Professional Documents
Culture Documents
www.elsevier.com/locate/dsw
Abstract
Manufacturing facility layout problem is an unstructured decision-making problem due to natural vagueness associated
with the inputs to the models. Arbitrary numerical ratings are assigned for relationship chart to determine facility selection
routine. This paper presents a distinct decision support system based on multifactor fuzzy inference system (FIS) for the
development of facility layout with fixed pickup/drop-off points. The algorithm searches several candidate points with
different orientation of incoming machine blocks in order to minimize flow cost, dead space and area required for the
development of layout. The proposed methodology is coded in C+ language and implemented in a Pentium III, 550-MHz
machine. The experimental results with a test problem are illustrated with encouraging result with its advanced soft
computational effectiveness.
D 2004 Elsevier B.V. All rights reserved.
Keywords: Facility layout; Fuzzy decision; Flow cost; Dead space; Minimum required area
0167-9236/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.dss.2003.12.007
306 S.K. Deb, B. Bhattacharyya / Decision Support Systems 40 (2005) 305–314
and their fuzzy relation. The integrated expert system development of facility layout. The applicability of
approach of Abdou and Dutta [1] determines the the suggested methodology is demonstrated with a
movement of material-handling equipments first to six-machine layout problem considering subjective
determine its effect on the layout. factors such as supervision, information and envi-
Recently, Dweiri [6] proposed a fuzzy decision- ronmental condition. The material flows between the
making application for developing relationship charts different departments are assumed as the objective
and comparing the layouts generated with them. The factor for the development of selection routine. The
concept has been applied to develop layout in the line heuristic search algorithms proposed in this paper
of computer-relative layout programme that was not take care of optimal placement of incoming facilities
efficient as compared to conventional layout proce- based on a multi-criteria optimization function. The
dure. He suggested further research work to improve performance of the proposed multifactor fuzzy facil-
the procedure or developing a new algorithm for the ity selection routine is compared with the multifactor
facility layout design. normalized facility selection routine for the develop-
The facility layout problems fall in to the class of ment of facility layout under the auspices of a
NP-complete solutions, and heuristic approaches are manufacturing environment.
usually adopted to develop the layout [8]. Most of the
models and algorithms available in the literature are
based on the quadratic assignment problem with an 2. Fuzzy set and decision-making system
objective to minimize transportation costs or maxi-
mize total closeness rating. The facility selection A fuzzy set can be thought of a class of concepts/
routine required for the development of layout was objects in which no well-defined boundary exists
solved by considering a single quantitative factor as between the concepts/objects that belong to the class
flow chart. Moreover, the move (distance) traversed is and those which do not belong. Formally, if
considered from center to center of the departments B={xi | i N} is a set of objects, then the fuzzy set
without considering the practical issue of entry and C on B is defined by its membership function fC(x)
exit of the departments. that assigns to each element x B a real number in
Most of the existing methods have been devel- the interval [0,1] which represents the grade of
oped on the grid-based system without considering membership of x in C or the degree to which x
actual dimensions of departmental block and entry/ belongs to C. Thus, C can be written as
exit locations, thus resulting in irregular shapes.
Deb et al. [3] have already developed a hybrid
modeling for the management of material-handling C fC x x x B ; B 0 1
equipment selection planning while generating a
manufacturing facility layout. The authors herein
have also proposed different projects of integrating Linguistic variables are words in natural language,
facility layout and material-handling equipment se- while numerical variables use numbers as values.
lection by using a knowledge base and optimization Since words are usually less precise than numbers,
approach [5]. They utilized the material flow inter- linguistic variables provide a method to characterize
action matrix in finding the facility placement complex systems that are ill structured to be de-
sequence. The authors have already developed a scribed in traditional quantitative terms. A linguistic
decision model and algorithm for material-handling variable is defined by the name of the variable x, and
equipment selection routine under facility layout the set term S(x) of the linguistic values of x with
planning by using fuzzy multi-criteria decision-mak- each value being a fuzzy number defined on U. For
ing methods [4]. example, if material flow (MF) is a linguistic vari-
Therefore, the present research work follows in able, its term set S(MF)={Very High (VH), High (H),
the same direction of author’s previous work to Medium (M), Low (L), Very Low (VL)}, where each
integrate various linguistic assessments to evaluate term is characterized by a fuzzy set in a universe of
facility selection routine and its impact on the discourse U.
