant.

127

The first type is called uninterrupted flow, and is flow regulated by vehicle-vehicle interactions and
interactions between vehicles and the roadway. For example, vehicles traveling on an interstate
highway are participating in uninterrupted flow.

The second type of traffic flow is called interrupted flow. Interrupted flow is flow regulated by an
external means, such as a traffic signal. Under interrupted flow conditions, vehicle-vehicle
interactions and vehicle-roadway interactions play a secondary role in defining the traffic flow.

7.3.2. Traffic Flow Parameters

Traffic flow is a difficult phenomenon to describe without the use of a common set of terms. The
following paragraphs will introduce most of the common terms that are used in discussions about
traffic flow.

Speed (v)

The speed of a vehicle is defined as the distance it travels per unit of time. Most of the time, each
vehicle on the roadway will have a speed that is somewhat different from those around it. In
quantifying the traffic flow, the average speed of the traffic is the significant variable. The average
speed, called the space mean speed, can be found by averaging the individual speeds of all of the
vehicles in the study area.

Volume

Volume is simply the number of vehicles that pass a given point on the roadway in a specified
period of time. By counting the number of vehicles that pass a point on the roadway during a 15-
minute period, you can arrive at the 15-minute volume. Volume is commonly converted directly to
flow (q), which is a more useful parameter.

Flow (q)

Flow is one of the most common traffic parameters. Flow is the rate at which vehicles pass a given
point on the roadway, and is normally given in terms of vehicles per hour. The 15-minute volume
can be converted to a flow by multiplying the volume by four. If our 15-minute volume were 100
cars, we would report the flow as 400 vehicles per hour. For that 15-minute interval of time, the
vehicles were crossing our designated point at a rate of 400 vehicles/hour.

Peak Hour Factor (PHF)

The ratio of the hourly flow rate (q
60
) divided by the peak 15 minute rate of flow expressed as an
hourly flow (q
15
). PBF =
q
60
q
1S


Density (k)

Density refers to the number of vehicles present on a given length of roadway. Normally, density is
reported in terms of vehicles per mile or vehicles per kilometer. High densities indicate that
individual vehicles are very close together, while low densities imply greater distances between
vehicles.

Headway, spacing, gap, and clearance are all various measures for describing the space between
vehicles. These parameters are discussed in the paragraphs below and are shown graphically in
figure 1.0.
128


Headway (h)

Headway is a measure of the temporal space between two vehicles. Specifically, the headway is the
time that elapses between the arrival of the leading vehicle and the following vehicle at the
designated test point. You can measure the headway between two vehicles by starting a
chronograph when the front bumper of the first vehicle crosses the selected point, and subsequently
recording the time that the second vehicles front bumper crosses over the designated point.
Headway is usually reported in units of seconds.

Spacing (s)

Spacing is the physical distance, usually reported in feet or meters, between the front bumper of the
leading vehicle and the front bumper of the following vehicle. Spacing complements headway, as it
describes the same space in another way. Spacing is the product of speed and headway.

Gap (g)

Gap is very similar to headway, except that it is a measure of the time that elapses between the
departure of the first vehicle and the arrival of the second at the designated test point. Gap is a
measure of the time between the rear bumper of the first vehicle and the front bumper of the second
vehicle, where headway focuses on front-to-front times. Gap is usually reported in units of seconds.

Clearance (c)

Clearance is similar to spacing, except that the clearance is the distance between the rear bumper of
the leading vehicle and the front bumper of the following vehicle. The clearance is equivalent to the
spacing minus the length of the leading vehicle. Clearance, like spacing, is usually reported in units
of feet or meters.

Figure 1.0: Explanation of Parameters


7.3.3. Speed-Flow-Density Relationship

Speed, flow, and density are all related to each other. The relationships between speed and density
are not difficult to observe in the real world, while the effects of speed and density on flow are not
quite as apparent.

Under uninterrupted flow conditions, speed, density, and flow are all related by the following
equation:

q = k v

Where
129


q = Flow (vehicles/hour)
v = Speed (miles/hour, kilometers/hour)
k = Density (vehicles/mile, vehicles/kilometer)

Because flow is the product of speed and density, the flow is equal to zero when one or both of
these terms is zero. It is also possible to deduce that the flow is maximized at some critical
combination of speed and density.

Two common traffic conditions illustrate these points. The first is the modern traffic jam, where
traffic densities are very high and speeds are very low. This combination produces a very low flow.
The second condition occurs when traffic densities are very low and drivers can obtain free flow
speed without any undue stress caused by other vehicles on the roadway. The extremely low density
compensates for the high speeds, and the resulting flow is very low.

7.3.4. Special Speed & Density Conditions

The discussion of the speed-flow-density relationship mentioned several speed-density conditions.
Two of these conditions are extremely significant and have been given special names.

Free Flow Speed

This is the mean speed that vehicles will travel on a roadway when the density of vehicles is low.
Under low-density conditions, drivers no longer worry about other vehicles. They subsequently
proceed at speeds that are controlled by the performance of their vehicles, the conditions of the
roadway, and the posted speed limit.

J am Density

Extremely high densities can bring traffic on a roadway to a complete stop. The density at which
traffic stops is called the jam density.

7.3.5. Greenshields Model [d]

Greenshield was able to develop a model of uninterrupted traffic flow that predicts and explains the
trends that are observed in real traffic flows. While Greenshields model is not perfect, it is fairly
accurate and relatively simple.

Greenshield made the assumption that, under uninterrupted flow conditions, speed and density are
linearly related. This relationship is expressed mathematically and graphically below. See figure
1.0.

v = A - B k

Where:

v = speed (miles/hour, kilometers/hour)
A, B = constants determined from field observations
k = density (vehicles/mile, vehicles/kilometer)

As noted above, you can determine the values of the constants A and B through field observations.
This is normally done by collecting velocity and density data in the field, plotting the data, and then
130

using linear regression to fit a line through the data points. The constant A represents the free flow
speed, while A/B represents the jam density.

Figure 1.0: Speed vs. Density


Inserting Greenshields speed-density relationship into the general speed-flow-density relationship
yields the following equations:

q = (A B k) k or q = A k B k
2

Where:

q = flow (vehicles/hour)
A, B = constants
k = density (vehicles/mile, vehicles/kilometer)

Figure 2.0: Flow vs. Density


This new relationship between flow and density provides an avenue for finding the density at which
the flow is maximized.

u
q
u
k
= A - 2 B k

setting
d
q
d
R
= u
yIcIds
----

k =
A
2 B


Therefore, at the density given above, the flow will be maximized. Substituting this maximized
value of k into the original speed-density relationship yields the speed at which the flow is
maximized.

v = A - B_
A
2 B
] oi v =
A
2


131

This indicates that the maximum flow occurs when traffic is flowing at half of free-flow speed (A).
Substituting the optimum speed and density into the speed-flow-density relationship yields the
maximum flow.

q =
A
2
_
A
2 B
] oi q =
A
2
4 B


Figure 3.0 shows the relationship between flow and speed graphically.

Figure 3.0: Flow vs. Speed



As you can see, Greenshields model is quite powerful. The following can be derived from
Greenshields model:

When the density is zero, the flow is zero because there are no vehicles on the roadway.
As the density increases, the flow also increases to some maximum flow conditions.
When the density reaches a maximum, generally called jam density, the flow must be zero
because the vehicles tend to line up end to end (parking lot conditions).

As the density increases the flow increases to some maximum value, but a continual increase in
density will cause the flow to decrease until jam density and zero flow conditions are reached.

7.3.6. Time-Space Diagrams [d]

A timespace diagram is commonly used to solve a number of transportation- related problems.
Typically, time is drawn on the horizontal axis and distance from a reference point on the vertical
axis. The trajectories of individual vehicles in motion are portrayed in this diagram by sloping lines,
and stationary vehicles are represented by horizontal lines. The slope of the line represents the
speed of the vehicle. Curved portions of the trajectories represent vehicles undergoing speed
changes such as deceleration.

Diagrams that show the position of individual vehicles in time and in space are very useful for
understanding traffic flow. These diagrams are especially useful for discussions of shock waves and
wave propagation.

The time-space diagram is a graph that describes the relationship between the location of vehicles in
a traffic stream and the time as the vehicles progress along the highway. The following diagram is
an example of a time-space diagram.
132




Time-space diagrams are created by plotting the position of each vehicle, given as a distance from a
reference point, against time. The first vehicle will probably start at the origin, while the vehicles
that follow wont reach the reference point until slightly later times. Reductions in speed cause the
slopes of the lines to flatten, while increases in speed cause the slopes to become greater.
Acceleration causes the time-space curve for the accelerating vehicle to bend until the new speed is
attained. Curves that cross indicate that the vehicles both shared the same position at the same time.
Unless passing is permitted, crossed curves indicate collisions.

7.3.7. Shock Waves [d]

Shock waves that occur in traffic flow are very similar to the waves produced by dropping stones in
water. A shock wave propagates along a line of vehicles in response to changing conditions at the
front of the line. Shock waves can be generated by collisions, sudden increases in speed caused by
entering free flow conditions, or by a number of other means. Basically, a shock wave exists
whenever the traffic conditions change.

The equation that is used to estimate the propagation velocity of shock waves is given below.

v
sw
=
q
b
- q
a
k
b
- k
a


Where

v
sw
= propagation velocity of shock wave (miles/hour)
q
b
= flow prior to change in conditions (vehicles/hour)
q
a
= flow after change in conditions (vehicles/hour)
k
b
= traffic density prior to change in conditions (vehicles/mile)
k
a
= traffic density after change in conditions (vehicles/mile)

Note the magnitude and direction of the shock wave.

(+) Shock wave is travelling in same direction as traffic stream.
(-) Shock wave is traveling upstream or against the traffic stream.

For example, lets assume that an accident has occurred and that the flow after the accident is
reduced to zero. Initially, the flow was several vehicles per hour. Also, the density is much greater
after the accident. Substituting these values into the shock wave equation yields a negative (-)
propagation velocity. This means that the shock wave is traveling against the traffic. If you could
133

look down on this accident, you would see a wave front, at which vehicles began to slow from their
initial speed, passing from vehicle to vehicle back up the traffic stream. The first car would notice
the accident first, followed an instant later by the second car. Each vehicle begins slowing after its
driver recognizes that the preceding vehicle is slowing.

7.3.8. Queuing Theory

Greenshields model was developed to aid our understanding of uninterrupted flow. Unfortunately,
Greenshields model is unable to cope with the added complexities that are generated under
interrupted flow conditions. Interrupted flow requires an understanding of Queuing Theory, which
is an entirely separate model of traffic flow.

