Beruflich Dokumente
Kultur Dokumente
A Thesis Submitted in
Partial Fulfilment of the Requirement for the Degree of
Master of Technology
In
CAD-CAM & Automation
Submitted by
Bablu Kumar Mandal
15-22-305
Under the supervision of
Dr. Saikat Ranjan Maity
I declare that the presented thesis represents mostly my own ideas and work in my own
words. Where other ideas or words have been included, I have adequately citied and listed
in the reference materials. The thesis has been prepared without resorting to plagiarism. I
have adhered to all principles of academic honesty and integrity. No falsified or fabricated
data have been presented in the thesis. I understand that any violation of the above will cause
for disciplinary action by the Institute, including revoking the conferred degree, if conferred,
and can also evoke penal action from the sources which have not been properly cited or from
whom proper permission has not been taken.
-------------------------
Bablu Kumar Mandal
Scholar No.: 15-22-305
Date: ----/----/------
i
CERTIFICATE
This is to certify that the work contained in this thesis entitled Application of decision
making methods for selection of Advance Manufacturing Systems submitted by Bablu
Kumar Mandal, Registration number 15-22-305 for the award of Master of Technology
is absolutely based on his own work carried out under my supervision and that this thesis
has not been submitted elsewhere for any degree.
-------------------
Dr. Saikat Ranjan Maity
Assistant Professor
Date: ----/----/------- Department of Mechanical Engineering
National Institute of Technology Silchar
ii
ABSTRACT
Various Multi-Criteria Decision Making (MCDM) methods have been developed to support
the decision-making process. The main aim of all MCDM methods is to obtain a ranking of
the alternatives and select the best one under conflicting criteria. Selection of advanced
manufacturing systems (AMS) is a multi-criteria decision-making problem, based on
assessing a large number of conflicting quantitative and qualitative criteria. Decision making
for AMS selection has become more complicated due to the availability of a wide range of
feasible alternatives. In the present work three decision making methods (Measuring
Attractiveness by Categorically Based Evaluation Technique (MACBETH), Grey
Technique of Order Preference by Similarity to Ideal Solution (GTOPSIS) and Grey
Additive Ratio Assessment (G-ARAS) are consider to solve four AMS selection problems
(e.g., a Flexible manufacturing system, Material handling equipment, Green supplier, a
Rapid prototyping process) having both qualitative and quantitative attributes. Four
performance tests are conducted for the ranking performance comparison and also for
measuring the degree of agreement between the rankings derived by the considered methods.
In all these cases, the rankings of the alternatives obtained almost corroborate with those
derived by the past researchers, proving the feasibility of these method as an effective
MCDM techniques.
iii
ACKNOWLEDGEMENT
I would like to take this opportunity to express my sincere gratitude and regards to my
supervisor Dr. Saikat Ranjan Maity, Assistant Professor, Department of Mechanical
Engineering, National Institute of Technology Silchar, for his constant support and guidance
in completion of the project work.
At the end, I would like to thank the Almighty God for bestowing upon me with its gracious
blessings to successfully completing the project work.
iv
Table of Contents
Declaration i
Certificate ii
Abstract iii
Acknowledgement iv
Table of Contents v
Nomenclature x
List of Abbreviations xi
v
4.3 Flexible Manufacturing System Selection using Macbeth method 22
4.4 Flexible Manufacturing System Selection using Grey TOPSIS 24
4.5 Flexible manufacturing system selection using Grey ARAS method 27
4.6 Ranking Performance analysis for FMS selection 28
References 30
Chapter 5: Material Handling Equipment selection 31-40
5.1 Need for Material Handling Equipment Selection 31
5.2 Literature Survey on Material handling Equipment Selection 31
5.3 Material Handling Equipment Selection using MACBETH method 32
5.4 Material Handling Equipment Selection using GTOPSIS 35
5.5 Material Handling Equipment Selection using Grey ARAS method 37
5.6 Ranking Performance analysis for Material handling Equipment 37
References 39
Chapter 6: Green Supplier Selection 41-50
6.1 Need for Green Supplier Selection 41
6.2 Literature survey on Green Supplier Selection 41
6.3 Green Supplier Selection using Macbeth method 42
6.4 Green supplier selection using GTOPSIS method 45
6.5 Green supplier selection using ARAS method 48
6.6 Ranking Performance analysis for Green supplier selection 49
References 50
Chapter 7: Rapid Prototyping Process Selection 51-60
7.1 Need for Rapid prototyping Process Selection 51
7.2 Literature survey on rapid prototyping process selection 51
7.3 Rapid prototyping process selection using Macbeth method 52
7.4 Rapid prototyping process selection using G-TOPSIS method 55
7.5 Rapid prototyping process selection using ARAS method 56
7. 6 Ranking Performance analysis for Rapid prototyping process selection 57
References 59
Chapter 8: Conclusion and Future Scope 61-62
8.1 Conclusion 61
8.2 Future scope 62
List of Publications 63
vi
List of Figures
vii
List of Tables
viii
Table 6.5 Weighted normalized decision matrix in ARAS method for 48
Green supplier selection problem
Table 6.6 Si and Ui values in ARAS method for Green supplier selection 48
problem
Table 6.7 Comparison table for Green supplier selection 49
Table 6.8 Comparative study on ranking performance for Green supplier 49
selection
Table 7.1 Initial decision matrix for RP process selection 52
Table 7.2 Grey decision matrix for RP process selection 55
Table 7.3 Normalized values of alternatives and positive/negative ideal 56
values for RP process selection
Table 7.4 Separation measures and the relative closeness of each 56
alternative for RP process selection
Table 7.5 Weighted normalized decision matrix in ARAS method for 57
RP process selection
Table 7.6 Si and Ui values in ARAS method for RP process selection 57
Table 7.7 Comparison table for RP process selection 58
Table 7.8 Comparative study on ranking performance for RP process 58
selection
ix
Nomenclature
Symbol Description
Ui Utility degree
Si Optimality function
x
List of Abbreviations
xi
CHAPTER
1
Introduction
1
robotics, rapid prototyping, environmentally sustainable technologies, etc., which have
become an integral part of manufacturing. Parallel to this is rapid strides in the
development of new products, and the emergence of an open economy leading to global
competition. Manufacturing industries are compelled to move away from traditional setups
to more responsive and dynamic ones. Many new concepts have emerged from these
changes, sustained by strategies aimed at meeting the challenges arising from global
markets. Product attributes like quality, reliability, cost, life-cycle prediction, and the
organizational ability to meet market pressures like delivery and service, have come into
focus. A long array of emerging technologies has opened up the potential for a variety of
new products. Fast-changing technologies on the product front cautioned the need for an
equally fast response from the manufacturing industries. The old, traditional model of
unfocused, short-term views and non-holistic vision is becoming replaced by the
enlightened approach of focused, holistic and strategic vision.
To meet the challenges, manufacturing industries have to select appropriate product
designs, manufacturing strategies, manufacturing processes, work piece and tool materials,
machinery and equipment, etc. The selection decisions are more complex, as decision
making is challenging today. Necessary conditions for achieving effective decision making
consist in understanding the current and upcoming events and factors influencing the
whole manufacturing environment, in examine the nature of decision-making processes
and the reach of different typologies of techniques and methods, and finally in structuring
appropriately the decision-making approach based on a wide range of issues related to
manufacturing systems design, planning, and management. Decision makers in the
manufacturing sector frequently face the problem of assessing a wide range of alternatives,
and selecting one of them based on a set of conflicting criteria.
In manufacturing sector there is wide range of alternative option for decision makers.
Some of the important decision-making situations in the manufacturing environment are
listed below:
Material selection for a given engineering problem
Evaluation of best product designs
Evaluation of machinability for work materials
Selection of cutting fluid for a given machining application
Selection and evaluation of modern machining methods
Selection and evaluation of flexible manufacturing systems
Selection of machines group in a flexible manufacturing cell
Analysis of cause of failure of machine tools
Selection of robot for a given industrial application
Selection of automated inspection systems
Selection of material handling equipment
Selection of a rapid prototyping process in rapid product development
Selection of software for design and manufacturing applications
Selection of the most appropriate welding process for a given job
Mouldability analysis of parts
2
Evaluation of metal stamping layouts
Selection of forging conditions for a given component
Evaluation and examine of environmentally conscious manufacturing process
Environmental impact estimation of manufacturing processes
Evaluation of risk in green manufacturing
Selection of best product end-of-life scenario
Integrated project selection and evaluation
Selection of facility location
Selection of vendor in a supply chain environment
It must be noted that in choosing the right alternative, there is not always a single
definite criterion of selection, so decision makers have to take into account a large number
of criteria including economic, ethical, political, legal, social factors and technological.
There is a need for systematic, simple, logical methods and mathematical tools to guide
decision makers in considering a number of selection criteria and their interrelations.
The objective of any selection procedure is to identify the appropriate selection
criteria, and find the most appropriate combination of criteria in conjunction with the real
requirement. Thus, efforts need to be developed to identify those criteria that influence an
alternative selection for a given problem, using simple and logical methods, to eliminate
unsuitable alternatives, and to select the perfect alternative to strengthen existing selection
procedures.