S.K. Deb, B. Bhattacharyya / Decision Support Systems 40 (2005) 305–314 307
A fuzzy decision-making system (FDMS) consists point of edges of the rectangular machine blocks. The
of four main components [14] as shown in Fig. 1. The generation of model for layout construction is a
four components of FDMS are given below: critical step because of its unstructured and vast
nature. The complexity increases further due to mul-
(i) Fuzzification: In this interface, the different input tifactor influence on the development of facility
and output variables are measured and converted selection routine for its sequential placement in the
into natural language; open plane. Out of various types of facility layout
(ii) Knowledge base: In this component, the mem- environment, very less work has been carried out
bership functions are decided by the experts under the continual planner approach because of its
based on their knowledge of the system; complexity and flexibility in the generation process
(iii) Decision rules: In this interface, the experts’ starting from the center of the plane considering zero
decision-making ability is simulated based on a base area allocation [3 – 5]. The inherent difficulties of
fuzzy concept. The connective ‘and’ is imple- generating such kind of layout are the generation of
mented as a fuzzy conjunction in a Cartesian dead space. As the layout expands from the center of
product space in which the input variables take in the continual plane, it is very difficult to predict the
their respective universe of discourses. The possible location for incoming facilities. The formu-
minimum operator is used, and the rules are in lation of facility layout design problem mainly con-
the form of IF– THEN statement; and sists of two modules:
(iv) Defuzzification: In this interface, the fuzzy
outputs are converted into crisp (no fuzzy) values (i) facility selection routine and
by center-of-area (COA) method. (ii) facility placement routine
as rules in the knowledge base of the FIS. The tion link and environmental link are considered as
development of rules may be time consuming. An trapezoidal membership functions. The number of
intuitively developed strategy for finding the rating of rules (N) used in controlling the system using fuzzy
each move based on the values of their relationships control is represented by:
can be summarized as follows:
m n
N Li
(i) If the material flow (MF) relationship between j 1
i 1
two facilities is very high, then they should be
located very close to each other, i.e., rating given where, m = number of set of rules, Li = number of
is ‘A’. membership functions or levels, N = number of input
(ii) If the supervision link (SL) between two facilities variables used in one set of rules. When m = 1, n = 4
is very high, then they should be located very and Li = 5, then number of rules (N) becomes:
close to each other, i.e., rating given is ‘A’. 5 5 5 5 = 625.
(iii) If the environmental link (EL) between two The following steps are established to find the
departments is very hazardous, then they should selection routine of facilities in an open field:
be located very far to each other, i.e., rating
given is ‘U’. 1. Prepare the input values for all moves (activities).
(iv) If the information flow (IF) between two facilities The total activities are n (n 1), where n is
is very strong, then they should be located very number of departments.
close to each other, i.e., rating given is ‘A’. 2. Find the minimum values of the input variables’
membership values using minimum operator for
The mapping of the inputs to the outputs for a every activity.
fuzzy system is in part characterized by a set of 3. Scan the heuristic design rules for all the moves,
condition action rules in the form of IF– THEN. and find the crisp output of the rating (Rijc) by using
The connective ‘and’ is implemented as a fuzzy center-of-area (COA) method.
conjunction in a Cartesian product space in which 4. Determine the fuzzy rating matrix as Rn*n =
the input variables take on their respective universe of [Rijc] i,j = 1,2 n and Rijc = 0 for i = j.
discourses. For this study, the [minimum] operator 5. Calculate the total fuzzy rating of the ith
will be used. The membership value of the control department with the other departments and find
action of each rule is the minimum value of the input the maximum value ( Fk) to select the first facility
variables’ membership values. In this paper, multi- as department ‘k’.
input, single-output (MISO) is considered under heu-
ristic design rules in the following form: Fi Rcij Rcji i j 1 2...n
j
IF MF is VH and IF is VH Fk max Fi i 1 2...n
THEN rating is A
6. Next, find the department that has maximum fuzzy
IF MF is H and EL is hazardous rating value with the facility already included in
the selection routine.