Queuing Theory can be used to analyze the flow of traffic on the approach to and through an
intersection controlled by a traffic signal. This is accomplished by analyzing the cumulative passage
of vehicles as a function of time. The queuing diagram for interrupted flow shows the flow on one
intersection approach. Traffic is stopped from time t
1
to t
2
during the red signal interval. At the start
of the green interval (t
2
), traffic begins to leave the intersection at the saturation flow rate (q
G
), and
continues until the queue is exhausted. Thereafter, the departure rate D(t), equals the arrival rate,
A(t), until t
3,
which is the beginning of the next red signal. At this point, the process starts over.

For further information on Queing Theory, consult the chapter entitled "Signal Timing Design."



Queuing Diagram for Interrupted Flow
Papacostas , C.S. and Prevedouros, P.D., Transportation Engineering and Planning, 2 nd Edition,
Prentice Hall, Englewood Cliffs, New Jersey, 1993

7.4. Professional Practice

In order to supplement your knowledge about the various concepts within Traffic Flow Theory, and
in order to give you a glimpse of how these various topics are discussed in the professional
environment, we have included selected excerpts from professional design aids. The Transportation
Research Board maintains an Internet website to provide information on all facets of the
transportation industry. The professional practice material in this module is excerpted from TRBs
updated Special Report on Traffic Flow Theory, which is published on their website.

134

7.4.1. Traffic Flow Parameters

The following excerpt is taken from Chapter 2 (pp. 5-11) of the Transportation Research Board
Special Report on Traffic Flow Theory, published on the website http://www.tfhrc.gov/its/tft/tft.htm.

In general, traffic streams are not uniform, but vary over both space and time. Because of that,
measurement of the variables of interest for traffic flow theory is in fact the sampling of a random
variable. . . . In reality, the traffic characteristics that are labeled as flow, speed, and concentration
are parameters of statistical distributions, not absolute numbers.

Flow Rates

Flow rates are collected directly through point measurements, and by definition require
measurement over time. They cannot be estimated from a single snapshot of a length of road. Flow
rates and time headways are related to each other as follows. Flow rate, q, is the number of vehicles
counted, divided by the elapsed time, T:

q =
N
T


. . . Flow rates are usually expressed in terms of vehicles per hour, although the actual measurement
interval can be much less. Concern has been expressed, however, about the sustainability of high
volumes measured over very short intervals (such as 30 seconds or one minute) when investigating
high rates of flow. The 1985 Highway Capacity Manual (HCM 1985) suggests using at least 15-
minute intervals, although there are also situations in which the detail provided by five minute or
one minute data is valuable. . . .

Speeds

Measurement of the speed of an individual vehicle requires observation over both time and space. ...
In the literature, the distinction has frequently been made between different ways of calculating the
average speed of a set of vehicles. . . . The first way of calculating speeds, namely taking the
arithmetic mean of the observation,

u
t
=
1
N
u
I
N
I=1


is termed the time mean speed because it is an average of observations taken over time.

The second term that is used in the literature is space mean speed, but unfortunately there are a
variety of definitions for it, not all of which are equivalent. . . . Regardless of the particular
definition put forward for space mean speed, all authors agree that for computations involving mean
speeds to be theoretically correct, it is necessary to ensure that one has measured space mean speed,
rather than time mean speed. . . . Under conditions of stop-and-go traffic, as along a signalized
street or a badly congested freeway, it is important to distinguish between these two mean speeds.
For freely flowing freeway traffic, however, there will not be any significant difference between the
two. . . . When there is great variability of speeds, as for example at the time of breakdown from
uncongested to stop and go conditions, there will be considerable difference between the two.
Wardrop (1952) provided an example of this kind (albeit along what must certainly have been a
signalized roadway Western Avenue, Greenford, Middlesex, England), in which speeds ranged
from a low of 8 km/h to a high of 100 km/h. The space mean speed was 48.6 km/h; the time mean
135

speed 54.0 km/h. . . . For relatively uniform flow and speeds, the two mean speeds are likely to be
equivalent for practical purposes. Nevertheless, it is still appropriate to specify which type of
averaging has been done, and perhaps to specify the amount of variability in the speeds (which can
provide an indication of how similar the two are likely to be).

. . . At least for freeways, the practical significance of the difference between space mean speed and
time mean speed is minimal. However, it is important to note that for traffic flow theory purists, the
only correct way to measure average travel velocity is to calculate space-mean speed directly.

Only a few freeway traffic management systems acquire speed information directly, since to do so
requires pairs of presence detectors at each of the detector stations on the roadway, and that is more
expensive than using single loops. Those systems that do not measure speeds, because they have
only single-loop detector stations, sometimes calculate speeds from flow and occupancy data, using
a method first identified by Athol (1965). . . .

Concentration

Concentration has in the past been used as a synonym for density. For example, Gerlough and
Huber (1975, 10) wrote, "Although concentration (the number of vehicles per unit length) implies
measurement along a distance." In this chapter, it seems more useful to use concentration as a
broader term encompassing both density and occupancy. The first is a measure of concentration
over space; the second measure concentration over time of the same vehicle stream.

Density can be measured only along a length. If only point measurements are available, density
needs to be calculated, either from occupancy or from speed and flow. Gerlough and Huber wrote
(in the continuation of the quote in the previous paragraph), that " . . .traffic engineers have
traditionally estimated concentration from point measurements, using the relationship

k =
q
u
s


. . . The difficulty with using this equation to estimate density is that the equation is strictly correct
only under some very restricted conditions, or in the limit as both the space and time measurement
intervals approach zero. If neither of those situations holds, then use of the equation to calculate
density can give misleading results, which would not agree with empirical measurements. These
issues are important, because this equation has often been uncritically applied to situations that
exceed its validity . . . .

Real traffic flows, however, are not only made up of finite vehicles surrounded by real spaces, but
are inherently stochastic (Newell 1982). Measured values are averages taken from samples, and are
therefore themselves random variables. Measured flows are taken over an interval of time, at a
particular place. Measured densities are taken over space at a particular time. Only for stationary
processes (in the statistical sense) will the time and space intervals be able to represent conditions at
the same point in the time-space plane. Hence it is likely that any measurements that are taken of
flow and density (and space mean speed) will not be very good estimates of the expected values that
would be defined at the point of interest in the time space plane. . . .

Speeds within a lane are relatively constant during uncongested flow. Hence the estimation of
density from occupancy measurements is probably reasonable during those traffic conditions, but
not during congested conditions. . . . In short, once congestion sets in, there is probably no good
way to estimate density; it would have to be measured.

136

Temporal concentration (occupancy) can be measured only over a short section (shorter than the
minimum vehicle length), with presence detectors, and does not make sense over a long section.
Perhaps because the concept of density has been a part of traffic measurement since at least the
1930s, there has been a consensus that density was to be preferred over occupancy as the measure
of vehicular concentration. . . .

It would be fair to say that the majority opinion at present remains in favor of density, but that a
minority view is that occupancy should begin to enter theoretical work instead of density. There are
two principal reasons put forward by the minority for making more use of occupancy. The first is
that there should be improved correspondence between theoretical and practical work on freeways.
If freeway traffic management makes extensive use of a variable that freeway theory ignores, the
profession is the poorer. The second reason is that density, as vehicles per length of road ignores the
effects of vehicle length and traffic composition. Occupancy, on the other hand, is directly affected
by both of these variables, and therefore gives a more reliable indicator of the amount of a road
being used by vehicles. There are also good reasons put forward by the majority for the continued
use of density in theoretical work. Not least is that it is theoretically useful in their work in a way
that occupancy is not. . . .

7.4.2. Speed-Flow-Density Relationships

The following excerpt is taken from Chapter 2 (pp. 20-26) of the Transportation Research Board
Special Report on Traffic Flow Theory, published on the website http://www.tfhrc.gov/its/tft/tft.htm.

Speed-Density Model

This subsection deals with mathematical models for the

u = u
I
_1 -
k
k
j
_

speed-density relationship, going back to as early as 1935. Greenshields (1935) linear model of
speed and density was mentioned in the previous section. . . . The most interesting aspect of this
particular model is that its empirical basis consisted of half a dozen points in one cluster near free-
flow speed, and a single observation under congested conditions. . . . The linear relationship comes
from connecting the cluster with the single point. . . . What is surprising is not that such simple
analytical methods were used in 1935, but that their results (the linear speed-density model) have
continued to be so widely accepted for so long. While there have been studies that claimed to have
confirmed this model they tended to have similarly sparse portions of the full range of data,
usually omitting both the lowest flows and flow in the range near capacity. . . .

A second early model was that put forward by Greenberg (1959), showing a logarithmic
relationship:

u = c ln_
k
k
j
_

His paper showed the fit of the model to two data sets, both of which visually looked very
reasonable. However, the first data set was derived from speed and headway data on individual
vehicles, which "was then separated into speed classes and the average headway was calculated for
each speed class". In other words, the vehicles that appear in one data point(speed class) may not
even have been traveling together! While a density can always be calculated as the reciprocal of
137

average headway, when that average is taken over vehicles that may well not have been traveling
together, it is not clear what that density is meant to represent. . . .

Duncan (1976, 1979) showed that the tree step procedure of (1) calculating density from speed and
flow data, (2) fitting a speed-density function to that data, and then (3) transforming the speed-
density function into a speed-flow function results in a curve that does not fit the original speed-
flow data particularly well. . . . Duncans 1979 paper expanded on the difficulties to show that
minor changes in the speed-density function led to major changes in the speed-flow function. This
result suggests the need for further caution in using this method of double transformations to
calibrate a speed-flow curve. . . .

The car-following models gave rise to four of the speed-density models tested by Drake et al. The
results of their testing suggest that the speed-density models are not particularly good. Logic says
that if the consequences of a set of premises are shown to be false, then one (at least) of the
premises is not valid. It is possible, then, that the car-following models are not valid for freeways.
This is not surprising, as they were not developed for this context.

Flow-Concentration Model

Although Gerlough and Huber did not give the topic of flow-concentration models such extensive
treatment as they gave the speed-concentration models, they nonetheless thought this topic to be
very important. . . .

Edie was perhaps the first to point out that empirical flow-concentration data frequently have
discontinuities in the vicinity of what would be maximum flow, and to suggest that therefore
discontinuous curves might be needed for this relationship. . . .

Koshi et al. (1983) gave an empirically-based discussion of the flow-density relationship, in which
they suggested that a reverse lambda shape was the best description of the data. . . .

These authors also investigated the implications of this phenomenon for car-following models, as
well as for wave propagation.