3
By the help of MCDM, decision-makers are able to handle complex and difficult
decisions.
i) Firstly, they can compare end results using their intuition and using an MCDM
analysis supported by user-friendly software leading to enhancing their level of
understanding and learning, and subsequently improving their decision-making.
ii) Secondly, the steps in MCDM can be undertaken in an organization
without difficulties or interpersonal conflict, and encourage people to shift
from their intuition to rational decision making.
iii) Thirdly, a MCDM analysis can be conducted for group decision-making,
accommodating many participants at different time and places.
4
CHAPTER
2
Literature review
5
affecting the performance of advanced manufacturing technologies in Indian manufacturing
organisations and proposed certain initiatives for enhancing the performance of those
organisations in pursuit of agility-based competitiveness. Saen [14] used DEA for selecting the
best technologies in presence of dual-role factors. Rao et al. [15] adopted an MCDM approach
for robot selection using integrated weights of the selection attributes by combining subjective
and objective weights of importance. Kreng et al. [16] used an extended AHP method for
strategic justification of AMS by taking into account both the tangible and intangible attributes.
Dhal et al. [17] proposed a fuzzy MCDM approach based on grey theory to systematically
evaluate FMS alternatives. Joseph and Sridharan [18] developed a discrete-event simulation
model for scheduling decision-making problems in an FMS and applied the preference selection
index (PSI) method to rank different combinations of scheduling rules for part launching, part
routing and part sequencing decisions. Khalili-Damghani and Abtahi [19] presented a fuzzy
DEA-based procedure to measure the efficiency of a complicated process in just-in-time
manufacturing environment. Maniya and Bhatt [20] applied PSI method for selection of
appropriate FMS alternative. Three different FMS selection problems were considered, and
examined to demonstrate, validate and check the reliability of PSI method. Koulouriotis and
Ketipi [21] developed a fuzzy digraph method for robot evaluation and selection according to a
given industrial application. Taha et al. [22] proposed an AHP-based approach for selection of
AMS in an aircraft industry based on tangible and intangible criteria. Tansel [23] integrated
technique for order preference by similarity to ideal solution (TOPSIS) with design of
experiments to solve different AMS selection problems, and identified the critical AMS selection
attributes and their interactions by fitting a polynomial to the experimental data in a multiple
linear regression analysis which would greatly reduce cost, time and calculations as involved in
TOPSIS model. Ilangkumaran et al. [24] developed an evaluation model based on fuzzy AHP
and fuzzy visekriterijumskokompromisnorangiranje (VIKOR) for selecting the best machine tool
among various alternatives. Aya and zdemir [25] applied modified TOPSIS and analytic
network process (ANP) methods to solve a machine tool selection problem. The ANP method
was used to determine the relative weights of a set of evaluation criteria, as the modified TOPSIS
method was utilised to rank the competing machine tool alternatives in terms of their overall
performance. et al. [26] developed a two-phase robot selection decision support system
(ROBSEL) to help the decision makers in robot selection decisions. In the first phase of the
decision support system, the user would obtain the feasible set of robots by providing limited
values for 15 requirements. ROBSEL would then use fuzzy AHP to rank the feasible robots in
the second phase.
6
efficient in dealing with tangible attributes but they cannot perform very well if the criteria
values are expressed qualitatively. TOPSIS method uses the Euclidean distance algorithm in
principle, but this algorithm does not consider the correlation of attributes. On the other hand,
AHP can deal with tangible as well as non-tangible attributes, specially, when the subjective
judgements of different individuals play an important role in the decision-making process. But, it
also becomes very difficult to keep the consistency of the pair-wise comparison matrices, when it
is used with a large number of attributes or alternatives. As the number of alternatives increases,
the amount of calculations rises quite rapidly and the computational process becomes quite
complicated. AHP also needs ratio scales for pair-wise comparisons. The GRA method reflects
the trend relationship between an alternative and the ideal alternative, but it cannot exhibit the
situational relationship between the alternative and the ideal alternative. In case of DEA, the
results are potentially sensitive to the selection of inputs and outputs, so their relative importance
needs to be analysed prior to calculation. However, there is no way to test their appropriateness.
The number of efficient decision-making units on the frontier tends to increase with the number
of input and output variables. On the other hand, it is extremely sensitive to variable selection,
data uncertainty and errors. In ELECTRE method, exploitation of alternatives becomes quite
difficult in presence of multiple actions. PROMETHEE method can only be used with criteria
where differences in evaluations are meaningful. For criteria with an ordinal scale, these
differences have no mathematical meaning and such criteria cannot be used in PROMETHEE
method-based applications. Also, it is not possible to take discordance into account when
constructing the outranking relation in PROMETHEE method. Furthermore, implementing
ELECTRE and PROMETHEE methods require to determine values of some preference
parameters, like indifference threshold, veto threshold and concordant coalition parameter, etc.
So, there is a need to search for an efficient and accurate method that can give more precise
ranking of AMS alternatives, because most of the previously applied MCDM methods require
detailed information with respect to criteria weights, preference functions, etc.
7
deals with Material Handling Equipment selection for a south east India company. Chapter 6
deals with selection of green supplier selection for an Iranian diary company and chapter 7 deals
with Rapid prototyping process selection for a fictitious decision maker. Finally, concluding
remarks of this dissertation have been presented in subsequent chapter end.
References
[1] Talluri, S. and Yoon, K.P. (2000) A cone-ratio DEA approach for AMT justification, International
Journal of Production Economics, Vol. 66, No. 2, pp.119129
[2] Karsak, E.E. and Tolga, E. (2001) Fuzzy multi-criteria decision making procedure for evaluating advanced
manufacturing system investments, International Journal of Production Economics, Vol. 69, No. 1,
pp.4964.
[3] Yurdakul, M. (2004) Selection of computer-integrated manufacturing technologies using a combined
analytic hierarchy process and goal programming model, Robotics and Computer-Integrated
Manufacturing, Vol. 20, No. 4, pp.329340.
[4] Saen, R.F. (2006) A decision model for technology selection in the existence of both cardinal and ordinal
data, Applied Mathematics and Computation, Vol. 181, No. 2, pp.16001608.
[5] Chan, F.T.S., Chan, H.K., Chan, M.H. and Humphreys, P.K. (2006) An integrated fuzzy approach for the
selection of manufacturing technologies, International Journal of Advanced Manufacturing Technology,
Vol. 27, Nos. 78, pp.747758.
[6] Armillotta, A. (2008) Selection of layered manufacturing techniques by an adaptive AHP decision model,
Robotics and Computer-Integrated Manufacturing, Vol. 24, No. 3, pp.450461.
[7] Ahmari, A.M.A. (2008) A methodology for selection and evaluation of advanced manufacturing
technologies, International Journal of Computer Integrated Manufacturing, Vol. 21, No. 7, pp.778789.
[8] Karsak, E.E. (2008) Using data envelopment analysis for evaluating flexible manufacturing systems in the
presence of imprecise data, International Journal of Production Research, Vol. 35, No. 3, pp.867874.
[9] Durn, O. and Aguilo, J. (2008) Computer-aided machine-tool selection based on a fuzzy-AHP approach,
Expert Systems with Applications, Vol. 34, No. 3, pp.17871794.
[10] Chuu, S.J. (2009) Selecting the advanced manufacturing technology using fuzzy multiple attributes group
decision making with multiple fuzzy information, Computers & Industrial Engineering, Vol. 57, No. 3,
pp.10331042.
[11] Amin, G.R. (2009) Optimal solution of technology selection model: a computational efficient form,
International Journal of Advanced Manufacturing Technology, Vol. 43, Nos. 910, pp.10461050.
[12] Wang, T.Y., Shaw, C.F. and Chen, Y.L. (2000) Machine selection in flexible manufacturing cell: a fuzzy
multiple attribute decision-making approach, International Journal of Production Research, Vol. 38, No. 9,
pp.20792097.
[13] Singh, H. and Khamba, J.S. (2009) Evolving the barriers for enhancing the utilisation level of advanced
manufacturing technologies (AMTs) in Indian manufacturing industry, International Journal of Advanced
Operations Management, Vol. 1, Nos. 2/3, pp.135150.
[14] Saen, R.F. (2010) Technology selection in the presence of dual-role factors, International Journal of
Advanced Operations Management, Vol. 2, Nos. 3/4, pp.249262.
[15] Rao, R.V., Patel, B.K. and Parnichkun, M. (2011) Industrial robot selection using a novel decision making
method considering objective and subjective preferences, Robotics and Autonomous Systems, Vol. 59,
No. 6, pp.367375.
[16] Kreng, V.B., Wu, C.Y. and Wang, I.C. (2011) Strategic justification of advanced manufacturing
technology using an extended AHP model, International Journal of Advanced Manufacturing Technology,
Vol. 52, Nos. 912, pp.11031113.