THEN rating is
7. Repeat step 6 until all facilities are included in the
The values of linguistic variables are considered selection routine.
within a designed weighing scale [0, 10] with levels
[VL, L, M, H, VH], and the values of the output 3.2. Facility placement routine
rating are designed within a weighing scale [0,6]
with generally accepted levels [X,U,O,I,E,A]. The Heuristic is deterministic and hence suboptimal.
triangular membership function is considered for Not that this is a bad thing. The generalized QAP
objective variable material flow and rating score. facility layout formulation is NP complete. Indeed, the
The subjective variables supervision link, informa- heuristic is novel and provides an interesting approach
310 S.K. Deb, B. Bhattacharyya / Decision Support Systems 40 (2005) 305–314
Table 2 Table 4
Fuzzy system input data and output rating Multifactor-normalized relation matrix
Move MF SL EL IL Rcij Rnij Machines M1 M2 M3 M4 M5 M6
number M1 – 0.20 0.16 0.13 0.16 0.16
1-2 1 5 9 10 3.00 0.20 M2 0.18 – 0.17 0.01 0.11 0.16
2-1 5 8 2 2 3.00 0.18 M3 0.16 0.21 – 0.15 0.12 0.11
1-3 2 3 8 5 1.00 0.15 M4 0.19 0.05 00 – 0.38 0.20
3-1 2 2 7 6 1.99 0.15 M5 0.11 0.17 00 0.13 – 0.11
1-4 1 1 6 8 1.00 0.13 M6 0.12 0.16 0.07 0.20 0.16 –
4-1 4 6 1 9 3.00 0.19
1-5 2 8 5 2 3.00 0.15
5-1 1 3 4 5 1.99 0.11
1-6 3 5 2 8 3.00 0.16 3.3. Steps of algorithm for facility placement routine
6-1 0 7 3 6 3.00 0.12
2-3 1 2 5 7 1.99 0.17
3-2 3 2 6 8 1.99 0.21
1. Find the facility selection routine applying
2-4 2 9 9 5 3.00 0.01 multifactor FIS.
4-2 0 1 0 0 1.04 0.05 2. Locate the first block at the center of the open
2-5 1 2 1 2 1.09 0.11 plane horizontally.
5-2 2 3 4 3 1.99 0.17 3. Select the next block for placement according to
2-6 2 4 6 9 3.00 0.15
6-2 2 6 5 4 3.00 0.16
placement routine of facilities.
3-4 3 8 3 3 3.00 0.15 4. Select the candidate point, and check the feasible
4-3 0 0 0 0 1.04 00 quarter. If not feasible, go to step 7, else go to next
3-5 2 1 5 6 1.99 0.12 step.
5-3 0 0 0 0 1.04 00 5. Locate the block according to placement possibil-
3-6 1 1 2 2 1.99 0.11
6-3 0 2 0 0 3.00 0.07
ities, and check for nonoverlapping. If not satisfied,
4-5 1 1 1 5 1.00 0.28 repeat next possible placement as explained in
5-4 5 8 9 9 3.00 0.13 Section 4, else go to step 6.
4-6 2 5 6 1 3.00 0.20 6. Calculate the value of objective function, if it is
6-4 2 8 8 6 3.00 0.20 better than previous update configuration and
5-6 1 3 5 4 1.99 0.11
6-5 10 9 7 3 1.99 0.16
objective function value. Go to step 5 for searching
other possibilities at the candidate point.
MF = material flow, SL = supervision link, EL = environmental link,
IL = information link. 7. Select next candidate point. If all candidate points
of the selected block are already considered, go to
step 8; otherwise, go to step 4.
unit flow cost coefficient (i.e., cij = 1) and unit penalty 8. Select the next block. If all blocks are selected, go
cost coefficient for the dead space (i.e., Pc = 1). The to step 9, else go to step 3.
penalty cost coefficient is defined as the cost of dead
space per unit area that is considered as a parameter,
which varies from place to place.