. . . there appears to be strong evidence that traffic operations on a freeway can move from one
branch of the curve to the other without going all the way around the capacity point. This is an
aspect of traffic behavior that none of the mathematical models . . . either explain or lead one to
expect. Nonetheless, the phenomenon has been at least implicitly recognized since Lighthill and
Withams (1955) discussion of shock waves in traffic, which assumes instantaneous jumps from
one branch to the other on a speed-flow or flow-occupancy curve. As well, queuing models (e.g.
Newell 1982) imply that immediately upstream from the back end of a queue there must be points
where the speed is changing rapidly from the uncongested branch of the speed-flow curve to that of
the congested branch. It would be beneficial if flow-concentration (and speed-flow) models
explicitly took this possibility into account.

One of the conclusions of the paper by Hall et al. (1986), . . . is that an inverted V shape is a
plausible representation of the flow-occupancy relationship. Although that conclusion was based on
limited data from near Toronto, Hall and Gunter (1986) supported it with data from a larger number
of stations. Banks (1989) tested their proposition using data from the San Diego area, and
confirmed the suggestion of the inverted V. He also offered a mathematical statement of this
proposition and a behavioral interpretation of it (p. 58): The inverted-V model implies that drivers
maintain a roughly constant average time gap between their front bumper and back bumper of the
vehicle in front of them, provided their speed is less than some critical value. Once their speed
138

reaches this critical value (which is as fast as they want to go), they cease to be sensitive to vehicle
spacing. . . .

7.4.3. Greenshields' Model

The following excerpt is taken from Chapter 2 (pp. 17-20) of the Transportation Research Board
Special Report on Traffic Flow Theory, published on the website http://www.tfhrc.gov/its/tft/tft.htm.

Speed-Flow Model

The problem for traffic flow theory is that these curves are empirically derived. There is not really
any theory that would explain these particular shapes, except perhaps for Edie et al. (1980), who
propose qualitative flow regimes that relate well to these curves. The task that lies ahead for traffic
flow theorists is to develop a consistent set of equations that can replicate this reality. . . .

It is instructive to review the history of depictions of speed-flow curves in light of this current
understanding. Probably the seminal work on this topic was the paper by Greenshields in 1935, in
which he derived the following parabolic equation for the speed-flow curve on the basis of a linear
speed-density relationship together with the equation, flow = speed density:

q = k
j
_u -
u
2
u
I
_

where u
f
is the free-flow speed, and k
j
is the jam density. . . . In short, Greenshields model
dominated the field for over 50 years, despite at least three problems. The most fundamental is that
Greenshields did not work with freeway data. Yet his result for a single lane of traffic was adopted
directly for freeway conditions. (This of course was not his doing.) The second problem is that by
current standards of research the method of analysis of the data, with overlapping groups and
averaging prior to curve-fitting, would not be acceptable. The third problem is that despite the fact
that most people have used a model that was based on holiday traffic, current work focuses on
regular commuters who are familiar with the road, to better ascertain what a road is capable of
carrying. . . .

Speed-flow models are now recognized to be important for freeway management strategies, and
will be of fundamental importance for ITS implementation of alternate routing; hence there is
currently considerably more work on this topic than on the remaining two bivariate topics. . . .
Hence, it is sensible to turn to discussion of speed-concentration models, and to deal with any other
speed-flow models as a consequence of speed-concentration work, which is the way they were
developed.

7.4.4. Shock Waves and Continuum Flow Models

The following excerpt is taken from Chapter 5 (pp. 1-4) of the Transportation Research Board
Special Report on Traffic Flow Theory, published on the website http://www.tfhrc.gov/its/tft/tft.htm.

Since the conservation equation describes flow and density as a function of distance and time, one
can immediately see that continuum modeling is superior to input-output models used in practice
(which are only one dimensional, because they essentially ignore space). In addition, because flow
is assumed to be a function of density, continuum models have a second major advantage, (e.g.
compressibility). The simple continuum model referred to in this text consists of the conservation
equation and the equation of state (speed-density or flow density relationship). If these equations
are solved together with the basic traffic flow equation (flow equals density times speed), then we
139

can obtain speed, flow, density at any time and point of the roadway. Knowing these basic traffic
flow variables we know the state of the traffic system and can derive measures of effectiveness,
such as delays stops, total travel, travel time, and others that allow engineers to evaluate how well
the system is performing. . . .

A shock wave is a discontinuity of flow or density, and has the physical implication that cars
change speeds abruptly without time to accelerate or decelerate. This is an unnatural behavior that
could be eliminated by considering high order continuum models. These models add a momentum
equation that accounts for the acceleration and inertia characteristics of traffic mass. In this manner,
shock waves are smoothed out and the equilibrium assumption is removed. . . . In spite of this
improvement, the most widely known high order models still require an equilibrium speed-density
relationship. . . .

_u
I
- 2u
I

k
k
j
_
k
x
+
k
t
= u

. . . where u
f
represents the free flow speed and k
j
the jam density . . . is a first order quasi-linear,
partial differential equation which can be solved by the method of characteristics. . . . In practical
terms, the solution . . . suggests that:

The density k is a constant along a family of curves called characteristics or waves; a wave
represents the motion (propagation) of a change in flow and density along the roadway.
The characteristics are straight lines emanating from the boundaries of the time-space domain.
The slope of the characteristics is:

ux
ut
= f(k) + k|f(k)] =
uq
uk


This implies that the characteristics have slope equal to the tangent of the flow-density curve at
the point representing the flow conditions at the boundary from which the characteristic
emanates.
The density at any point x,t of the time space domain is found by drawing the proper
characteristic passing through that point.
The characteristics carry the value of density (and flow) at the boundary from which they
emanate.
When two characteristic lines intersect, then density at this point should have two values which
is physically unrealizable; this discrepancy is explained by the generation of shock waves. In
short, when two characteristics intersect, a shock wave is generated and the characteristics
terminate. A shock then represents a mathematical discontinuity (abrupt change) in k, q, or u.
The speed of the shock wave is:

u
w
=
q
d
-q
u
k
d
-k
u


. . . where k
d
, q
d
represent downstream and k
u
, q
u
upstream flow conditions. In the flow
concentration curve, the shock wave speed is represented by the slope of the line connecting the
two flow conditions (i.e., upstream and downstream).

It should be noted that when u
w
is positive, the shock wave moves downstream with respect to the
roadway; conversely, when u
w
is negative, the shock is moving upstream. Furthermore, the mere
fact that a difference exists in flow conditions upstream and downstream of a point does not imply
140

that a shock wave is present unless the characteristics intersect. Generally this occurs only when the
downstream density is higher than upstream. When density downstream is lower than upstream, we
have diffusion of flow similar to that observed when a queue is discharging. When downstream
density is higher than upstream, then shock waves are generated and queues are generally being
built even though they might be moving downstream.

Figure 5.2, taken from Gerlough and Huber (1975), demonstrates the use of traffic waves in
identifying the occurrence of a shock wave and following its trajectory. The process follows the
steps of the solution of the conservation equation as outlined above. The top of the figure represents
a glow-concentration curve; the bottom figure represents trajectories of the traffic waves. On the q-k
curve, point A represents a situation where traffic flows at near capacity implying that speed is well
below the free-flow speed. Point B represents an uncongested condition where traffic flows at a
higher speed because of the lower density. Tangents at points A and B represent the wave velocities
of these two situations. The areas where conditions A and B prevail are shown by the characteristics
drawn in the bottom of Figure 5.2. This figure assumes that the faster flow of point B occurs later in
time than that of point A; therefore, the characteristics (waves) of point B will eventually intersect
with those of point A. The intersection of these two sets of waves has a slope equal to the chord
connecting the two points on the q-k curve, and this intersection represents the path of the shock
wave shown at the bottom of Figure 5.2.

It is necessary to clarify that the waves of the time-space diagram of Figure 5.2 are not the
trajectories of vehicles but lines of constant flow and speed showing the propagation of conditions
A and B. The velocities of individual vehicles within A and B are higher because the speed of the
traffic stream is represented by the line connecting the origin with A and B in the q-k curve.

Figure 5.2 Shock Wave Formation Resulting from the Solution of the Conservation Equation





141

7.4.5. Queuing Theory

The following excerpt is taken from Chapter 5 (pg. 6 ) of the Transportation Research Board
Special Report on Traffic Flow Theory, published on the website http://www.tfhrc.gov/its/tft/tft.htm.

Consider a single-lane queue at the beginning of the effective green at a signalized intersection. If
the number of cars in the queue (i.e., the queue size) at this time is x and the average space headway
is h, then the estimated queue length (i.e., the space occupied by the x cars) is xh. Suppose now that
shortly after the beginning of green, N
1
cars join the queue while N are discharged in front. Then
following the same logic, the queue length should be [x + (N
1
-N
2
)]h. However, generally this is not
the case, since shortly after the commencement of green the queue length is growing regardless of
the net difference N
1
N
2
; for instance, if N
1
= N
2
the effective queue size continues to be x, but the
queue length can no longer be estimated from the product xh.

Clearly the average space headway is a function of time because of compressibility (i.e., the
changing density within the queue in both time and space). This observation leads to the conclusion
that although input-output analysis can be used for describing the evolution of queuing situation in
time, they yield crude estimation of another important state variable (i.e., the queue length). For
fixed-time control such approximations may suffice, but when further accuracy or realism is
required, more rigorous modeling is necessary. Another disadvantage of input-output analysis is
that the assumption of compact queues leads to miscalculations of the queue size itself and therefore
results in miscalculations of delays (Michalopoulos and Pisharody 1981). The simple continuum
model offers the advantage of taking compressibility into account

(pp. 9-11)

A major benefit of the continuum modeling is the fact that compressibility is built into the state
equations since speed or flow is assumed to be a function of density. This suggests that as groups of
cars enter areas of higher density, the continuum models exhibit platoon compression
characteristics; conversely, when they enter areas of lower density we observe diffusion or
dispersion. This phenomenon has been shown analytically in Michalopoulos and Pisharody (1980),
where it is demonstrated that by using continuum models we do not have to rely on empirical
dispersion models such as the ones employed today in most signal control packages. The result is a
more realistic and elegant modeling that should lead to more effective control.

The advantage of the analytical results presented thus far is that they visually depict the effects of
downstream disturbances on upstream flow. Thus they provide a good insight on the formation and
dissipation of queues and congestion in time and space in both freeways and arterials; further, they
can be used to demonstrate that platoon dispersion and compression are inherent in this modeling. .
. . The disadvantage of the analytical solution lies in the oversimplifications needed in the
derivations.