8
[17] Dhal, P.R., Datta, S. and Mahapatra, S.S. (2011) Flexible manufacturing system selection based on grey
relation under uncertainty, International Journal of Services and Operations Management, Vol. 8, No. 4,
pp.516534.
[18] Joseph, O.A. and Sridharan, R. (2011) Ranking of scheduling rule combinations in a
flexiblemanufacturing system using preference selection index method, International Journal ofAdvanced
Operations Management, Vol. 3, No. 2, pp.201216.
[19] Khalili-Damghani, K. and Abtahi, A-R. (2011) measuring efficiency of just in time implementation using
a fuzzy data envelopment analysis approach: real case of Iranian dairy industries, International Journal of
Advanced Operations Management, Vol. 3, Nos. 3/4, pp.337354.
[20] Maniya, K.D. and Bhatt, M.G. (2011) The selection of flexible manufacturing system using preference
selection index method, International Journal ofIndustrial and Systems Engineering, Vol. 9, No. 3,
pp.330349.
[21] Koulouriotis, D.E. and Ketipi, M.K. (2011) A fuzzy digraph method for robot evaluation and selection,
Expert Systems with Applications, Vol. 38, No. 9, pp.1190111910.
[22] Taha, Z., Banakar, Z. and Tahriri, F. (2011) Analytical hierarchy process for the selection of advanced
manufacturing technologiesin an aircraft industry, International Journal of Applied Decision Sciences,
Vol. 4, No. 2, pp.148170.
[23] Tansel, Y. (2012) An experimental design approach using TOPSIS method for the selection of computer-
integrated manufacturing technologies, Robotics and Computer-Integrated Manufacturing, Vol. 28, No. 2,
pp.245256.
[24] Ilangkumaran, M., Sasirekha, V., Anojkumar, L. and Boopathi Raja, M. (2012) Machine toolselection
using AHP and VIKOR methodologies under fuzzy environment, InternationalJournal of Modelling
inOperations Management, Vol. 2, No. 4, pp.409436.
[25] Aya, Z. and zdemir, R.G. (2012) Evaluating machine tool alternatives through modified TOPSIS and
alpha-cut based fuzzy ANP, International Journal of Production Economics, Vol. 140, No. 2, pp.630636.
[26] , Y.T., Yurdakul, M. and Dengiz, B. (2013) Development of a decision support system for robot
selection, Robotics and Computer-Integrated Manufacturing, Vol. 29, No. 4, pp.142157.
9
10
CHAPTER
3
Mathematical models
11
Table 3.1: Semantic scale of Macbeth
12
Where is a coefficient necessary to meet the condition that v(L1) and v(L2) [0,100]. The
quantified MACBETH scores are obtained by solving the related equations for all the
performance levels. Let the decision maker decides the strengths of performance, as
expressed in Table 3.2, and when all the strengths of performance levels are provided, the
matrix of judgments is ready for quantification of the data.
Table 3.2 Strengths of performance levels for kth criterion
Performance L3 L2 L1 L4
level
L3(Good) No Very strong P P
L1 No weak P
L2 No Moderate
L4(Neutral) No
From the judgments provided in Table 3.2, the following system of equations can be
extracted:
v(L3) - v(L1) = 5 , (3.2)
v(L1) - v(L2) = 2 , (3.3)
v(L2) - v(L4) = 3 , (3.4)
On solving Eqns. (3.2-3.4), the obtained solutions are = 10, v(L1) = 50 and v(L2) = 30.
The quantification of performance levels for all the remaining criteria as well as the
corresponding criteria weights can be obtained adopting the same procedure. The next step
of MACBETH method is to select the alternatives and their performance with respect to
different criteria, and enter the relevant data into M-MACBETH software. Finally, the global
attractive scores are obtained to rank the considered alternatives. The additive value model
of the following type is used to determine the overall global score of an alternative.
V X i w j v j
n
(3.5)
j 1
With,
n
v j x j 100
w j 1, w j >0 and
(3.6)
j 1 v j x j 0
x : x [0,1]
G G (3.7)
( x) : x [0,1]
G
Where,
G
x G , x X , X R (3.8)
If only the lower limit of x can be possibly estimated, x is defined as lower limit grey number
and if only the upper limit of X can be estimated, X is defined as upper limit grey number.
If the lower and upper limits of X can be estimated, X is defined as interval grey number.
The basic operation laws of grey numbers X 2 x1 , x 2
and X 2 x1 , x 2 are
expressed as follows:
Addition: X X [ X X , X X ] (3.9)
1 2 1 2 1 2
Subtraction: X X [ X X , X X ] (3.10)
1 2 1 2 1 2
min X1 X 2 , X1 X 2 , X1 X 2 , X1 X 2 ,
max X1 X 2 , X1 X 2 , X1 X 2 , X1 X 2
Multiplication: X1 X 2 (3.11)
X , X
1 1
Division: X / X , (3.12)
1 2
1
1 X X
2 2
14
x11 x12 x1n
x 21 x 22 x 2 n
X
(3.13)
xm1
xm 2 xmn
Where X ij denotes the grey evaluations of the ith alternative with respect to the jth attribute.
[ X i1 , X i 2 ,..., X im ] is the grey number evaluation series of the i-th alternative, i= 1 to
n , j=1 to m.
Step 3: To Construct the normalized grey decision matrices. The normalized values of
maximizing attributes are calculated as:
xij wij bij
xij
maxi (bij ) max (b ) max (b )
, (3.14)
,b
i ij i ij
Where wij and bij are lower and upper values of attributes, respectively.
The normalized values of minimizing attributes are calculated as:
xij wij bij
xij 1 1 ,1 (3.15)
,w max i (bij ) max (b ) max (b )
i ij i ij
Step 4: To Determine weights of the criteria q j .
Step 5: To construct the grey weighted normalized decision-making matrix.
Step 6: To Determine the positive and negative ideal alternatives for each decision-maker.
The positive ideal alternative A , and the negative ideal alternative A .
A max b / j J , min w / j J / i n (3.16)
i ij i ij
x1 , x2
,.........., x
m
A mini wij / jJ , maxi bij / jJ ' / in (3.17)
x1 , x2 ,.........., xm
Step 7: To Calculate the separation measure from the positive and negative ideal alternatives,
d i and d i , for the group. There are two sub-steps to be considered: the first one concerns
the separation measure for individuals; the second one aggregates their measures for the
group. Accordingly, the measures from the positive and negative ideal alternatives should
be calculated individually.
1 m q {( x w )2 ( x b )2 }
2 j
di j j ij j ij (3.18)
1
1 m q {( x w )2 ( x b )2}
2 j
di j j ij j ij (3.19)
1
15
Step 8: To Calculate the relative closeness ci to the positive ideal alternative for the group.
The aggregation of relative closeness for the ith alternative with respect to the positive ideal
alternative for the group can be expressed as:
di
ci (3.20)
di di
0 ci 1
Where , the larger the index value is, the better the evaluation of alternative.
Step 9: Rank the preference order. A set of alternatives now can be ranked by the descending
order of the value of ci .
Where X ij denotes the grey evaluations of the ith alternative with respect to the jth attribute.
[ X i1 , X i 2 ,., X im ] is the grey number evaluation series of the ith alternative, i=1 to
n , j=1 to m.
Step 2: To Determine the optimal value of each criterion. Let x 0j be the optimal value of jth
Criterion. If the optimal value of jth criterion is known, then
x max x for beneficial criterion (3.22)
0j i ij
x min x (3.23)
0j i ij
Step 3: To Construct the normalized grey decision matrices. The normalized values of
beneficial criteria are calculated as
16
x
ij (3.24)
x
ij m
xij
i0
S
U i , i 0, m (3.28)
i S
o
It is quite obvious that the calculated values of U lie in the interval of [0, 1] and can be
i
ordered in an increasing sequence to provide a complete ranking of the considered
alternatives.
Figure 3.1 presents the flow diagram of the proposed method. It starts from data
collection for the particular decision making method. Next probable alternatives are
determined and also the set of decisive criteria upon which the decision is based is obtained.
After that the decision matrix is formed and it is solved using the three MCDM method as
shown in the flow chart.
17
Figure 3.1: Flow chart of the proposed model
18
Reference
[1] Karande P. and Chakraborty, S. (2014). A facility layout selection model using MACBETH method,
Proceedings of the 2014 International Conference on Industrial Engineering and Operations
Management, January 7 9, Bali, Indonesia.
[2] Bana e Costa, C.A., & Oliveira, M.D. (2012). A multicriteria decision analysis model for faculty
evaluation. Omega, vol. 40 no.4, pp. 424-436
[3] Chen, M.-F., Tzeng, G.-H. (2004). Combining grey relation and TOPSIS concepts for selecting an
expatriate host country. Mathematical and Computer Modelling, vol. 40 no.13, pp. 14731490
[4] Oztaysi, B. (2014) A decision model for information technology selection using AHP integrated
TOPSIS-Grey: The case of content management systems Knowledge-Based Systems vol. 70, pp. 44-
54
[5] TURSKIS, Z., ZAVADSKAS, E. (2010) A Novel Method for Multiple Criteria Analysis: Grey
Additive Ratio Assessment (ARAS-G) Method INFORMATICA, Vol. 21, No. 4, pp. 597610
19
20
CHAPTER
4
Flexible Manufacturing System Selection
21
a need for sophisticated and applicable technique to help the decision makers for selecting
the proper FMS in a manufacturing organization.
