A CASE STUDY ON DEMAND FORCASAT: ELECTRICITY IN BANGLADESH

ABSTRACT:

IN this case study we had tried to figure out the demand of service on future periods.
For this we had select electricity sector of Bangladesh. We had collect data of demand on
previous periods and had utilized it for projecting future demands. We used various
quantitative methods like time series and trend analysis for predicting future demands and
different error analysis also included. The results of the study are useful for the energy
planning of the country.

1. INTRODUCTION:

Bangladesh is a country under development and moving towards the industrialization.
Although Bangladesh is predominantly an agricultural country but a large number of
large-scale Industries based on both indigenous and imported raw materials have been set
up. With huge population and limited resources Bangladesh is facing notable crisis of
energy. Energy is the main factor that makes every machinery run and electricity is the
main source of energy. Day by day the demand for electricity is increasing but production
is not enough. Effort on decreasing System losses and increasing production capacity has
been a major matter to be focused for Bangladesh.

2. Current situation of Bangladesh:

Only around 20% of the population (25% in urban areas and 10% in rural areas)
has access to electricity, and per capita commercial energy consumption is among
the lowest in the world.

Surface Area: 144, 0 1000 sq km
Population: 164, 7 Millions of inhabitants - 2009 (estimates after 2007)
Current GDP: 67, 8 Billions of Euros - 2009 (estimates after 2008)
GDP per capita: 411, 4 Euros - 2009 (estimates after 2007)

Electricity - production by source:
fossil fuel: 93.7%
hydro: 6.3%
nuclear: 0%
other: 0% (2001)
Electricity - consumption: 14.25 billion kWh (2001)








3. About forecasting:

Forecasting is the process of making statements about events whose actual outcomes
(typically) have not yet been observed. A commonplace example might be estimation of
the expected value for some variable of interest at some specified future date. Prediction
is a similar, but more general term. Both might refer to formal statistical methods
employing time series, cross-sectional or longitudinal data, or alternatively to less formal
judgmental methods. Usage can differ between areas of application: for example in
hydrology, the terms "forecast" and "forecasting" are sometimes reserved for estimates of
values at certain specific future times, while the term "prediction" is used for more
general estimates, such as the number of times floods will occur over a long period. Risk
and uncertainty are central to forecasting and prediction; it is generally considered good
practice to indicate the degree of uncertainty attaching to forecasts. The process of
climate change and increasing energy prices has led to the usage of Egain Forecasting of
buildings. The method uses Forecasting to reduce the energy needed to heat the building,
thus reducing the emission of greenhouse gases. Forecasting is used in the practice of
Customer Demand Planning in every day business forecasting for manufacturing
companies. The discipline of demand planning, also sometimes referred to as supply
chain forecasting, embraces both statistical forecasting and a consensus process. An
important, albeit often ignored aspect of forecasting, is the relationship it holds with
planning. Forecasting can be described as predicting what the future will look like,
whereas planning predicts what the future should look like.There is no single right
forecasting method to use. Selection of a method should be based on your objectives and
your conditions (data etc.).A good place to find a method, is by visiting a selection tree.
An example of a selection tree can be found here.




There are four steps to forecasting:


1
Set objectives for forecast - what decisions will forecast influence, and how - what
aspects of plan are vulnerable to surprise - how accurate must forecast be to be of any
use?
2 Obtain forecast - from existing sources or commissioned study.
3
Evaluate forecast, including assessing the sensitivity of planned actions on the
accuracy of the forecast - what exactly is being forecast - what assumptions does the
forecast itself make - source of forecast and past reliability - what objectives of
forecasters, and how do they compare with your own?
4 Disseminate the forecast, persuade others of its accuracy and appropriateness.









4.FORECASTING TECHNIQUES WE HAVE USED:

FOR the projection of future demand on the data we had collected, we find time series
analysis and casual model best fits. Forecast model we selected is illustrated on following
flow chart.