These include simple initial flow conditions, as well as arrival and departure patterns, absence of
sinks or sources, and uncomplicated flow-concentration relationships. Most importantly,
complexities frequently encountered in real situations such as turning lanes, side streets, or freeway
entrances and exits cannot be treated analytically with ease. As in similar problems of compressible
flow, these difficulties can be resolved by developing numerical solutions for the state equations.

Clearly, a numerical methodology is needed for numerical implementation of the conservation
equation in practical situations. This allows for inclusion of complexities one is likely to encounter
in practice (turning lanes, sinks and sources, spillbacks, etc.) treatment of realistic arrival and
departure patterns, more complicated u-k models, as well as inclusion of empirical considerations.
142

Numerical computation of k, u, and q proceeds by discretizing the roadway under consideration into
small increments x (in the order of 9 to 45 meters) and updating the values of these traffic flow
variables on each node of the discretized network at consecutive time increments t (in the order of
one second or so). . . .

. . . It should be emphasized that this discretization is not physical and is only performed for
computational purposes. . . . density on any node j except the boundary ones . . . at the next time
step n+1 is computed from density in the immediately adjacent cells (both upstream and
downstream j-1 and j+1 respectively) at the current time step n according to the relationship:

k
j
n+1
=
1
2
(k
j+1
n
+ k
j-1
n
) -
t
2x
(q
j+1
n
- q
j-1
n
) +
t
2
(q
j+1
n
- q
j-1
n
)

in which:

k
j
n
, q
j
n
= density and flow rate on node j at t = t
o
+nt
t
o
= the initial time
t, x = the time and space increments respectively such that t/x > free flow speed.
g
j
n
= is the generation (dissipation) rate at node j at t = t
o
+nt; if no sinks or sources exist
g
j
n
= 0 and the last term . . . vanishes.

Once the density is determined, the speed at t+t (i.e., at n+1) is obtained from the equilibrium
speed density relationship u
e
(k), i.e.,

u
j
n+1
= u
c
(k
j
n+1
)

For instance, for the Greenshields (1934) linear model,

u
j
n+1
= u
I
_1 -
k
j
n+1
k
jam
_

where u
f
is the free flow speed and k
jam
the jam density if an analytical expression is not
available, then u can easily be obtained numerically from the u-k curve. Finally, flow at t+t is
obtained from the fundamental relationship:

q
j
n+1
= k
j
n+1
u
j
n+1


. . . It can be demonstrated (Michalopoulos 1988) that measures of effectiveness such as delays,
stops, total travel, etc., can be derived from k, u, and q. . . .

In conclusion it is noted that more accurate numerical methods can be developed for solving the
conservation Equation 9; such methods are not recommended as they lead to sharp shocks which
are unrealizable in practice. . . .

7.5. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable concepts. The more concerned you
are about your understanding of a topic, the more seriously you will want to approach the example
problem for that topic.
143


7.5.1. Greenshield's Model

Inspection of a freeway data set reveals a free flow speed of 60 mph, a jam density of 180 vehicles
per mile per lane, and an observed maximum flow of 2000 vehicles per hour. Determine the linear
equation for velocity for these conditions, and determine the speed and density at maximum flow
conditions. How do the theoretical and observed conditions compare?

Solution

v
s
= v
I
-
v
I
k
j
k (mph)

v
s
= 6u -
6u
18u
k = 6u -u.SSSk

q = v
s
k

q = 6uk - u.SSSk
2


uq
uk
= 6u - 2 u.SSSk

60 = 2(0.333)k

k = 90 = k
j
/2 half of jam density

v
m
= 6u -
6u
18u
9u = Su mph =
v
I
2
= half of fiee flow speeu

q = v
s
k

q = Su 9u = 27uu vph > 2uuu:ph

The theoretical value does not account for the field conditions that influence maximum flow.

7.5.2. Shock Waves

A slow moving truck drives along the roadway at 10 MPH. The existing conditions on the roadway
before the truck enters are shown at point 1 below: 40 mph, flow of 1000 vehicles per hour, and
density of 25 vehicles per mile. The truck enters the roadway and causes a queue of vehicles to
build, giving the characteristics of point 2 below: flow of 1200 vehicles per hour and a density of
120 vehicles per mile. Using the information provided below, find the velocity of the shockwave at
the front and back of the platoon.

144



Point 1: Normal flow ( u
s
= 40 MPH, k=25 veh/mi, q= 1000 vph.)

Point 2: Slow Truck: ( u
s
= 10 MPH, k=120 veh/mi, q= 1200 vph.)

Solution

Figures 3.6.2 and 3.6.3, shown below, illustrate the behavior of the vehicles that are impacted by the
shockwave.

The speed of the shockwave in front of the truck at point A-A ( q
b
= 0, k
b
= 0) can be found by
substituting the correct values into the general shockwave equation. Upon substitution, as shown
below, we find that the shockwave is moving at the same speed as the truck, or 10 MPH
downstream with reference to a stationary point on the roadway.

u
sw
=
u - 12uu
u -12u
= +1u NPB

145



Solving for the speed of the shockwave at the end of the platoon (B-B) is accomplished by
substituting the correct values into the general shockwave equation.

q
a
= 1000 vph, k
a
=25 vpm
q
b
= 1200 vph, k
b
=120 vpm



The (+) sign indicates that the shockwave is moving downstream with respect to a fixed observer.

A-A moves forward relative to the roadway at 10 MPH
B-B moves forward relative to the roadway at 2.1 MPH

Platoon Growth: 10 - 2.1 = 7.9 MPH

Problem adapted from:
Papacostas, C.S., and Prevedourous P.D., Transportation Engineering and Planning, 2nd Edition,
Prentice Hall, pages 151-157

146

7.5.3. Traffic Flow Model

A study of freeway flow at a particular site has resulted in a calibrated speed-density relationship, as
follows:

U
s
= 57.5 (1 - 0.008k)

From this relationship:

a. Find the free-flow speed and jam density
b. Derive the equations describing flow versus speed and flow versus density.
c. Determine the capacity of the site mathematically

Solution

A) To solve for free-flow speed and jam density:

u
s
= 57.5 0.46k Notice that this equation is linear with respect to space mean speed and density
and is of the form of Greenshields equation.

Greenshields equation: u
s
= u
I
-
u
I
k
]
k

Free flow speed u
f
= 57.5 MPH

To calculate jam density:
u
I
k
]
= u.46 gives k
j
= 125 vpm

B) To derive the equations for flow as a function of density:

q= u
s
k

q = 57.5k - 0.46k
2
vph gives flow as a function of density (note that it is a quadratic in k)

To derive flow as a function of speed:

0.46k = 57.5 - u
s

k =
S7.S - u
s
u.46
= 12S -
u
s
u.46


q = u
s
[12S -
u
s
u.46
= 12Su
s
-
u
s
2
u.46
vph (note that it is a quauiatic in u
s
)

C) To determine the capacity of the site:

Need to determine the maximum flow:

uq
uk
= S7.S -u.46 2 k = u

S7.S = u.46 2 k

147

k =
S7.S
u.46 2
= 62.S veh pei mile = k
m
= uensity at maximum flow

q = S7.Sk - u.46k
2


q = S7.S 62.S -u.46 62.S
2


q = SS9S.7S - 1796.87S

q = 1796.87S vehhoui = q
m


speed at maxium flow = u
m
= 57.5 0.46 (62.5) = 28.75 mph

7.6. Glossary

Density: the number of vehicles occupying a road lane per unit length at a given instant.

Flow: the number of vehicles passing a point per unit of time; often called volume when the time
unit is one hour.

Gap: the time interval between the passage of consecutive vehicles moving in the same stream,
measured between the rear of the lead vehicle and the front of the following vehicle.

Headway: the time interval between passage of consecutive vehicles moving in the same stream,
measured between corresponding points (e.g. front bumper) on successive vehicles.

I nterrupted Flow: occurs when flow is periodically interrupted by external fixtures, primarily
traffic control devices.

J am Density: the density when speed and flow are zero.

PHF (Peak Hour Factor): This describes the relationship between hourly volume and the
maximum rate of flow within the hour: PHF = hourly volume/maximum rate of flow. For the 15
minute periods, PHF = volume / [4 x (maximum 15 minute volume within the hour)]

Shockwaves: Shockwaves occur as a result of differences in flow and density which occur when
there are constrictions in traffic flow. These constrictions are called bottlenecks. The speed of
growth of the ensuing queue is the shockwave, and is the difference in flow divided by the
difference in density.

Space Mean Speed: the arithmetic mean of the speed of those vehicles occupying a given length of
road at a given instant.

Spacing: the distance between vehicles moving in the same lane, measured between corresponding
points (front to front) of consecutive vehicles.

Speed: the time rate of change of distance.

Time Mean Speed: the arithmetic mean of the speed of vehicles passing a point during a given
time interval.

148

Travel Time: the total time required for a vehicle to travel from one point to another over a
specified route under prevailing conditions.

Uninterrupted Flow: occurs when vehicles traversing a length of roadway are not required to stop
by any cause external to the traffic stream, such as traffic control devices.

Volume: Traffic volume is the most basic and widely used parameter in traffic engineering,
vehicles per mile, or vehicles per kilometer.












































149

8. Travel Demand Forecasting

8.1. I ntroduction

Travel Demand Forecasting is a key component of the transportation engineers technical
repertoire. It allows the engineer to predict the volume of traffic that will use a given transportation
element in the future, whether that element is an existing highway or a potential light-rail route.

Like many other predictive sciences, Travel Demand Forecasting is continually evolving. Special
refinements based on experience and research are proposed each year, but the general ideology
behind Travel Demand Forecasting has remained relatively untouched.

The travel demand forecasting process can be confusing. This chapter is designed to introduce the
fundamentals of Travel Demand Forecasting to undergraduate engineering students by dividing the
process into manageable steps.

8.2. Lab Exercises

These exercises will help increase your understanding of Travel Demand Forecasting, by presenting
more complicated problems that require more thorough analysis.

8.2.1. Lab Exercise 1: The Gravity Model

The four-zone city of Wocsoms trip generation characteristics are shown below, in addition to a
travel network for the city. There are two major activities in this lab assignment:

a. Use the trip generation information provided to distribute the trips between the four zones.
b. Using the results of the trip distribution analysis and the results of the network analysis, assign
the trips to the various links.


The four-zone city has the following productions and attractions:


150

Zone Productions Attractions
A
B
C
D
1000
2000
3000
4000
3000
3000
2000
2000

Travel Time (min)
Zone A B C D
A
B
C
D
2
5
7
10
5
3
8
12
7
8
2
11
10
12
11
3

Travel Time (min) F
ij

2
3
5
7
8
11
12
3.0
2.5
2.3
1.5
1.2
0.95
0.90

Tasks to be Completed

Task 1. Distribute the trips for the city of Wocsom using the gravity model. Use the given data to
develop a trip table for the four-zone city of Wocsom.