Table 3
Fuzzy crisp activity relation matrix
Machines M1 M2 M3 M4 M5 M6
M1 – 3.00 1.00 1.00 3.00 3.00
M2 3.00 – 1.99 3.00 1.09 3.00
M3 1.99 1.99 – 3.00 1.99 1.99
M4 3.00 1.04 1.04 – 1.00 3.00
M5 1.99 1.99 1.04 3.00 – 1.99
M6 3.00 3.00 3.00 3.00 1.99 –
Fig. 4. Membership function for material flow.
312 S.K. Deb, B. Bhattacharyya / Decision Support Systems 40 (2005) 305–314
Table 5
Experimental results using fuzzy and normalized approach
Facility Methods Flow Minimum Dead Value
layout applied cost required space of
selection (T-m) area of (m2) fuzzy
routine layout score
1-2-6-4-5-3 (A) Multifactor 2998 17280 7580 7.14
normalized
6-4-2-1-5-3 (B) Multifactor 3015 17199 5499 7.32
fuzzy method
Fig. 5. Membership function for supervision link.
obtained by applying the proposed methodology
9. Locate the block, which provides the best value of and normalized rating obtained by applying the
objective function. methodology of Harmonosky and Tothero [7] for
each move are presented in Table 2. The fuzzy
activity relation matrix and normalized activity rela-
4. Experimentation and results tion matrix are shown in Tables 3 and 4 respectively.
Membership functions of material flow, supervision
The experimentation was carried out in order to link and score are shown in Figs. 3– 5. A triangular
investigate the applicability and effectiveness of the membership function is considered for the material
proposed methodology based on multifactor fuzzy flow and score (Fig. 6). For other variables, trape-
inference system, and the results have been com- zoidal membership functions are considered. The
pared with the facility selection routine developed as values of material flow cost, minimum required area
an extension of the multifactor plant layout method- and dead space for different selection routine are
ology by Harmonosky and Tothero [7]. The algo- presented in Table 5. Figs. 7 and 8 show the layouts
rithm was coded in C+ language, and the problem developed by using the proposed algorithm under
was run on an IBM Pentium III, 550-MHz machine. multifactor fuzzy inference system and multifactor
The data table related to machine dimensions, P/D
locations and move characteristics used for computer
simulation having six machines, 30 moves under the
consideration of four influencing factors are taken
from earlier research work [3– 5] and listed in Tables
1 and 2. The values of subjective variables supervi-
sion link (SL), information link (IL), environment
link (EL) and objective variable material flow (MF)
are taken arbitrarily within designed weighing scale
[0,10]. The values of crisp output rating (R ijc )
Fig. 6. Membership function for output variable score. Fig. 7. Layout based on multifactor fuzzy selection routine.
S.K. Deb, B. Bhattacharyya / Decision Support Systems 40 (2005) 305–314 313
6. Future research
normalized selection routine as discussed in Section A.1 . Minimization of material flow cost
3. The values of FC, DS and MRAL are converted
to fuzzy score and its value for the six-machine Minimize Zj1 cij fij xjp xid yjp yid
layout problem is shown in Table 5. cji fij xip xjd yip yjd j 2 3 . . . n.
in a mathematically strict and rigorous way. The where xbj = Max{xbi} i = 1,2, n; xtj = Max{xti}
experimental results obtained from the computer sim- i = 1,2, j
n; yb= Max{ ybi} i = 1,2, n; ytj =
ulation illustrate that the proposed methodology has Max{ yti} i = 1,2, n; li and wi are the length and
been very effective in reducing MRAL and DS, while width of the ith rectangular machine block; ji 1 li
developing green-field layouts. The value of flow cost wi is the sum of all the machine blocks.