4.1 Time series analysis:
4.1.1 Simple moving average: Moving average techniques forecast demand by
calculating an average of actual demands from a specified number of prior periods each
new forecast drops the demand in the oldest period and replaces it with the demand in the
most recent period; thus, the data in the calculation "moves" over time
Time series analysis Casual model
Simple moving average Linear regression
TREND
ANALYSIS

DATA COLLECTION
Fig; Forecasting model
simple moving average:
where N = total number of periods in the average
forecast for period t+1: F
t+1
= A
t

Key Decision: N - How many periods should be considered in the forecast
Tradeoff: Higher value of N - greater smoothing, lower responsiveness
Lower value of N - less smoothing, more responsiveness
- the more periods (N) over which the moving average is calculated, the less susceptible
the forecast is to random variations, but the less responsive it is to changes
- a large value of N is appropriate if the underlying pattern of demand is stable
- a smaller value of N is appropriate if the underlying pattern is changing or if it is
important to identify short-term

4.1.2 Trend analysis: There are no proven "automatic" techniques to identify trend
components in the time series data; however, as long as the trend is monotonous
(consistently increasing or decreasing) that part of data analysis is typically not very
difficult. If the time series data contain considerable error, then the first step in the
process of trend identification is smoothing.
Smoothing. Smoothing always involves some form of local averaging of data such that
the nonsystematic components of individual observations cancel each other out. The most
common technique is moving average smoothing which replaces each element of the
series by either the simple or weighted average of n surrounding elements, where n is the
width of the smoothing "window" (see Box & Jenkins, 1976; Velleman & Hoaglin,
1981). Medians can be used instead of means. The main advantage of median as
compared to moving average smoothing is that its results are less biased by outliers
(within the smoothing window). Thus, if there are outliers in the data (e.g., due to
measurement errors), median smoothing typically produces smoother or at least more
"reliable" curves than moving average based on the same window width. The main
disadvantage of median smoothing is that in the absence of clear outliers it may produce
more "jagged" curves than moving average and it does not allow for weighting.
In the relatively less common cases (in time series data), when the measurement error is
very large, the distance weighted least squares smoothing or negative exponentially
weighted smoothing techniques can be used. All those methods will filter out the noise
and convert the data into a smooth curve that is relatively unbiased by outliers (see the
respective sections on each of those methods for more details). Series with relatively few
and systematically distributed points can be smoothed with bicubic splines.
Fitting a function. Many monotonous time series data can be adequately approximated
by a linear function; if there is a clear monotonous nonlinear component, the data first
need to be transformed to remove the nonlinearity. Usually a logarithmic, exponential, or
(less often) polynomial function can be used.

4.2 Casual modeling:
A reliable method of prediction becomes possible if you have, through research, obtained
a model which not only describes (as in the previous section) the development of the
phenomenon to be predicted but also explains it, in other words enumerates the reasons
why it happens. In the best case the reasons and their outcomes are assembled as a model
defining the dynamic invariance of change in the process to be predicted.
The weather, for example, need today no more be predicted on the basis of a statistical
association of air pressure and weather only. The science of meteorology has lately
advanced so much that we now know and can make use of the invariable structure of
moving cyclones which explains the changes both in air pressure and in weather. Even
the proverb about red skies has now been given an explanation:
"Because the weather patterns in North America generally move from west to east, when
clouds arrive overhead at sunrise the sky will appear red, signalling a storm "moving in".
When the storm eventually passes, the sky will clear in the western sky. If sunset occurs
simultaneously, the light will cast a red glow on the clouds above, now moving towards
the east."
The most elementary method of forecasting on the basis of a causal model is to use the
model just like a statistical association, explained earlier. This is particularly easy when
one of the variables in the model is time: then you just insert the right year into the
model, and it immediately becomes the desired forecast.
If time is not included in the causal model, the model may still be helpful, because you
can often predict the development of the independent variable easier than the future of the
dependent variable or of the entire system - not least because of the fact that a reason
normally precedes its result and it is thus not so distant in the future as the outcome will
be.
When you know the causal relationships between the variables you will be able to use
much more advanced methods of forecasting than mere statistical models would permit.
These include:
You can better assess whether the model remains valid also in the future.
You can assess with sensitivity analysis the probable error of the forecast.
You will also be able to modify the model, with a high degree of reliability,
according to the requirements of the situation.
If you want not only to forecast but also to change the future, you can pinpoint
those changes in the independent variables that are needed to cause the desired
change in the dependent variables.
The causal model is often so complicated
that it is best managed by using a
computer. Even then, you will usually need
an illustrative presentation of your model
to clarify your thinking and finally to be
presented in the report. In such an
illustration you will need a notation system
to describe the various logical relations
between the variables. The computer
program will often be able to print out the
model, using its in-built notations. If you
can find no suitable ready made notation
systems, you can devise one.