Task 2. Find the shortest path from nodes A,B,C, and D to all other nodes and intersections.

Task 3. Using the trip table (veh/hr) below, load the network and find the total volume on each link
assuming all or nothing assignment.

From\ To A B C D
A
B
C
D
--
30
90
60
50
--
80
70
40
80
--
50
20
10
20
--

Task 4. Using the trip table resulting from the gravity model above, load the network and find the
total volume on each link, assuming all or nothing assignment.

Task 5. Prepare a brief report documenting your analysis, and be sure to explain the differences and
similarities in the results of Task 3 and Task 4.

Assumptions

For your analysis using the gravity model, assume the socioeconomic factor K
ij
=1.0.

151

8.2.2. Lab Exercise 2: Cross-Classification

Twenty households in the city of Scoretown were sampled for household income, autos per
household and trips produced.

Households Trips I ncome(dollars) Autos
1
2
3
4
5
2
4
10
5
5
4000
6000
17,000
11,000
4,500
0
0
2
0
1
6
7
8
9
10
15
7
4
6
13
17,000
9,500
9,000
7,000
19,000
3
1
0
1
3
11
12
13
14
15
8
9
9
11
10
18,000
21,000
7,000
11,000
11,000
1
1
2
2
2
16
17
18
19
20
11
12
8
8
9
13,000
15,000
11,000
13,000
15,000
2
2
1
1
1

Tasks to be Completed

Task 1. Develop matrices relating income to automobiles available.

Task 2. Draw a graph relating trips per household to income.

Task 3. Using the results of tasks 1 and 2, calculate how many trips a household with an income of
$10,000, owning one auto, will make per day?

Task 4. Perform a similar analysis for the same household, but with an additional vehicle.

Task 5. Prepare a brief report documenting your analysis of the data for Scoretown, and the impact
of increased auto ownership on travel demand patterns.

Task 6. Prepare a 3-5 page report discussing the important issues involved in managing and even
reducing travel demand, as measured by vehicle miles driven for the city of Scoretown or your own
community.

8.3. Theory and Concepts

Travel Demand Forecasting can seem like a long and daunting process when viewed as a whole. It
is much easier to approach when broken into small steps. The discussions below should help you
develop a basic understanding of the Travel Demand Forecasting process.

152

8.3.1. Overview of the TDF Process

Travel Demand Forecasting is a multi-stage process, and there are several different techniques that
can be used at each stage. Generally, Travel Demand Forecasting involves five interrelated tasks.

1. Break the area that requires prediction of future travel demand into study zones that can be
accurately described by a few variables.
2. Calculate the number of trips starting in each zone for a particular trip purpose. (Trip
Generation Analysis)
3. Produce a table of the number of trips starting in each zone and ending up in each other zone.
(Trip Distribution Analysis)
4. Complete the allocation of the various trips among the available transportation systems (bus,
train, pedestrian, and private vehicles). (Modal Choice Analysis)
5. Identify the specific routes on each transportation system that will be selected by the travelers.
(Trip Assignment Analysis)

Once these five steps have been completed, the transportation engineer will have a clear picture of
the projected travel demand for an existing or proposed transportation system.

8.3.2. Description of the Study Area

Study Boundary

Before forecasting the travel for an urban area or region, the planner must clearly define the exact
area to be considered. These areas may be defined by the urban growth boundary (UGB), county
lines or town centers. The planning area generally includes all the developed land, plus undeveloped
land that the area will encompass in the next 20 to 30 years.

The cordon line denotes the boundary of the planning area. In addition to considering future
growth, the establishment of the cordon line might take into account political jurisdictions, census
area boundaries, and natural boundaries. The cordon line should intersect a minimum number of
roads.

Zones

The study area must then be divided into analysis units, or zones. This will enable the planner to
link information about activities, travel, and transportation to the physical locations in the study
area. The transportation analysis zones (TAZ) vary in size depending on the density or nature of the
development. In an urban area the TAZ may be as small as a city block, but in rural areas the TAZ
may be as large as 10 or more square miles. The zones attempt to encompass homogeneous urban
activities, which are all residential, all commercial, or all industrial. Zones are designed to be
relatively homogeneous traffic generators and are sized so that only 10-15% of the trips are intra-
zonal.

An important consideration in establishing zones is their compatibility with the transportation
network. As a general rule, the network should form the boundaries of the zones.

A study area that has been divided into zones is shown below.

153



Links and Nodes

Normally, a simple representation of the geometry of the available transportation systems is
included on the map of the study area. A system of links and nodes, or a network, indicates
roadways and other transportation routes. Links represent sections of roadway (or railway etc.) that
are homogeneous, while nodes are simply points at which links meet. Usually, transit networks are
developed independently of truck and automobile networks. In the network description, zone
centroids (centers of activity) are identified; they are connected to nodes by imaginary links called
centroid connectors. Centroids are used as the points as which trips are "loaded" onto the network.
A diagram of a transportation network is shown below.

154


The figures on this page came from:
Garber, N.J. and Hoel, L.A., Traffic and Highway Engineering, Revised 2nd Edition, PWS, Pacific
Grove, CA. 1999. Pg. 499 and 501

155

8.3.3. Trip Generation Analysis

Once the study area has been broken into zones, the next task involves quantifying the number of
trips that each zone will produce or attract. The number of trips to and from an area or zone is
related to the land use activities of the zone and the socioeconomic characteristics of the trip-
makers.

There are at least three characteristics of land use and trip-makers that are important. The density or
intensity of the land use is important. Many studies begin by determining the number of dwellings,
employees, or tenants per acre. The intensity can be related to an average number of trips per day,
based on experience with the type of land use at hand. Next, the social and economic character of
the users can influence the number of trips that are expected. Character attributes like average
family income, education, and car ownership influence the number of trips that will be produced by
a zone. Finally, location plays an important role in trip production and attraction. Street congestion,
parking, and other environmental attributes can increase or decrease the number of trips that an area
produces or attracts.

The three major techniques used for Trip Generation Analysis are Cross-Classification, Multiple
Regression Analysis, and Experience Based Analysis. Each of these techniques is discussed as a
separate concept within this section.

Cross-Classification

The three major techniques used for Trip Generation Analysis are Cross-Classification, Multiple
Regression Analysis, and Experience Based Analysis. Cross-Classification procedures measure the
changes in one variable (trips) when other variables (land use etc.) are accounted for. Cross-
Classification resembles multiple regression techniques. Cross-Classification is essentially non-
parametric, since no account is taken of the distribution of the individual values. One problem with
the Cross-Classification technique is that the "independent" variables may not be truly independent,
and the resultant relationships and predictions may well be invalid.

The FHWA Trip Production Model uses Cross-Classification and has the following sub-models.

a. Income sub-model: reflects the distribution of households within various income categories (e.g.
high, medium and low).
b. Auto ownership sub-model: relates the household income to auto ownership.
c. Trip production sub-model: establishes the relationship between the trips made by each
household and the independent variables.
d. Trip purpose sub-model: relates the trip purposes to income in such a manner that the trip
productions can be divided among various purposes. These models are developed using origin-
destination travel surveys.

A considerable amount of research and development has focused on the area of disaggregate models
for improved travel demand forecasting. The difference between the aggregate and disaggregate
techniques is mainly in the data efficiency. Aggregate models are usually based upon home
interview origin and destination data that has been aggregated into zones; then the "average" zonal
productions and attractions are derived. The disaggregate approach is based on large samples of
household types and travel behaviors and uses data directly. There are savings in the amount of data
required and some of the data can be transferred to other applications. The disaggregate approach
expresses non-linear relationships and is more easily understood. The tables shown below show
several steps of a cross-classification analysis.

156


The above figures are from:
Paul Wright, Highway Engineering, 6
th
ed. Wiley, 1996.pp55, 56, and 58

8.3.4. Multiple Regression Analysis

The three major techniques used for Trip Generation Analysis are Cross-Classification, Multiple
Regression Analysis, and Experience Based Analysis. Multiple Regression Analysis is based on trip
generation as a function of one or more independent variables. The approach is mathematical and
all of the variables are considered random, and with normal distributions.

For example, consider the following equation:

T
i
= 0.34 (P) + 0.21 (DU) + 0.12 (A)

A
j
= 57.2 + 0.87 (E)

Where:

T
i
= Total number of trips produced in zone I
A
j
= Total number of trips attracted in zone j
P = Total Population for zone I
DU = Total number of dwelling units for zone I
A = Total number of automobiles in zone I
E = Total employment in zone j

157

Multiple Regression Analysis is relatively simple to understand. First, data regarding the actual
number of productions and attractions is coupled with data about the area that is thought to impact
the production and attraction of trips. For instance, the total population is believed to impact the
number of trips produced. If we know the number of trips produced and the population for the
present and a few time periods in the past, it is possible to develop a relationship between these
parameters using statistical regression. Once we are satisfied with the relationship that has been
developed, we can extrapolate into the future by plugging the future population into our relationship
and solving for the number of productions. The process is called Multiple Regression, because there
are normally several variables that impact trip production and attraction.

8.3.5. Experience Based Analysis

The three major techniques used for Trip Generation Analysis are Cross-Classification, Multiple
Regression Analysis, and Experience Based Analysis. Experience Based Analysis, one of the most
commonly used techniques, is founded primarily on experience. The Institute of Transportation
Engineers Manual of Trip Generation is one of the best sources of generalized trip generation rates.
The manual is a compilation of data from all over North America on many different types of land
uses. Within the manual, productions and attractions for each type of land use are related to some
measurable variable. For example, a shopping center might produce a certain number of trips for
each employee. Simply asking for the employment roster would allow a transportation engineer to
estimate the total number of trips that are generated by the shopping center employees. To establish
local credibility, a survey of similar land uses in the area may also need to be conducted.

8.3.6. Trip Distribution Analysis

Once the trip productions and attractions for each zone are computed, the trips can be distributed
among the zones using Trip Distribution Models. Trip Distribution has traditionally been based on
the gravity model, but other models are gaining popularity as well. This module will discuss the
logit model and the gravity model.

8.3.7. The Logit Model

The logit model, which will be discussed again later in the Mode Choice module, has been used by
the Portland, Oregon metropolitan area. The probability of selecting a particular destination zone is
based on the number of trip attractions estimated for that destination zone, relative to the total
attractions in all possible destination zones. The probability is applied to trip productions estimated
for the origin zone, making it conceptually similar to the gravity model.