22
Table 4.1: Decision matrix for FMS selection problem
Alternative Criteria
Table 4.1 shows the decision matrix for this FMS selection problem. For analysis
of this problem using M-MACBETH software, the MACBETH value tree is first
developed, as displayed in Fig. 4.1. In this tree, RLC, RWP, RSC, IMR and IQ are added
under the beneficial criteria node, whereas, CMC and FSU are allocated under the Non-
beneficial criteria node. Two reference levels, i.e., upper and lower reference levels are
selected for all the considered criteria. For benefit criteria, the upper reference level is the
highest performance criteria value and for cost criteria, it is the lowest performance criteria
value.
23
case, the lower reference level is selected just below the lowest performance value for
benefit criteria and just above the highest performance value for non-beneficial criteria.
The reference levels for all the considered criteria are shown in Fig. 4.2.
24
Figure 4.4: MACBETH weighing judgements for FMS selection problem
Figure 4.5 shows the criteria scores and overall scores for the eight FMS
alternatives. Based on these scores, the final ranking is obtained for all the FMS
alternatives. FMS7 is the top ranked alternative, followed by FMS4. FMS8 is the worst
chosen alternative.
Rao and Parnichkun [6] also obtained FMS 7 as the best alternative and FMS 8 as
the worst. Fig. 4.6 exhibit the results of sensitivity analysis for criteria CMC as it is
Figure 4.6: Sensitivity analysis for CMC criterion for FMS selection problem
observed to be the most important criteria in this example with a maximum weight of
0.2861. From Fig. 4.6, it is found that when the weight of CMC is varied between 0.246
and 0.431, FMS7 remains as the best choice. On reducing its weight below 0.246 replaces
FMS4 as the best choice, while increasing the weight above 0.431 selects FMS 3 as the
best choice and beyond 0.653 FMS 5 becomes the most preferred choice.
25
4.4 Flexible Manufacturing System Selection using Grey TOPSIS
Now the same problem has been solved by GTOPSIS method. Rao and Parnichkun applied
AHP method to calculate the normalized weights of the criteria as WRLC = 0.1129, WRWP =
0.1129, WRSC = 0.0445, WIMR = 0.1129, WIQ = 0.2861, WCMC = 0.2861 and WFSU =
0.0445. The same weight is consider here. The decision matrix with 20% greyness is
exhibited in Table 4.2. As the seven considered FMS selection criteria are having different
units, it is necessary to normalize their values and obtain the normalized decision matrix.
In order to develop the normalized grey decision matrix Eqs. (3.14, 3.15) is used. The
positive-ideal and the negative-ideal solutions are determined by choosing the maximum
and minimum criteria values depending on whether the criterion is beneficial or non-
beneficial in nature and is shown in Table 4.2.
Table 4.2: Grey decision matrix for FMS selection
Alternative Criteria
FMS1 24 36 18.4 27.6 4 6 0.596 0.894 0.596 0.894 1200 1800 4000 6000
FMS2 14.4 21.6 10.4 15.6 12 18 0.596 0.894 0.596 0.894 1040 1560 4800 7200
FMS3 12 18 9.6 14.4 8 12 0.400 0.600 0.400 0.600 760 1140 5600 8400
FMS4 20 30 16 24 10.4 15.6 0.596 0.894 0.596 0.894 960 1440 3200 4800
FMS5 11.2 16.8 14.4 21.6 11.2 16.8 0.204 0.306 0.596 0.894 760 1140 2800 4200
FMS6 13.6 20.4 12 18 7.2 10.8 0.596 0.894 0.400 0.600 1000 1500 4200 6300
FMS7 18.4 27.6 14.4 21.6 16 24 0.400 0.600 0.596 0.894 880 1320 2400 3600
FMS8 12.8 19.2 6.4 9.6 11.2 16.8 0.204 0.306 0.400 0.600 1200 1800 2400 3600
The next step is calculating the separation measure of the positive and negative
ideal alternatives using Eq. (3.18, 3.19) and d+ and d- are found and finally it is used to find
the relative closeness C+. The calculated relative closeness values are shown in Table 4.4.
According to the results of Table 4.4, the priority of the alternatives are determined as
FMS7>FMS4>FMS1>FMS2>FMS5>FMS6>FMS3>FMS8. The calculated results showed
that the FMS7 is the best option for this FMS problem followed by FMS2. The worst
option was FMS8. Rao & Parnichkun [6] also obtained FMS7 as the best vehicle and
FMS8 as the worst one. From the initial decision matrix of Table 4.1, it is observed that
FMS7 and FMS2 outperforms most of the other alternative with respect to higher values of
(IQ), (RWP), (RSC), (LC) and lower value of (CMC) and (FSU). On the other hand, the
main reason behind the underperformance of FMS8 is its very low value of (IQ) criteria.
Itis also identified that this FMS8 has very high values of non-beneficial criteria such as
(CMC) and (FSU).
26
Table 4.3: Normalized values of alternatives and positive/negative ideal values for FMS selection
RLC
0.67 1.00 0.40 0.60 0.33 0.50 0.56 0.83 0.31 0.47 0.38 0.57 0.51 0.77 0.36 0.53 1.00 0.33
RWP
0.67 1.00 0.38 0.57 0.35 0.52 0.58 0.87 0.52 0.78 0.44 0.65 0.52 0.78 0.23 0.35 1.00 0.23
RSC
0.22 0.33 0.50 0.75 0.33 0.50 0.43 0.65 0.47 0.70 0.30 0.45 0.67 1.00 0.47 0.70 1.00 0.17
IMR
0.67 1.00 0.67 1.00 0.45 0.67 0.67 1.00 0.23 0.34 0.67 1.00 0.45 0.67 0.23 0.34 1.00 0.23
IQ
0.67 1.00 0.67 1.00 0.45 0.67 0.67 1.00 0.67 1.00 0.45 0.67 0.67 1.00 0.45 0.67 1.00 0.45
CMC
0.33 0.00 0.42 0.13 0.58 0.37 0.47 0.20 0.58 0.37 0.44 0.17 0.51 0.27 0.33 0.00 0.58 0.00
FSU
0.52 0.29 0.43 0.14 0.33 0.00 0.62 0.43 0.67 0.50 0.50 0.25 0.71 0.57 0.71 0.57 0.71 0.00
Table 4.4: Separation measures and the relative closeness of each alternative for FMS selection
Alternative d+ d- C+ Rank
27
Table 4.5: Weighted normalized decision matrix in ARAS method for FMS selection
Alternative Criteria
FMS1 0.017 0.026 0.016 0.025 0.002 0.003 0.015 0.002 0.033 0.049 0.023 0.034 0.004 0.006
FMS2 0.010 0.015 0.009 0.014 0.005 0.008 0.015 0.002 0.033 0.049 0.026 0.039 0.003 0.005
FMS3 0.009 0.013 0.009 0.013 0.004 0.005 0.010 0.002 0.022 0.033 0.036 0.054 0.003 0.004
FMS4 0.014 0.021 0.014 0.021 0.005 0.007 0.015 0.002 0.033 0.049 0.028 0.042 0.005 0.007
FMS5 0.008 0.012 0.013 0.019 0.005 0.008 0.005 0.001 0.033 0.049 0.036 0.054 0.005 0.008
FMS6 0.010 0.015 0.011 0.016 0.003 0.005 0.015 0.002 0.022 0.033 0.027 0.041 0.004 0.005
FMS7 0.013 0.020 0.013 0.019 0.007 0.011 0.010 0.002 0.033 0.049 0.031 0.046 0.006 0.009
FMS8 0.009 0.014 0.006 0.009 0.005 0.008 0.005 0.001 0.022 0.033 0.023 0.034 0.006 0.009
Alternative Si Ui Rank
A0 0.6400 1.0000
FMS1 0.1266 0.1979 4
FMS2 0.1171 0.1830 5
FMS3 0.1070 0.1672 6
FMS4 0.1319 0.2061 2
FMS5 0.1275 0.1992 3
FMS6 0.1041 0.1627 7
FMS7 0.1345 0.2102 1
FMS8 0.0913 0.1426 8
28
Table 4.7: Comparison table for FMS selection
Alternative Macbeth G-TOPSIS G-ARAS Rao and
FMS1 4 3 4 3
Parnichkun[6]
FMS2 5 4 5 4
(2008)
FMS3 6 7 6 7
FMS4 2 2 2 2
FMS5 3 5 3 5
FMS6 7 6 7 6
FMS7 1 1 1 1
FMS8 8 8 8 8
Using Spearmans rank correlation coefficient (rs) value, the similarity between two
sets of rankings can be measured. Usually, its value lies between 1 and +1, where the
value of +1 denotes a perfect match between two rank orderings. Table 4.8 shows the
Spearmans rank correlation coefficients when the rankings of FMS alternatives as
obtained using all the three preference ranking methods are compared between themselves
and also with respect to the rank ordering as derived by Rao and Parnichkun [6]. It is
observed that the rs value ranges from 0.9048 to 1.