5.2.1 Linear regression: simple linear regression is fitting a straight line to a set of
paired observation.Mathematical expression for straight line is

Y=mx+c
Where
Y=dependent variable
m=slope of fitted line
x=independent variable
c=intercept on y axis
This technique of fitting straight line can be used as forecasting model by plotting
independent variable to figure out dependent variable. Slop and intercept of fitted line can
be determined by following formula

m=(nx
i
y
i
-xiyi)/(nx
i
2
-(x
i
)
2
) and

c=y
i
/n-mx
i
/n
























5.3 Error calculations techniques:
5.3.1 Mean absolute deviation:

=
=
n
t
t t
F D
n
MAD
1
1

Where
n=number of observation
D
t
=current demand
F
t
=current forecast
5.3.2 Mean square error
2
1
) (
1
t
n
t
t
F D
n
MSE =

=

Where
n=number of observation
D
t
=current demand
F
t
=current forecast
5.3.3 Mean absolute percentage error:

=
n
t
t
t t
D
F D
n
MAPE
1
100


Where
n=number of observation
D
t
=current demand
F
t
=current forecast
5.3.4 Cumulative Sum of Forecast Error:

5.3.5 Mean forecast error:
) (
1
1
t
n
t
t
F D
n
MFE =

=



5.4 Tracking signal:


6.OUR STEPWISE OPERATIONS:

6.1 DATA COLLECTION;
We had collected all data used in our forecasting model from published e-books,
Bangladesh bureau of statistics and websites. We collected data on population, electricity
production, system loss and load shedding previous periods.



Fig: Data obtained
6.2 FORMULATION OF DATA: We then calculate shortage and demand from this data.





Fig: Calculated demand and shortage


Fig: demand vs population

6.3 TIMESERIES ANALYSIS:

6.3.1 Simple moving average:
We plotted obtained data with reference of time and used simple moving average of one
period to forecast short period demand in future.

Fig: Simple moving average




Fig; Excel sheet for calculation of single period moving average


6.3.2 Trend analysis: Further trend analysis had been done to obtain forecasting on
long horizon.

Fig; Excel sheet for trend analysis;

Fig; Excel sheet for trend analysis;







Fig : Trend analysis









Fig: Yearly increasing demand
















6.3 Casual model:
6.3.1 Linear regression:








uemand(v) lorecasL 8SlL
Cum
MAu
1rack
Slanal

1996 1980.86

1997 2143.34 1980.86 164.49 164.49 1

1998 2178.43 2143.34 197.38 98.79 2

1999 2444.22 2178.43 463.36 134.43 3

2000 2338.86 2444.22 378 137.18 2.76

2001 2832.31 2338.86 831.63 204.47 4.17

2002 2706.33 2832.31 723.68 191.39 3.79

2003 3037.64 2706.33 1036.78 211.33 3

2004 3428.97 3037.64 1448.12 233.83 6.19

2003 3666.13 3428.97 1683.29 234.22 7.2

2006 3903.62 3666.13 1922.77 234.34 8.2

1C1ALS 30683.13

AvL8ACL 2789.38

nexL perlod forecasL 3903.62










3.6
n 8las MAu MSL SLandard error MAL
1 192.28 234.34 72170.61 283.18 0.08
2 289 289 109937 331.68 0.09
3 402.72 402.72 193163.6 469.83 0.13
4 497.31 497.31 286803.7 378.43 0.13
3 629.11 629.11 424164.3 713.44 0.19
6 713.13 713.13 330333.2 829.36 0.21
7 872.94 872.94 778017.4 1018.31 0.23
8 1021.66 1021.66 1043372 1232.34 0.28
9 1123.11 1123.11 1261999 1388.71 0.3
10 1223.67 1223.67 1302273

0.31













6.3.2 Demand vs. population regression analysis:


Fig: Excel sheet


Fig: scatter plot demand vs.. Population

Fig: After adding trend line


Conclusion :
The most sensitive issue corresponding to modern human civilization is electricity
except which everything will be destroyed. By our forecasting case study we have
evaluate trend of electricity demand after 14
th
periods (2020) is nearly seven thousand
megawatt per hour. Due to the lacking of exact data of previous periods our forecasted
model might not be totally reliable. Our forecasting model is enough able to demonstrate
the demand patterns of electricity.


References:
1. Bangladesh power sector data book
Prepared by: Engr. Abdur Rouf, Engr. KM Nayeem Khan, Engr. Shazibul Hoque
Published date: june 2006
2.3rd International Conference on Electrical & Computer Engineering
ICECE 2004, 28-30 December 2004, Dhaka, Bangladesh
3. Source: IndexMundi.com


Software used for calculation and forecasting: Microsoft excel