P
Ij
=
e
V
]
e
V
]
z


where

P
ij
= probability of trips from zone i choosing destination j
V
ij
= A
j
- a t
Ij
+b t
Ij
2
where a and b are parameters to be estimated
A
j
= trip attractions estimated for zone j
t
ij
= highway travel time to zone j from zone i
Z = total number of zones

Multiplying the probability of traveling from zone i to zone j by the number of trips produced by
zone i will yield the number of trips produced by zone i that will travel to zone j.

158

8.3.8. The Gravity Model

The gravity model is much like Newton's theory of gravity. The gravity model assumes that the
trips produced at an origin and attracted to a destination are directly proportional to the total trip
productions at the origin and the total attractions at the destination. The calibrating term or "friction
factor" (F) represents the reluctance or impedance of persons to make trips of various duration or
distances. The general friction factor indicates that as travel times increase, travelers are
increasingly less likely to make trips of such lengths. Calibration of the gravity model involves
adjusting the friction factor.

The socioeconomic adjustment factor is an adjustment factor for individual trip interchanges. An
important consideration in developing the gravity model is "balancing" productions and attractions.
Balancing means that the total productions and attractions for a study area are equal.

Standard form of gravity model

T
Ij
=
A
j
F
Ij
K
Ij
A
x
F
Ij
K
Ix aIIzoncs
P
I


Where:

T
ij
= trips produced at i and attracted at j
P
i
= total trip production at i
A
j
= total trip attraction at j
F
ij
= a calibration term for interchange ij, (friction factor) or travel time factor (F
Ij
=
C
t
]
n
)
C = calibration factor for the friction factor
K
ij
= a socioeconomic adjustment factor for interchange ij
i = origin zone
n = number of zones

Before the gravity model can be used for prediction of future travel demand, it must be calibrated.
Calibration is accomplished by adjusting the various factors within the gravity model until the
model can duplicate a known base years trip distribution. For example, if you knew the trip
distribution for the current year, you would adjust the gravity model so that it resulted in the same
trip distribution as was measured for the current year.

8.3.9. Modal Choice Analysis

After completing the Trip Distribution Analysis, we need to determine what transportation system
each of those travelers will use. Mode choice models estimate how many people will use public
transit and how many will use private automobiles. The most common form of the mode choice
model is the logit model. The logit mode choice relationship states that the probability of choosing a
particular mode for a given trip is based on the relative values of a number of factors such as cost,
level of service, and travel time. The most difficult part of employing the logit mode choice model
is estimating the parameters for the variables in the utility function. The estimation is often
accomplished using one or more multivariate statistical analysis programs to optimize the accuracy
of estimates of the coefficients of several independent variables.

In regions where there are several alternative modes available, the mode choice model may require
a special form called the "nested" logit. This form attempts to represent the choices presented to the
159

traveler in a more structured manner. Nesting is necessary when there are major competing
alternatives within, as well as between, principal modes.

Logit Model

P
It
=
e
U
t
e
U
]t
aIIj


Where:

P
it
= probability of individual t choosing mode i
U
it
= utility of mode i to individual t
U
jt
= utility of mode j to individual t

For example:

U
auto
= 1.0 - 0.1 (TT
auto
) - 0.05 (TC
auto
)
U
bus
= - 0.1 (TT
bus
) - 0.05 (TC
bus
)
U
walk
= - 0.5 - 0.1 (TT
walk
)
TT = travel time by mode in minutes
TC = travel cost by mode in dollars

8.3.10. Trip Assignment Analysis

Once you have determined the number of trips that will enter and leave each zone, as well as the
transportation modes that the travelers will use, you can identify the exact roadways or routes that
will be selected for each trip. Trip assignment involves assigning traffic to a transportation network
such as roads and streets or a transit network.

Traffic is assigned to available transit or roadway routes using a mathematical algorithm that
determines the amount of traffic as a function of time, volume, capacity, or impedance factor. There
are three common methods for trip assignment: all or nothing, diversion, and capacity restraint.

All-or-Nothing

All-or-nothing is often referred to as the minimum path algorithm. The minimum path, or tree,
represents the minimum time path between two zone centroids and is assigned all of the traffic
volume between the zones in question. As volumes and travel times increase, the results of this
method become more unreliable.

As an example of this method, imagine that zones A and B are connected by ten separate routes.
Route 3.0 has the shortest travel time which means that, according to this model, all trips from A to
B will use route 3.0.

Diversion is the allocation of trips to two or more possible routes in a designated proportion that
depends on some specified criterion. In most cases the criterion that is used is time, although some
also use distance and generalized cost. Diversion is very similar to the all-or-nothing method,
except that portions of the total number of trips are allocated to different routes, with fewer trips
being given to those routes with longer travel times.



160

Capacity Restraint

Many different capacity restraint equations have been developed and tested and are available for
use. There are two basic characteristics common to capacity restraint models; (i) they are non-linear
relationships and (ii) they use the volume-capacity ratio or v/c as a common factor. The underlying
premise of a capacity restraint model is that the travel time on any link is related to the traffic
volume on that link. This is analogous to the level of service (LOS) criterion, where LOS A
corresponds to a low v/c and a higher vehicle speed. LOS E and the corresponding v/c = 1
represents capacity.

Capacity restraint models assign traffic to possible routes in an iterative manner:

1. A portion of the total traffic volume is assigned to the link with the shortest travel time.
2. Travel times for all possible links are calculated again, since volumes have changed.
3. Another portion of the traffic volume remaining to be assigned is allocated to the link that now
has the shortest travel time.
4. The travel time for all links are calculated and revised if changes result.
5. The process of incremental assignments, followed by calculation of revised shortest travel
times, by link, continues until all trips have been assigned.

The capacity restraint model used by FHWA is applied in an iterative manner. The adjusted link
speed and/or its associated travel impedance is computed using the following capacity restraint
function:

T = T
o
_1 +u.1S_
v
C
]
4
_

Where:

T = balance travel time (at which traffic V can travel on a highway segment)
T
o
= free flow travel time: observed travel time (at practical capacity) times 0.87
V = assigned volume
C = practical capacity

8.3.11. Results

Once you have completed the trip assignment analysis, you have a picture of the volume of traffic
that each element of your transportation system can expect to service in the future. This gives you
insight into the ramifications of changing the transportation system. For example, widening a
highway will increase capacity and shift more traffic onto that highway in the future. Using travel
demand forecasting, you can explore the impacts of alternatives before their construction.

8.4. Professional Practice

The Transportation Planning Handbook serves as the primary source for information concerning
Travel Demand Forecasting. Published by the Institute of Transportation Engineers, this manual
serves as a general reference for professional engineers. It has been developed extensively to
encompass all aspects of planning and provides essential knowledge for the transportation engineer.

We referenced the 1992 publication of the Transportation Planning Handbook because it is
generally accepted as "the" authority on planning in the professional realm of engineering. For more
a extensive analysis the Institute of Transportation Engineers also publishes a Travel Demand
161

Forecasting manual. This manual provides in-depth analyses for various aspects of travel demand
and impact studies and is also frequently referenced in the professional arena.

8.4.1. Zones and Zoning

The following excerpt was taken from the Transportation Planning Handbook published in 1992 by
the Institute of Transportation Engineers (pp. 100-102).

Data processing of information describing the urban area and the transportation system requires
identifying that information with a numerical code to facilitate automated retrieval. To do this the
area being studied is divided into small geographic areas called zones, and the boundaries of each
zone are drawn on a base map of convenient scale. A unique numerical code, usually consecutive
starting with number one, is assigned to each zone. . . . The time, cost, and capacity for computer
processing dictate that there should usually not be more than 1,000 analysis zones. . . . In large
metropolitan areas, the recommended limitation on the number of zones may yield zones that are
too large for detailed transportation analysis. The approach that has been chosen by some agencies
to overcome this difficulty is to define zones that are small enough to perform the most detailed
analysis anticipated. These small zones are aggregated to larger zones of an appropriate size for
analyses requiring less detail.

8.4.2. Networks and Nodes

The following excerpt was taken from the Transportation Planning Handbook published in 1992 by
the Institute of Transportation Engineers (pp. 102-103).

The transportation system is represented by a network of lines; each road and transit route is drawn
as a line on a map or overlay at the same scale as the zone map. The intersections of the
transportation lines are called nodes. Each node is assigned a unique number, starting with a
number somewhat greater than the highest zone number. . . . The two node numbers at the ends of
any link identify that link. A roadway is defined by the links along its path. The links are stored in
the computer according to their identifying node numbers. Transit routes are identified and stored in
a similar manner, as the string or series of node numbers along the transit route. For transit routes
operating on roadways, the string of node numbers identifies the roadway nodes traversed by the
transit route.

(p. 103) The characteristics of the roadway represented by each link are coded as attributes of that
link. The attributes usually include the length of the link (in miles), the vehicle capacity of the
roadway, and the speed or time of movement along the link. Depending on the computer program
being used, other attributes of the link or activity along the link may also be coded, such as adjacent
land use, character of the area in which the link is located, whether parking is permitted, and the
classification of the facility that the link represents.

8.4.3. Trip Generation Analysis

The following excerpt was taken from the Transportation Planning Handbook published in 1992 by
the Institute of Transportation Engineers (pp. 108-112).

Trip Generation Models

(p. 110) There are two kinds of trip generation models: production models and attraction models.
Trip production models estimate the number of home-based trips to and from zones where trip
makers reside. Trip attraction models estimate the number of home-based trips to and from each
162

zone at the non-home end of the trip. Different production and attraction models are used for each
trip purpose. Special generation models are used to estimate nonhome-based, truck, taxi, and
external trips.

Cross-Classification

Over time the profession has come to understand that considerable predictive power and accuracy
can be gained by disaggregate analysis of influential variables. . . . This means that the models use
factors describing individual sample units (e.g., persons, households or workplaces) rather than an
average value of each factor for each analysis zone. The result is trip generation models with trip
rates for sample units having specific characteristics, such as households of one, two, or more
family members, owning one, two, or more vehicles. These models are based on the trip rates for
individual sample households having those particular discrete characteristics. . . .

(p. 112) Most trip production models are two- or three-way cross-classification tables with the
dependent variable being trips per household or trips per person. The independent variables are
most often income, auto ownership, and household size. . . . Virtually all of the trip attraction
models use employment and an identifier of location as independent variables.

Multiple Regression

(p. 110) Early trip generation models were commonly developed by regression analysis because of
its power and simplicity. The independent variables in such models were usually zonal averages of
the various factors of influence. Trip generation equations developed by regression are still used by
some planning agencies, more commonly for attraction models than for production models. This is
because only zonal averages of trip attracting characteristics are usually available since most travel
surveys do not survey at trip destinations. Obtaining more detailed data for individual attraction
zones requires a survey of trip attractors, such as a workplace survey.