Table 4.8: Comparative study on the ranking performance for FMS selection
Rao and
0.9048(1,2,#), 37.5 1.000 (1,2,3),100 0.9048(1,2,#), 37.5
Parnichkun[6]
29
alternatives considered. These results are also shown in Table 4.8. It is observed that all
these methods are quite capable to deal with both the cardinal and ordinal data, and can
provide the total ranking of the considered alternatives, although they have different
mathematical treatments and operational approaches.
References
[1] Kulak, O. and Kahraman, C. (2005) Multi-attribute comparison of advanced manufacturing systems
using fuzzy vs. crisp axiomatic design approach, International Journal of Production Economics,
Vol. 95, No. 3, pp.415424.
[2] Rao, R.V. (2006) A decision-making framework model for evaluating flexible manufacturing
systems using digraph and matrix methods, International Journal of Advanced Manufacturing
Technology, Vol. 30, Nos. 1112, pp.11011110
[3] Bhattacharya, A., Abraham, A., Vasant, P. and Grosan, C. (2007) Evolutionary artificial neural
network for selecting flexible manufacturing systems under disparate level-of-satisfaction of
decision maker, International Journal of Innovative Computing, Information and Control, Vol. 3,
No. 1, pp.131140.
[4] Liu, S-T. (2008) A fuzzy DEA/AR approach to the selection of flexible manufacturing systems,
Computers & Industrial Engineering, Vol. 54, No. 1, pp.6676.
[5] Karsak, E.E. (2008) Using data envelopment analysis for evaluating flexible manufacturing systems
in the presence of imprecise data, International Journal of Advanced Manufacturing Technology,
Vol. 35, Nos. 910, pp.867874.
[6] Rao, R.V. and Parnichkun, M. (2008) Flexible manufacturing system selection using a
combinatorial mathematics-based decision-making method, International Journal of Production
Research, Vol. 47, No. 24, pp.69816998.
[7] Chuu, S-J. (2009) Selecting the advanced manufacturing technology using fuzzy multiple attributes
group decision making with multiple fuzzy information, Computers & Industrial Engineering, Vol.
57, No. 3, pp.10331042
[8] Balaji, C.M., Gurumurthy, A. and Kodali, R. (2009) Selection of machine tool for FMS using
ELECTRE III a case study, Proc. of 5th Annual IEEE Conference on Automation Science and
Engineering, pp.2225, India.
[9] Biswas, S. and Mahapatra, S.S. (2009) An improved metaheuristic approach for solving the
machine loading problem in flexible manufacturing systems, International Journal of Services and
Operations Management, Vol. 5, No. 1, pp.7693.
[10] Rezaie, K., Nazari-Shirkouhi, S., Alem, S.M. and Hatami-Shirkouhi, L. (2010) Using data
envelopment analysis and analytical hierarchy process model to evaluate flexible manufacturing
systems, Journal of Applied Sciences Research, Vol. 6, No. 12, pp.6461 6469.
[11] Maniya, K.D. and Bhatt, M.G. (2011) The selection of flexible manufacturing system using
preference selection index method, International Journal of Industrial and Systems Engineering,
Vol. 9, No. 3, pp.330349.
[12] Dhal, P.R., Datta, S. and Mahapatra, S.S. (2011) Flexible manufacturing system selection based on
grey relation under uncertainty, International Journal of Services and Operations Management, Vol.
8, No. 4, pp.516534.
[13] Mondal, S. and Chakraborty, S. (2011) Selection of flexible manufacturing systems using data
envelopment analysis, Journal of the Institution of Engineers, Production Engineering Division,
Vol. 91, No. 18, pp.3139
[14] Todi, V., Zeljkovi, M., Tepi, J., Miloevi, M. and Luki, D. (2012) Techno-economic method
for evaluation and selection of flexible manufacturing systems (FMS), Metalurgija, Vol. 51, No. 3,
pp.349353
[15] Momeni, M., Fathi, M.R. and Khanmohammadi, E. (2012) Flexible manufacturing system
selection using of logarithmic fuzzy preference programming and ELECTRE method, New York
Science Journal, Vol. 5, No. 5, pp.9499.
30
CHAPTER
5
Material Handling Equipment selection
31
specific handling environment. Sujono and Lashkari [5] proposed a method for
simultaneously determining the operation allocation and MH system selection in a flexible
manufacturing environment with multiple performance objectives. Onut et al. [6] proposed
an integrated fuzzy analytic network process (F-ANP) and fuzzy technique for order
performance by similarity to ideal solution (F-TOPSIS) methodology for evaluating and
selecting the most suitable MH equipment types for a manufacturing organization.
Komljenovic and Kecojevic [7] applied coefficient of technical level and AHP methods for
selection of rail-mounted boom type bucket wheel reclaimers and stacker-reclaimers as
used for material handling at the stockyards. Tuzkaya et al. [8] suggested an integrated F-
ANP and fuzzy preference ranking organization method for enrichment evaluation (F-
PROMETHEE) approach for solving the MH equipment selection problems. Sawant et al.
[9] applied preference selection index (PSI) method to choose automated guided vehicle
(AGV) in a given manufacturing environment. Maniya and Bhatt [10] used AHP to assign
the relative importance between different AGV selection criteria and then applied modified
grey relational analysis (M-GRA) method to determine the corresponding index values for
AGV selection. Cho and Edbelu [11] developed a web-based system, called as
DESIGNER for the design of integrated MH systems in a manufacturing environment,
which could model and automate the MH system design process, including the selection of
MH equipment. Mirhosseyni and Webb [12] presented a hybrid method for selection and
assignment of the most appropriate MH equipment.
Table 5.1: Decision matrix for Material Handling Equipment selection problem
ALTERNATIVE CRITERIA
Traveling
Cost ($) Floor space Limitation Load capacity Life period
distance
32
This Material handling device selection problem is now solved using the free
downloaded version of M-MACBTH software. Initially, all the six evaluation criteria are
entered into the M-MACBTH software to develop the corresponding value tree, as shown
in Figure 5.1
Figure 5.1: MACBETH value tree for Material Handling equipment selection problem
In this value tree, the Beneficial criteria node contains all the beneficial attributes,
while the Non beneficial criteria node consists of all the attributes whose values is to be
diminished. The performances of alternatives are then input in the respective criterias
properties window. For Cost (C) criteria performances of the alternatives are expressed in
quantitative measures. Therefore, two reference levels, i.e. upper and lower are chosen for
this criteria. In case of beneficial criteria, the upper reference level is the highest
performance criteria value, whereas, for non-beneficial criteria, it is the lowest
performance criteria value. The lower reference level is selected 10% below the lowest
performance value in case of beneficial criteria and for non-beneficial criteria, it is chosen
10% above the highest performance value. The performance of qualitative criterion such as
(FS), (L), (LC), (TD) and (LP) are expressed using a five point scale, i.e. very high (VH),
high (H), average (A), low (L) and very low (VL). Fig. 5.2 shows the reference levels for
all the criteria as involved in this problem.
Figure 5.2: Reference levels for Material Handling equipment selection problem
Now, five alternatives (V1, V2, V3, V4, and V5) as well as performances of these
alternatives with respect to six considered criteria are entered into M-MACBETH software.
33
In order to determine the weights of the criteria with MACBETH method, criteria are
entered into M-MACBETH software in descending order of their importance from
left to right and top to bottom in the weighting matrix. The judgments of preference of
each criterion over the remaining criteria are selected using a seven point semantic scale.
Consistency checking is performed for the judgments. After considering the suggestions
for eliminating inconsistencies (if any), the judgments are made consistent and
subsequently, the criteria weights are computed. Fig. 5.3 shows the overall weighing
judgments and the weights for this example.
Figure 5.3: MACBETH weighing judgments Material Handling Equipment selection problem
The overall attractiveness scores for all the five alternatives are obtained using Eq.
(3.6) and quantitative performance scores for the alternatives with respect to six criteria are
shown in Fig. 5.4
Figure 5.4: MACBETH table of scores Material Handling device selection problem
Based on the overall attractiveness scores obtained using MACBETH method, the
ranking of Material Handling Device is obtained as V2-V5-V4-V1-V3 which shows that
V2 is the best choice. Bairagi et al. [6] also obtained the same ranking for the facility
layout alternatives using a Technique of Precise Order Preference (TPOP) method.
Sensitivity analysis is also performed to study the effects of changes in criteria
weights on the final rankings of the alternatives. The weight of a selected criterion is
changed in such a way that the weights of the remaining criteria are automatically adjusted
so that all the criteria weights always add up to one. In this example, it is observed that
LC is the most important criterion having a weight of 0.2728. The results of sensitivity
analysis with respect to LC criterion, as shown in Fig. 5.5, identifies that V2 remains as
the first choice till the weight of LC criterion is varied up to 0.2760 and above it V5
becomes the first choice.