obtained by using the proposed approach is slightly
higher than the layout developed with the selection A.3 . Minimization of flow cost and dead space
routine based on multifactor-normalized method. The
slightly higher value of flow cost is comparable to the Minimize Zj w1 Zj1 w2 Zj2
fuzzy decision-making method under the present
consideration of developing green-field layouts where where w1 and w2 are weights of FC and DS; (xip,yip)
MRAL and DS are considered very important evalu- is the input coordinate of ith department; (xdi ,ydi ) is
ating parameters. Applying the fuzzy scoring meth- the exit coordinate of the ith department; (xjp,yjp) is
odology typically yields higher scores for the multi- the input coordinate of the jth department; (xjd,yjd) is
factor fuzzy approach. the exit coordinate of the jth department; fij and fji
The firm can save initial investment costs by adopt- are the material flow between departments i –j and
ing the proposed facility layout methodology. The j–i; cij and cji are the flow cost coefficients between
methodology presented in this paper is simple and moves i–j and j– i; Pc is the penalty cost coefficients
can easily be implemented on a personal computer. for dead space.
314 S.K. Deb, B. Bhattacharyya / Decision Support Systems 40 (2005) 305–314
References [12] J.A. Tompkins, J.A. White, Facilities Planning, Wiley, New
York, 1984.
[1] G. Abdou, S.P. Dutta, An integrated approach to facilities [13] L.A. Zadeh, Fuzzy sets, Information and Control 8 (1965)
338 – 553.
layout design using expert system, International Journal of
Production Research 28 (1990) 685 – 708. [14] H.J. Zimmermann, Fuzzy Sets. Decision Making and Expert
[2] J.M. Apple, Plant Layout and Material Handling, Wiley, New Systems, Kluwer Academic Publishing, Boston, 1987.
York, 1977.
[3] S.K. Deb, B. Bhattacharyya, S.K. Sorkhel, Management of
machine layout and material handling system selection using S.K. Deb is assistant professor of the
hybrid approach, 1st International Conference on Logistic and Mechanical Engineering Department,
Jorhat Engineering College, Jorhat-7, In-
Supply Chain Management, PSG Tech, India, 2001.
dia. He secured his MTech and MBA
[4] S.K. Deb, B. Bhattacharyya, S.K. Sorkhel, Material Han-
dling Equipment Selection by Fuzzy Multi-Criteria Decision degree from IIT, Kharagpur and Gauhati
Making Methods. Lecture Notes in Artificial Intelligence, University, respectively. Recently, he has
obtained his PhD (Engg) from Jadavpur
Springer-Verlag, Berlin, 2002.
[5] S.K. Deb, B. Bhattacharyya, S.K. Sorkhel, Facility layout and University, Kolkata, India. The author has
material handling equipment selection planning using hybrid teaching and research experience of about
methodology, International Journal of Industrial Engineering 18 years. His area of specialization is
Facility Layout Planning and Operations
10 (3) (2003 September) 436 – 443.
[6] F. Dweiri, Fuzzy development of crisp activity relationship Management. He has published several research papers in national
charts for facilities layout, Computer and Industrial Engineer- and international journals.
ing 36 (1999) 1 – 16.
[7] C.M. Harmonosky, K. Tothero, Multi factors plant layout B. Bhattacharyya is professor and for-
methodology, International Journal of Production Research mer head of the Production Engineering
30 (1992) 1773 – 1789. Department, Jadavpur University, Kol-
[8] S.S. Heragu, A. Kusiak, Machine layout: an optimization and kata, India. He did his MProd (Engg)
knowledge-based approach, International Journal of Produc- and PhD (Engg) from Jadavpur Univer-
tion Research 28 (1990) 615 – 635. sity, Kolkata. At present, he is the coor-
[9] W. Kawasaki, G.W. Evans, A layout design heuristic employ- dinator of Center of Advanced Studies
ing theory of fuzzy set, International Journal of Production (CAS) and Quality Improvement Pro-
Research 25 (1987) 1431 – 1450. gram (QIP) of Jadavpur University. His
[10] A. Raoot, A. Rakshit, A linguistic pattern approach for mul- area of specialization is non-traditional
tiple criteria facility layout problems, International Journal of manufacturing and production manage-
Production Research 31 (1993) 203 – 222. ment. He has published about 50 research papers in international
[11] T.L. Saaty, The Analytical Hierarchy Process, McGraw-Hill, and national journals.
New York, 1980.