Experience Based

(p. 108) Early travel forecasting used extrapolation of past trends to estimate future travel. Such an
approach is still used occasionally for estimating future traffic on a single facility, in a relatively
isolated area, where only moderate and uniform growth or change in development pattern is
anticipated. One level of sophistication that can be added to trend analysis to respond to anticipated
growth is comparing the past traffic trend to the trend of development during the same period. This
provides understanding of how traffic on the subject facility will respond to expected development
changes. That relationship between the two trends is incorporated subjectively in the trend forecast.

8.4.4. Trip Distribution

The following excerpt was taken from the Transportation Planning Handbook published in 1992 by
the Institute of Transportation Engineers (pp. 112-114).

Trip distribution models connect the trip origins and destination estimated by the trip generation
models to create estimated trips. Different trip distribution models are developed for each of the trip
purposes for which trip generation has been estimated. The trip distribution models found most
often in practice today are "gravity models," so named because of their basis in Newtons law. . . .

The measure of separation between zones most commonly used for trip distribution is roadway
travel time, calculated from the computerized transportation networks. Most transportation planning
efforts use peak-period travel times as a measure of zonal separation for home-based work and
163

home-based school models. . . . Recent studies have tried to incorporate travel cost and transit travel
time into the separation measure. Cost has been considered in an attempt to estimate effects on trip
distribution of parking costs, vehicle operating costs, and tolls.

Logit Model

Other trip distribution models that have been used include "opportunity" models and logit models,
both of which estimate the probability that travelers will accept various destination options
available. The logit formulation has recently been used for the Portland, Oregon metropolitan area.
As shown in Figure 4.20, the probability of selecting a particular destination zone is based on the
number of trip attractions estimated for that destination zone relative to the total attractions in all
possible destination zones. The probability is applied to trip productions estimated for the origin
zone, making it conceptually similar to the gravity model.

Gravity model

Those models generally estimate the distribution of trips to be proportional to the number of trip
ends estimated by the trip generation models and inversely proportional to a measure of separation
between the origin and destination zones. The gravity model has achieved virtually universal use
because of its simplicity, its accuracy and due to its support from the U.S. Department of
Transportation. . . .

Developing a gravity model is a trial-and-error process that requires considerable care. This
process, often called calibration, identifies the appropriate decay function or "friction factor", that
represents the reluctance or impedance of persons to make trips of various durations or distances. . .
. The adjustments are made incrementally with successive iterations of the model until the trip
length frequency distribution produced by the model closely matches the frequency distribution
from the travel survey or demonstrates an acceptable shape and average trip length.

An important consideration in developing trip distribution models is "balancing" productions and
attractions. One aspect of balance is to assure that the total productions equal the total attractions in
the study area for each trip purpose. Deciding whether the productions or attractions should be the
control total depends on whether there is greater confidence in the production (usually population)
growth estimate or the attraction (usually employment) growth estimate. It is not unreasonable to
average the two (production and attraction) trip estimates. The productions and/or attractions for all
zones must then be factored so that their sum matches the control total. . . .

(p. 114) At each iteration of the gravity model, the total trips attracted to each zone is adjusted so
that the next iteration of the gravity model will send more or fewer trips to that attraction zone,
depending on whether the immediately previous total trips attracted to that zone was lower or
higher, respectively, than the trip attractions estimated by the trip generation model. . . . Any
unacceptable difference between the generation and distribution model estimates after five
iterations of the gravity model usually indicates an inconsistency in the assumptions or functions of
the trip distribution model and the growth allocation model.

One other consideration in developing a trip distribution model is how to handle unexplained and
unacceptable differences between observed and estimated travel patterns. Rather than conduct
extensive research to try to find an explanation for all such phenomena, the accepted practical
approach is to factor the model estimates to match observed patterns. . . . With the gravity model,
and often with other models in this situation, the adjustment factors are called "K" factors. The "K"
factors are developed for individual trip interchanges and are assigned values that adjust the
estimated trips for the interchanges of concern to match the observed values.
164


8.4.5. Modal Choice

The following excerpt was taken from the Transportation Planning Handbook published in 1992 by
the Institute of Transportation Engineers (pp. 114-115).

Mode choice models are usually the most complex of the sequential model structure. Typically
these models estimate how many persons will ride public transit and how many will use private
vehicles. Further sophistication of these models may include identifying submode choice among
different transit services and estimating the number of car pools or van pools of various sizes for
high occupancy vehicle facilities. . . .

Logit Model

Mode choice models are found in numerous formulations, but the most common are based on the
probabilities estimated by some variation or sophistication of the logit function. . . . The common
logit mode choice relationship states that the probability of choosing a particular mode for a given
trip is based on the relative values of the costs and levels of service on the competing modes for the
trip interchange being considered.

The level of service provided by a particular mode for a specific trip interchange is usually
represented in part by the travel time for that interchange as computed from the transit and roadway
networks. The travel time components used to represent level of service include the in-vehicle
travel time for each mode and the out-of-vehicle time required to use that mode, such as walking to
a transit stop or from a parking lot. The level of service also includes the waiting time likely to be
experienced, either to board transit or to transfer. The delay due to roadway traffic congestion is
included inherently by using attenuated speeds for congested roadway network links. . . .

The travel time and cost of a trip are usually combined using an estimate of the cost of time to
convert either cost or time to the terms of the other. The cost of time is usually a variable, based on
the economic level of the traveler. Although the mode choice model may be developed using the
economic level of individual travelers, forecasts of mode choice are prepared for different economic
groups, such as high, medium, and low income travelers. The resulting combination of time and
cost is commonly referred to as the "utility" or "generalized cost.". . . .

The logit formulation is not a complex mathematical function nor is the utility function it employs.
The difficulty in developing a logit model is encountered in estimating the considerable number of
parameters for variables in the utility function. The estimation is accomplished using one or another
multivariate statistical analysis program to optimize the accuracy of estimates of coefficients of
several independent variables.

8.4.6. Trip Assignment

The following excerpt was taken from the Transportation Planning Handbook published in 1992 by
the Institute of Transportation Engineers (pp. 115-117).

The traffic assignment process is somewhat different from the mathematical models used for trip
distribution and mode choice. Traffic is assigned to available transit or roadway routes using a
mathematical algorithm which determines the amount of traffic to allocate to each route. The traffic
allocation is usually based on the relative time to travel along each available path, computed from
the transit and roadway networks.

165

All or nothing

Historically all trips between two zones were assigned to the route having the minimum travel time,
regardless of the available capacity; this is termed an "all-or-nothing" assignment. Such an approach
is still used for identifying travel desire corridors as an initial step in locating new and improved
transportation facilities. For most transit assignments the all-or-nothing approach is still used since
there are rarely closely competing transit routes in an efficiently designed transit system. Similarly
the all-or-nothing approach is used for assigning high occupancy vehicle trip assignments.

Capacity Restraint

More common today for roadway assignments is the "capacity-restrained" assignment, a strategy
which assigns traffic in steps. One option in this approach is "proportional" assignment, which
allocates a portion of the trips between every origin-destination zone pair to the network at each
step. An alternative is the "incremental" assignment, which allocates all of the trips between a
subset of zone pairs at each step. In either case the travel times between all zone pairs are
recalculated after each assignment step, considering the traffic already assigned, to adjust the speeds
on all network links. The revised speeds on all links are determined by a speed-volume function that
indicates the maximum speed likely for a particular volume/capacity ratio. . . . Another assignment
step is then computed considering the revised travel times, after which the link speeds are again
adjusted as previously. This process is iterated until all trips have been assigned. Additional fully
iterated assignments may be necessary to reach an equilibrium in which there is little change in
speeds throughout the network at each assignment step.

8.4.7. Model Calibration and Validation

The following excerpt was taken from the Transportation Planning Handbook published in 1992 by
the Institute of Transportation Engineers (p. 116).

(p.116) The process of developing travel models is commonly called "calibration." Given the basic
form of a travel forecasting model, such as a gravity model or a logit model, calibration involves
estimating the values of various constants and parameters in the model structure. For this reason the
model development effort is sometimes termed "estimation."

Estimating model coefficients and constants is usually done by solving the model equation for the
parameters of interest after supplying observed values of both the dependent and independent
variables. The observed values of variables are obtained from the surveys of actual travel patterns.
As indicated previously, the estimation process is a trial and error effort that seeks the parameter
values which have the greatest probability or maximum likelihood of being accurate within
acceptable tolerance of error.

Such an effort is commonly accomplished with specialized statistical computer programs designed
for just such purposes. . . . Model calibration can also be accomplished by using values of constants
and parameters from models estimated for another location that is similar to the area being studied;
this strategy is referred to as "importing" model parameters and should be employed only by
experienced practitioners.

Once satisfactory estimates of the parameters for all models have been obtained, the models must be
checked to assure that they adequately perform the functions for which they are intended, that is, to
accurately estimate traffic volumes on transit and roadways. Verifying a calibrated model in this
manner is commonly called "validation." The validation process establishes the credibility of the
model by demonstrating its ability to replicate actual traffic patterns.
166


Validating the models requires comparing traffic estimated by the model to observed traffic on the
roadway and transit systems. Initial comparisons are for trip interchanges between quadrants,
sectors, or other large areas of interest. . . . The next step is to compare traffic estimated by the
models to traffic counts, including transit ridership, crossing contrived barriers in the study area.
These are commonly called screenlines, cutlines, and cordon lines and may be imaginary or actual
physical barriers. Cordon lines surround particular areas such as the central business district or other
major activity centers. . . . Transit ridership estimates are commonly validated by comparing them
to actual patronage crossing cordon lines around the central business district. . . .

The importance of traffic and transit counts for model validation underscores the need for careful
planning, thoroughness and accuracy of a traffic and transit data collection program that has this
purpose. As with the travel surveys, the resulting models and forecasts will be no better than the
data used for model estimation and validation.

8.5. Example Problems

It doesn't seem to matter how many times we read about a concept, most of us won't remember it or
fully understand it until we have worked with it. To encourage this extra level of comprehension,
we have provided an example problem for each of the applicable concepts. The more concerned you
are about your understanding of a topic, the more seriously you will want to approach the example
problem for that topic.

8.5.1. Cross Classification

The following cross-classification data have been developed for Beaver Dam Transportation Study
Area.