34
Figure 5.5: Sensitivity analysis with respect to LC criterion
ALTERNATIVE CRITERIA
Cost ($) Floor space Limitation Load capacity Traveling distance Life period
V1 360 540 6 9 6 9 9 10 1 3 4 5
V2 63.6 95.4 1 3 4 5 0 1 1 3 6 9
V3 416 624 1 3 9 10 1 3 4 5 4 5
V4 301.6 452.4 1 3 4 5 4 5 1 3 4 5
V5 401.6 602.4 4 5 1 3 6 9 4 5 4 5
The criteria weights were as followed WC =0.328, WFS =0.126, WL =0.128, WLC
=0.339, WTD =0.058, WLP =0.021. Following Eqs. (3.14, 3.15) the normalized grey
decision matrix are determined. To this end, the maximum upper limit of alternatives is
35
determined, and all evaluation values are divided by the maximum value. The criteria
weight for LC is maximum and for LP it is minimum. Negative and positive ideals are
calculated using Eqs. (3.16, 3.17) and shown in Table 5.3.
Table 5.3: Normalized values of alternatives and positive/negative ideal values for Material Handling
Equipment Selection
Criteria weight V1 V2 V3 V4 V5 A+ A-
C 0.328 0.423 0.135 0.898 0.847 0.333 0.000 0.517 0.275 0.356 0.035 0.898 0.000
FS 0.126 0.333 0.000 0.889 0.667 0.889 0.667 0.889 0.667 0.556 0.444 0.889 0.000
L 0.128 0.4 0.1 0.600 0.500 0.100 0.000 0.600 0.500 0.900 0.700 0.900 0.000
LC 0.339 0.900 1.000 0.000 0.100 0.100 0.300 0.400 0.500 0.600 0.900 1.000 0.000
TD 0.058 0.200 0.600 0.200 0.600 0.800 1.000 0.200 0.600 0.800 1.000 1.000 0.200
LP 0.021 0.444 0.556 0.667 1.000 0.444 0.556 0.444 0.556 0.444 0.556 1.000 0.444
The next step is calculating the separation measure of the positive and negative ideal
alternatives using Eqs. (3.18, 3.19) d+ and d- are found and finally it is used to find the
relative closeness C+. The calculated relative closeness values are shown in Table 5.4.
According to the results of Table 5.4, the priority of the alternatives are determined as V2-
V5-V1-V4-V3.The calculated results showed that the V2 is the best vehicle for this
material handling problem followed by V5. Among the alternative V3 appeared to be the
worst vehicle. Bairagi & Sarkar also obtained V2 as the best vehicle and V3 as the worst
one. As a conclusion, vehicle 2 should be selected by the company. From the initial
decision
Table 5.4: Separation measures and the relative closeness of each alternative for Material Handling
Equipment Selection
Alternative d+ d- C+ Rank
matrix of Table 5.1, it is observed that V2 and V5 outperforms most of the other alternative
vehicle with respect to higher values of (LC) and lower value of Cost criteria. On the other
hand, the main reason behind the underperformance of V3 is its very low value of (LC)
criteria. It is also identified that this V3 has very high values of non-beneficial criteria such
as (L) and (C)
36
5.5 Material Handling Equipment Selection using Grey ARAS method
From the weighted normalized decision matrix, as given in Table 5.5, and using Eq. (3.24,
3.25), the optimality function (Si) for each of the Material handling Equipment selection is
calculated. Then the corresponding values of the utility degree (Ui) are determined for all
the alternatives.
Table 5.5: weighted normalized decision matrix for material handling equipment selection
ALTERNATIVE CRITERIA
Traveling
Cost ($) Floor space Limitation Load capacity Life period
distance
A0 0.1215 0.1538 0.0285 0.0545 0.0461 0.0658 0.1130 0.0892 0.0276 0.0200 0.0066 0.0054
V1 0.0215 0.0181 0.0048 0.0061 0.0077 0.0073 0.1017 0.0892 0.0028 0.0060 0.0026 0.0027
V2 0.1215 0.1025 0.0285 0.0182 0.0115 0.0132 0.0000 0.0089 0.0028 0.0060 0.0039 0.0048
V3 0.0186 0.0157 0.0285 0.0182 0.0051 0.0066 0.0113 0.0268 0.0110 0.0100 0.0026 0.0027
V4 0.0256 0.0216 0.0285 0.0182 0.0115 0.0132 0.0452 0.0446 0.0028 0.0060 0.0026 0.0027
V5 0.0192 0.0162 0.0071 0.0109 0.0461 0.0219 0.0678 0.0803 0.0110 0.0100 0.0026 0.0027
The values of Si and Ui, and the ranking achieved by the Material handling Equipment
selection are exhibited in Table 5.6. It is revealed from this table that V2 is the best chosen
alternative and V3 obtains the last rank. V5 has the second rank.
Table 5.6: Si and Ui values in ARAS method for material handling equipment selection
37
(d) Number of ranks matched, as the percentage of the number of considered alternatives.
Table 5.7: Comparison table for Material Handling Equipment selection
Using Spearmans rank correlation coefficient (rs) value, the similarity between two sets of
rankings can be measured. Usually, its value lies between 1 and +1, where the value of +1
denotes a perfect match between two rank orderings. Table 5.8 shows the Spearmans rank
correlation coefficients when the rankings of Material handling Equipment selection
alternatives as obtained using all the three preference ranking methods are compared
between themselves and also with respect to the rank ordering as derived by Bairagi et al.
[13]. It is observed that the rs value ranges from 0.9 to 1.
Table 5.8: Comparative study on the ranking performance for Material Handling Equipment selection
The similarity of rankings obtained by these methods is also measured using Kendalls
coefficient of concordance (z). Its value lies between 0 and 1, where a value of 1 results in
a perfect match. In this case, the value of z is computed as 0.9501, which suggests that
there is an almost perfect agreement between the considered methods. When the ranking of
Material handling Equipment alternatives as derived by Bairagi et al. [13] is taken into
consideration, the z value is observed to be 0.9658, which is quite high suggesting a
similarity of the rank orderings between those obtained by the three preference ranking
methods and that of Bairagi et al. A high z value signifies the suitability of these methods
to solve the considered Material handling Equipment selection problem.
Sometimes, the decision maker may be interested to select the best Material
handling Equipment as the single choice. So, another test is performed based on the
agreement between the top three ranked Material handling Equipment alternatives. Here, a
result of (1,2,3) means the first, second and third ranks match, (1,2,#) means the
first and second ranks match, (1,#,#) means only the first ranks match, and (#,#,#) means
no match. Table 5.8 shows the results of this test, which indicates that Macbeth and
GARAS method has the maximum number of matches with respect to the ranking of the
top three Material handling Equipment alternatives. It is also quite interesting to note that
for all the methods, the top ranked alternative is V2.The last test is performed with respect
to the number of ranks matched, expressed as the percentage of the number of alternatives
considered. These results are also shown in Table 5.8. It is observed that all these methods
38
are quite capable to deal with both the cardinal and ordinal data, and can provide the total
ranking of the considered alternatives, although they have different mathematical
treatments and operational approaches.
Reference
[1] J. A. Tompkins, Facilities Planning, John Wiley and Sons, New York, NY, USA, 2010
[2] O. Kulak, (2005) A decision support system for fuzzy multi-attribute selection of material handling
equipments, Expert Systems with Applications, vol. 29, no. 2, pp. 310319.
[3] B. M. Beamon, (2006) Performance, reliability, and performability of material handling systems,
International Journal of Production Research, vol. 36, no. 2, pp. 377393.
[4] S. Chakraborty and D. Banik, (2006) Design of a material handling equipment selection model
using analytic hierarchy process, International Journal of Advanced Manufacturing Technology,
vol. 28, no. 11-12, pp. 12371245.
[5] S. Sujono and R. S. Lashkari, (2007) A multi-objective model of operation allocation and material
handling system selection in FMS design, International Journal of Production Economics, vol. 105,
no. 1, pp. 116133.
[6] S. Onut, S. S. Kara, and S. Mert, (2009) Selecting the suitable material handling equipment in the
presence of vagueness, International Journal of Advanced Manufacturing Technology, vol. 44, no.
7-8, pp. 818828.
[7] D. Komljenovic and V. Kecojevic, (2009) Multi-attribute selection method for materials handling
equipment, International Journal of Industrial and Systems Engineering, vol. 4, no. 2, pp. 151173,
2009.
[8] G. Tuzkaya, B. Glsn, C. Kahraman, and D. zgen, (2010) An integrated fuzzy multi-criteria
decision making methodology for material handling equipment selection problem and an
application, Expert Systems with Applications, vol. 37, no. 4, pp. 28532863.
[9] V. B. Sawant, S. S. Mohite, and R. Patil, (2011) A decision-making methodology for automated
guided vehicle selection problem using a preference selection index method, Communications in
Computer and Information Science, vol. 145, pp. 176181.