($000) HH (%) Autos/HH (%)
I ncome High Med Low 0 1 2 3
10 0 30 70 48 48 4 0
20 0 50 50 4 72 24 0
30 10 70 20 2 53 40 5
40 20 75 5 1 32 52 15
50 50 50 0 0 19 56 25
60 70 30 0 0 10 60 30

($000) Trip Rate/Auto Trips (%)
I ncome 0 1 2 3+ HBW HBO NHB.
10 2.0 6.0 11.5 17.0 38 34 28
20 2.5 7.5 12.5 17.5 38 34 28
30 4.0 9.0 14.0 19.0 35 34 31
40 5.5 10.5 15.5 20.5 27 35 38
50 7.5 12.0 17.0 22.0 20 37 43
60 8.0 13.0 18.0 23.0 16 40 44

Develop the family of cross classification curves and determine the number of trips produced (by
purpose) for a traffic zone containing 500 houses with an average household income of $35,000.
(Use high = 55,000; medium = 25,000; low = 15,000)
167


Solution

The solution to this type of problem is best described through the use of graphs and tables. The
graphs and tables used for this problem are shown below.






168






169



I ncome Households (%) HH/Zone Total HH
Low $under $20,000 13 500 65
Medium $20,000 - 45,000 72 500 360
High $45,000 - $60,000 15 500 75
100 500

Percentage of HH owning #vehicles
Auto Ownership
I ncome
Low Medium High
0 26 3 0
1 60 63 15
2 14 32 58
3+ 0 2 27
100 100 100

Trips per HH per I ncome Level and Auto Ownership
Auto Ownership
I ncome
Low Medium High
0 2 3 7
1 7 8 13
2 12 13 18
3+ 17 18 23


170

Number of HH owning #vehicles
Auto Ownership
I ncome
Low Medium High
0 17 11 0
1 39 227 11
2 9 115 44
3+ 0 7 20
65 360 75

Trips made by income level
Auto Ownership
I ncome
Low Medium High

0 34 32 0
1 273 1814 146
2 109 1498 783
3+ 0 130 466
416 3474 1395 5285

Trips by Trip Purpose %
I ncome
Low Medium High
HBW 38 37 18
HBO 34 34 38
NHB 28 29 44
100 100 100

Number of Trips by Purpose
I ncome
Low Medium High
HBW 158 1285 251 1695
HBO 141 1181 530 1853
NHB 116 1007 614 1738
Problem adapted from:
Garber, N.J. and Hoel, L.A., Traffic and Highway Engineering, Revised 2nd Edition, PWS, Pacific
Grove, CA. 1999. Page 545

8.5.2. Gravity Model

A study area consists of three zones. The data have been determined as shown in the following
tables. Assume a K
ij
=1.

Zone Productions and Attractions
Zone 1 2 3 Total
Trip Productions 140 330 280 750
Trip Attractions 300 270 180 750

171

Travel Time between zones (min)
Zone 1 2 3
1 5 2 3
2 2 6 6
3 3 6 5

Travel Time versus Friction Factor
Time (min) F
1 82
2 52
3 50
4 41
5 39
6 26
7 20
8 12

Determine the number of trips between each zone using the gravity model formula and the data
given above. Note that while the Friction Factors are given in this problem, they will normally need
to be derived by the calibration process described in the Theory and Concepts section.

Solution

First, determine the friction factor for each origin-destination pair by using the travel times and
friction factors given in the problem statement.

F
ij
as Determined from Travel Time
Zone 1 2 3
1 39 52 50
2 52 26 26
3 50 26 39

Once you have the friction factors for each potential trip, you can begin solving the gravity model
equation as shown below. Solving for the AFK term in a tabular form makes this process easier.
Study the equation below and the following table.

T
Ij
= P
I

A
j
F
Ij
K
Ij
A
I
F
Ij
K
Ij I


Where:

T
ij
= number of trips that are produced in zone i and attracted to zone j
P
i
= total number of trips produced in zone i
A
j
= number of trips attracted to zone j
F
ij
= a value which is an inverse function of travel time
K
ij
= socio economic adjustment factor for interchange ij



172

A
j
F
ij
K
ij
1 2 3 sum
1 11700 14040 9000 34740
2 15600 7020 4680 27300
3 15000 7020 7020 29040

Once the AFK terms for each origin-destination are tabulated, you can insert these values into
the gravity model equation and determine the number of trips for each origin-destination. The
following table illustrates this.

Zone to Zone First Iteration:
zone 1 2 3 P
1 47 57 36 140
2 189 85 57 330
3 145 68 68 280
A 380 209 161 750
given A 300 270 180 750

Since the total trip attractions for each zone dont match the attractions that were given in the
problem statement, we need to adjust the attraction factors. Calculate the adjusted attraction factors
according to the following formula:

A
jk
=
A
j
C
j(k1)
A
j(k1)


Where:

A
jk
= adjusted attraction factor for attraction zone (column) j iteration k.
A
jk
= A
j
when k=1
C
jk
= actual attraction (column) total for zone j, iteration k
A
j
= desired attraction total to attraction zone (column) j
j = attraction zone number
n = number of zones
k = iteration number

To produce a mathematically correct result, repeat the trip distribution computation using the
modified attraction values.

For example, for zone 1:

A
12
=
Suu
S8u
Suu = 2S7

Zone 1 2 3
A
j1
380 209 161
Given A 300 270 180
A
j2
237 349 201



173

A
j
F
ij
K
ij
1 2 3 sum
1 9237 18138 10062 37437
2 12316 9069 5232 26617
3 11842 9069 7848 28759

Zone to Zone Second Iteration:
zone 1 2 3 P
1 35 68 38 140
2 153 112 65 330
3 115 88 76 280
A 303 269 179 750
given A 300 270 180 750

Upon finishing the second iteration, the calculated attractions are within 5% of the given attractions.
This is an acceptable result and the final summary of the trip distribution is shown below.

The resulting trip table is:
zone 1 2 3
1 35 68 38
2 153 112 65
3 115 88 76

8.5.3. Logit Model

Given the utility expression:

U
K
= A
K
- 0.05 T
a
- 0.04T
w
- 0.02 T
r
- 0.01 C

Where:

T
a
is the access time
T
W
is the waiting time
T
r
is the riding time
C is the out of pocket cost

a) Apply the logit model to calculate the division of usage between the automobile mode
(A
K
= - 0.005) and a mass transit mode (A
K
= - 0.05). Use the data given in the table below for
your analysis.

Mode T
a
T
W
T
r
C
Auto 5 0 30 100
Transit 10 10 45 50

b) Estimate the patronage shift that would result from doubling the bus out-of-pocket cost.

Solution

Part A is solved by substituting the given values into the utility function and solving the logit
model equation. The calculations and results for part A are shown in the table below.
174


Part B is essentially identical to part A except for the change in the out-of-pocket cost for bus
travel. The preliminary calculations for part B are shown in the table below as well, while the final
calculations are located below the table.

Part A
Mode T
a
T
W
T
r
C A
k
U
k
e
U
R
P
Auto 5 0 30 100 - 0.0050 - 1.855 0.1565 0.621
Transit 10 10 45 50 - 0.0500 - 2.350 0.0954 0.379
0.2518 1.000
Part B
Mode T
a
T
W
T
r
C A
k
U
k
e
U
R
P
Auto 5 0 30 100 - 0.0050 - 1.855 0.1565 0.730
Transit 10 10 45 100 - 0.0500 - 2.850 0.0578 0.270
0.2143 1.000

A significant number of bus riders are predicted to shift to the automobile.

S8 - 27
S8
1uu = 29%

The increase in automobile use will be:

7S - 62
62
1uu = 17.7%

8.5.4. Traffic Assignment

Assign the vehicle trips shown in the following O-D trip table to the network, using the all-or-
nothing assignment technique. To summarize your results, list all of the links in the network and
their corresponding traffic volume after loading.

Origin-Destination Trip Table:
Trips between Zones
From/to 1 2 3 4 5
1 - 100 100 200 150
2 400 - 200 100 500
3 200 100 - 100 150
4 250 150 300 - 400
5 200 100 50 350 -







175

Highway Network:


Solution

The all-or-nothing technique simply assumes that all of the traffic between a particular origin and
destination will take the shortest path (with respect to time). For example, all of the 200 vehicles
that travel between nodes 1 and 4 will travel via nodes 1-5-4. The tables shown below indicate the
routes that were selected for loading as well as the total traffic volume for each link in the system
after all of the links were loaded.

Nodes From To Link Path Travel Time Volume
1
2 1-2 8 100
3 1-2, 2-3 11 100
4 1-5, 5-4 11 200
5 1-5 5 150
2
1 2-1 8 400
3 2-3 3 200
4 2-4 5 100
5 2-4, 4-5 11 500
3
1 3-2, 2-1 11 200
2 3-2 3 100
4 3-4 7 100
5 3-4, 4-5 13 150
4
1 4-5, 5-1 11 250
2 4-2 5 150
3 4-3 7 300
5 4-5 6 400
5
1 5-1 5 200
2 5-4, 4-2 11 100
3 5-4, 4-3 13 50
4 5-4 6 350









176

Link Volume
1-2 200
2-1 600
1-5 350
5-1 450
2-5 0
5-2 0
2-3 300
3-2 300
2-4 600
4-2 250
3-4 250
4-3 350
4-5 1300
5-4 700

8.6. Glossary

Centroids: Imaginary points within zones from which all departing trips are assumed to originate
and at which all arriving trips are assumed to terminate.

Cordon Line: An imaginary line that denotes the boundary of the study area.

Friction Factor: A mathematical factor that is used to describe the effort that is required to travel
between two points.

Link: An element of a transportation network that connects two nodes. A section of roadway or a
bus route could be modeled as a link.

Modal Choice Analysis: The process used to estimate the number of travelers who will use each of
the available transportation modes (train, car, bus) to reach their destination.

Nodes: Nodes are points at which links terminate. Links may terminate at destinations or at
intersections with other links.

Routes: Pathways through a network. Routes are composed of links and nodes.

Study Area: The region within which estimates of travel demand are desired.

Trip: The journey between one point and another.

Trip Assignment Analysis: The process used to estimate the routes (for each mode) that will be
used to travel from origin to destination. This process yields the total number of vehicles or
passengers that a particular route can expect to service.

Trip Distribution Analysis: The process used to determine the number of produced trips from
each zone that will be attracted by each of the remaining zones.

Trip Generation Analysis: A data collection and analysis process that is used to estimate the
number of trips that each zone will produce and attract.
177


Urban Growth Boundary (UGB): An imaginary boundary that encloses all of the land that is
expected to be developed at some point in the future.

Utility Function: A mathematical function that expresses the advantages and disadvantages of a
particular transportation mode.

Zones: Regions within the study area that contain homogenous land uses and can be described
accurately by only a few variables.

You might also like