[10] [K. D. Maniya and M. G. Bhatt, (2011) A multi-attribute selection of automated guided vehicle
using the AHP/M-GRA technique, International Journal of Production Research, vol. 49, pp. 6107
6124.
[11] C. Cho and P. J. Edbelu, (2005) Design of a web-based integrated material handling system for
manufacturing applications, International Journal of Production Research, vol. 43, pp. 375403.
[12] S. H. L. Mirhosseyni and P. Webb, (2009) A hybrid fuzzy knowledge-based expert system and
genetic algorithm for efficient selection and assignment of material handling equipment, Expert
Systems with Applications, vol. 36, no. 9, pp. 1187511887.
[13] B. Bairagi, B. Dey, B. Sarkar and S. K. sanyal, (2015) A De Novo multi-approaches multi-criteria
decision making technique with an application in performance evaluation of material handling
device Computer & Industrial Engineering, vol. 87 pp. 267-282.
39
40
CHAPTER
6
Green Supplier Selection
41
material, partnership with green organization, Green market share, management
commitment, adherence to environmental policies, green R&D projects, staff training and
etc. Bykzkan [11] described the framework of green supplier selection based on the
literatures about automotive industry, including g three aspects and 12 criteria. In sum,
most of the published studies have developed a framework for green supplier selection
based on previous studies and interviews, possibly because interviewing is the most
convenient way to link the reviewed studies with the real cases.
This selection problem is now solved using the free downloaded version of M-
MACBTH software. Initially, all the ten evaluation criteria are entered into the M-
MACBTH software to develop the corresponding value tree, as shown in Fig. 6.1. In this
value tree, the Beneficial criteria node contains all the beneficial attributes, while the
Non Beneficial criteria node consists of the non-beneficial attribute whose values is to be
lowered.
42
Figure 6.1: MACBETH value tree for Green supplier selection problem
43
In order to determine the weights of the criteria with MACBETH method, criteria
are entered into MMACBETH software in descending order of their importance
from left to right and top to bottom in the weighting matrix. Now, ten alternatives as
well as performances of these alternatives with respect to seven considered criteria are
entered into M-MACBETH software. The judgments of preference of each criterion over
the remaining criteria are selected using a seven point semantic scale. Consistency
checking is performed for the judgments. After considering the suggestions for eliminating
inconsistencies (if any), the judgments are made consistent and subsequently, the criteria
weights are computed. Fig. 6.3 shows the overall weighing judgments and the weight for
this example.
Figure 6.3: MACBETH weighing judgments for Green supplier selection problem
The overall attractiveness scores for all the four alternatives are obtained using
Eqn. (3.6) and quantitative performance scores for the alternatives with respect to seven
criteria are shown in Fig. 6.4.
Figure 6.4: MACBETH table of scores for Green supplier selection problem
44
Based on the overall attractiveness scores obtained using MACBETH method, the
ranking of green supplier selection is obtained as op7-op2-op3-op4-op1-op9-op8-op6-
op10-op5 which shows that op7 is the best choice and op5 as the worst choice. Yazdani et
al also obtained the same best and the worst option using COPRAS and MOORA method.
45
Table 6.2: Grey decision matrix for Green supplier selection
Opt.
max min min max max max max
direction
S1 0.054 0.082 0.053 0.079 0.120 0.180 0.078 0.118 0.125 0.187 0.091 0.137 0.078 0.118
S2 0.062 0.094 0.061 0.091 0.086 0.130 0.109 0.163 0.066 0.098 0.137 0.205 0.084 0.126
S3 0.126 0.188 0.091 0.137 0.102 0.154 0.066 0.100 0.086 0.130 0.090 0.136 0.105 0.157
S4 0.085 0.127 0.111 0.167 0.046 0.070 0.059 0.089 0.106 0.158 0.067 0.101 0.096 0.144
S5 0.082 0.124 0.150 0.224 0.100 0.150 0.141 0.211 0.059 0.089 0.051 0.077 0.046 0.068
S6 0.084 0.126 0.066 0.100 0.120 0.180 0.041 0.061 0.107 0.161 0.075 0.113 0.090 0.136
S7 0.110 0.164 0.102 0.152 0.045 0.067 0.106 0.160 0.098 0.146 0.095 0.143 0.091 0.137
S8 0.080 0.120 0.066 0.098 0.069 0.103 0.048 0.072 0.050 0.074 0.087 0.131 0.074 0.112
S9 0.042 0.064 0.042 0.062 0.034 0.052 0.080 0.120 0.040 0.060 0.062 0.094 0.050 0.076
S10 0.075 0.113 0.059 0.089 0.078 0.116 0.070 0.104 0.064 0.096 0.043 0.065 0.085 0.127
Table 6.3: Normalized values of alternatives and positive/negative ideal values for Green supplier selection
problem
46
0.582 0.547 0.751 0.504 0.521 0.464 0.58
S7
0.873 0.321 0.627 0.756 0.782 0.696 0.87
0.425 0.708 0.618 0.227 0.265 0.425 0.473
S8
0.637 0.561 0.427 0.341 0.397 0.637 0.71
0.225 0.815 0.809 0.379 0.214 0.304 0.321
S9
0.338 0.722 0.713 0.568 0.321 0.456 0.481
0.399 0.736 0.569 0.33 0.342 0.211 0.539
S10
0.599 0.604 0.353 0.494 0.513 0.316 0.809
A+ 1 0.815 0.809 1 1 1 1
A- 0.225 0 0 0.193 0.214 0.211 0.29
According to the results of Table 6.4, the priority of the alternatives are determined as S7 >
S2 > S3 > S4 > S1 > S4 > S8 > S6 > S10 > S5. The calculated results showed that the S7
as the best alternative and S5 as the worst alternative for the company. As a conclusion, S7
should be selected by the company. Yazdani et al also obtained S7 as the best supplier for
the company and S5 as the worst supplier.
Table 6.4: Separation measures and the relative closeness of each alternative for Green supplier selection
problem
Yazdani et al.
Supplier d+ d- C+ Rank
[1]
From the initial decision matrix of Table 6.1, it is observed that supplier S7 and S2
outperforms most of the other alternative suppliers with respect to higher values of QD and
RRR, and lower value of ENRC criteria. On the other hand, the main reason behind the
underperformance of S5 supplier is its very low RRR value, although it has amazingly
attractive values for QD and DS criteria. It is identified that this supplier has very less
capability in terms of energy and natural resource consumption (ENRC), delivery speed
(DS), green design (GD), re-use and recycle rate (RRR) and production planning (PP).
47
6.5 Green supplier selection using ARAS method
From the weighted normalized decision matrix, as given in Table 6.5, and using Eq. (3.24,
3.25), the optimality function (Si) for each of the Green Supplier selection alternative is
calculated. Then the corresponding values of the utility degree (Ui) are determined for all
the alternatives. The values of Si and Ui, and the ranking achieved by the Green Supplier
selection alternative are exhibited in Table 6.6. It is revealed from this table that S5 is the
best chosen alternative and S10 obtains the last rank. S2 has the second rank.
Table 6.5: Weighted normalized decision matrix for ARAS method for Green supplier selection problem
Table 6.6: Si and Ui values in ARAS method for Green supplier selection problem
Supplier Si Ui Rank
A0 0.1592 1.0000
S1 0.0979 0.6148 4
S2 0.1127 0.7083 2
S3 0.1013 0.6362 3
S4 0.0950 0.5970 5
S5 0.0831 0.5219 9
S6 0.0873 0.5488 8
S7 0.1161 0.7297 1
S8 0.0876 0.5503 7
S9 0.0947 0.5949 6
48
6.6 Ranking Performance analysis for Green supplier selection
Table 6.7 shows the overall comparison among the three method. In order to validate the
applicability and suitability of the three considered ranking methods to solve this Green
supplier selection problem, their relative ranking performance is compared using the
following measures:
(a) Spearmans rank correlation coefficient,
(b) Kendalls coefficient of concordance,
(c) Agreement between the top three ranked alternatives, and
(d) Number of ranks matched, as the percentage of the number of considered alternatives.
Yazdani et
Alternative Macbeth GTOPSIS GARAS
al. [1]
S1 5 5 4 5
S2 2 2 2 2
S3 3 3 3 4
S4 4 4 5 6
S5 10 10 9 10
S6 8 8 8 8
S7 1 1 1 1
S8 7 7 7 7
S9 6 6 6 3
S10 9 9 10 9
Using Spearmans rank correlation coefficient (rs) value, the similarity between two sets of
rankings can be measured. Usually, its value lies between 1 and +1, where the value of +1
denotes a perfect match between two rank orderings. Table 6.8 shows the Spearmans rank
Table 6.8: Comparative study on ranking performance for Green supplier selection problem
49
by Yazdani et al. [1] is taken into consideration, the z value is observed to be 0.9892,
which is quite high suggesting a similarity of the rank orderings between those obtained by
the three preference ranking methods and that of Bairagi et al. A high (z) value signifies
the suitability of these methods to solve the considered Green supplier equipment selection
problem.
Sometimes, the decision maker may be interested to select the best Green
supplier selection as the single choice. So, another test is performed based on the
agreement between the top three ranked Green suppliers Equipment. Here, a result of
(1,2,3) means the first, second and third ranks match, (1,2,#) means the first and
second ranks match, (1,#,#) means only the first ranks match, and (#,#,#) means no match.
Table 7.8 shows the results of this test, which indicates that GTOPSIS and GARAS
method has the maximum number of matches with respect to the ranking of the top three
Green supplier selection alternatives. It is also quite interesting to note that for all the
methods, the top ranked alternative is S7.The last test is performed with respect to the
number of ranks matched, expressed as the percentage of the number of alternatives
considered. These results are also shown in Table 6.8. It is observed that all these methods
are quite capable to deal with both the cardinal and ordinal data, and can provide the total
ranking of the considered alternatives, although they have different mathematical
treatments and operational approaches.
Reference
[1] M. Yazdani, P. Chatterjee, E.K. Zavadskas, S.H Zolfani, (2017) Integrated QFD-MCDM
framework for green supplier selection, Journal of Cleaner Production, vol. 142 pp. 3728-3740.
[2] K. Govindan, S. Rajendran, J. Sarkis, and P. Murugesan, (2015) Multi criteria decision making
approaches for green supplier evaluation and selection: a literature review, Journal of Cleaner
Production, vol. 98, pp. 6683.
[3] R. J. Kuo, Y. C. Wang, and F. C. Tien, (2010) Integration of artificial neural network and MADA
methods for green supplier selection, Journal of Cleaner Production, vol. 18, no. 12, pp. 1161
1170.
[4] J. Sarkis, Q. Zhu, and K.-H. Lai, (2011) An organizational theoretic review of green supply chain
management literature, International Journal of Production Economics, vol. 130, no. 1, pp. 115.
[5] H. I. Lee, H. Y. Kang, C. F. Hsu, and H. C. Hung, () 2009 A green supplier selection model for
high-tech industry, Expert Systems with Applications, vol. 36, no. 4, pp. 79177927.
[6] R. J. Kuo, Y. C. Wang, and F. C. Tien, (2010) Integration of artificial neural network and MADA
methods for green supplier selection, Journal of Cleaner Production, vol. 18, no. 12, pp. 1161
1170.
[7] M. L. Tseng and A. S. F. Chiu, (2013) Evaluating firm's green supply chain management in
linguistic preferences, Journal of Cleaner Production, vol. 40, pp. 2231.
[8] Lu, L.Y.Y., Wu, C.H., Kuo, T.C., (2007) Environmental principles applicable to green supplier
evaluation by using multi-objective decision analysis. Int. J. Prod. Res. Vol. 45, no. (18-19), pp.
4317-4331
[9] Tuzkaya, G., Ozgen, A., Ozgen, D., Tuzkaya, U., (2009) Environmental performance evaluation of
suppliers: a hybrid fuzzy multi-criteria decision approach. Int. J. Environ. Sci. Technol. 6 (3), pp.
477-490
[10] Awasthi, A., Chauhan, S.S., Goyal, S.K., (2010) A fuzzy multicriteria approach for evaluating
environmental performance of suppliers. Int. J. Prod. Econ. Vol.126 no.2, pp. 370-378
[11] Bykzkan, (2012) An integrated fuzzy multi-criteria group decision-making approach for
green supplier evaluation, International Journal of Production Research, vol. 50, no. 11, pp. 2892
2909.
50
CHAPTER
7
Rapid Prototyping Process selection
51
Methodology for the assortment of Rapid Prototyping Process. Their study defined SLS
technology as the preferred and intended one. Makhesana [8] demonstrated the
methodology of a decision-making method for selection of rapid prototyping process and
compared it with previously developed methods. The methodology of improved complex
proportional assessment (COPRAS) method was used.
3DP Very Less High Very high Very high 65 2 Less hazardous
SGC Very high Very less Very less Very less 37 62 Hazardous
The process selection problem is now solved using the free downloaded version of M-
MACBTH software. Initially, all the seven evaluation criteria are entered into the M-
MACBTH software to develop the corresponding value tree, as shown in Fig. 7.1. In this
value tree, the beneficial criteria node contains all the beneficial attributes, while the
non-beneficial node consists of the criteria whose value is to be minimized. The
performances of alternatives are then entered in the respective criterias properties window.
For SR and C criteria, performances of the alternatives are expressed in quantitative
measures. Two reference levels, i.e. upper and lower are chosen for these two criteria. In
case of beneficial criteria, the upper reference level is the highest performance criteria
value, whereas, for non-beneficial criteria, it is the lowest performance criteria value.
52
Figure 7.1: MACBETH value tree for RP process selection
The lower reference level is selected 10% below the lowest performance value in case of
beneficial criteria and for non-beneficial criteria, it is chosen 10% above the highest
performance value. The performance of qualitative criterion PV, PR, F, L, SR, and
H is expressed using a seven point semantic scale, i.e. excellent (Ex), very high (VH),
high (H), average (A), low (L), very low (VL) and none (N). These scales are entered as
performance levels with Ex identified as upper reference level while N set as lower
reference level. Fig. 7.2 shows the reference levels for all the criteria as involved in this
problem. The green box indicates upper level while the blue box indicates the lower value.
In order to quantify the qualitative performance levels, used to express the performance of
alternatives with respect to Location of available technology (L) criterion, seven semantic
scales, provided by MACBETH method, are used to represent the attractiveness of a
specific level over the other. After selecting the preferences of attractiveness, consistency
of judgments is checked. Consistent judgments lead to quantification of the performance
levels. Fig. 7.3 shows the comparison of attractiveness between the preference levels as
well as the converted MACBETH scale for L criterion. Now, six alternatives (SLA, SLS,
FDM, LOM, 3DP, SGC) as well as performances of these alternatives with respect to
seven considered criteria are entered into M-MACBETH software. MACBETH method
computes criteria weights by pair-wise comparison procedure. Initially, all the criteria are
arranged according to their decreasing importance, i.e. PV, PR, F, L, SR, C and
H from left to right and top to bottom in a matrix.
53
Figure 7.3: Comparison of attractiveness between reference levels for L criterion.
The judgments of preference of each criterion over the remaining criteria are selected using
a seven point semantic scale. Consistency checking is performed for the judgments. After
considering the suggestion for eliminating inconsistencies (if any), the judgments are made
consistent and subsequently, the criteria weights are computed
Figure 7.4 shows the overall weighing judgments. The overall attractiveness scores for all
the four alternatives are obtained and quantitative performance scores for the alternatives
with respect to seven criteria are shown in Figure 7.5.
Based on the overall attractiveness scores obtained using MACBETH method, the ranking
of RP process selection is obtained as FDM-SLS-3DP-SLA-LOM-SGC which shows that
FDM is the best choice.
54
Figure 7.6: Sensitivity analysis with respect to PV criterion
Alternatives PV PR F L S C D
Optimization
max max max max min min min
direction
SLA 0.5 0.6 0.9 1 0.1 0.2 0.3 0.4 0.9 0.8 0.3 0.2 0.125 0
SLS 0.7 0.8 0.5 0.6 0.3 0.4 0.7 0.8 0.9 0.8 0.5 0.4 0.625 0.5
FDM 0.7 0.8 0.5 0.6 0.5 0.6 0.7 0.8 0.9 0.8 0.1 0 0.625 0.5
LOM 0.3 0.4 0.3 0.4 0.7 0.8 0.5 0.6 0.5 0.4 0.9 0.8 0.125 0
3DP 0.1 0.2 0.7 0.8 0.9 1 0.9 1 0.1 0 0.9 0.8 0.625 0.5
SGC 0.9 1 0.1 0.2 0.1 0.2 0.1 0.2 0.5 0.4 0.1 0 0.125 0
As the seven considered RP process selection criteria are having different units, it is
necessary to normalize their values using and obtain the normalized decision matrix.
Following Eqs. (3.14, 3.15) the normalized grey values are determined. To this end, the
55
maximum upper limit of alternatives is determined, and all evaluation values are divided
by the maximum value. The criteria weight for (PV) is maximum and for (D) it is
minimum. Negative and positive ideals are calculated using Eqs. (3.16, 3.17) and shown in
Table 7.3.
Table 7.3: Normalized values of alternatives and positive/negative ideal values for RP process selection
PV 0.2553 0.5 0.6 0.7 0.8 0.7 0.8 0.3 0.4 0.1 0.2 0.9 1 1 0.1
PR 0.2383 0.9 1 0.5 0.6 0.5 0.6 0.3 0.4 0.7 0.8 0.1 0.2 1 0.1
F 0.1872 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.1 0.2 1 0.1
L 0.1489 0.3 0.4 0.7 0.8 0.7 0.8 0.5 0.6 0.9 1 0.1 0.2 1 0.1
S 0.1021 0.9 0.8 0.9 0.8 0.9 0.8 0.5 0.4 0.1 0 0.5 0.4 0 0.9
C 0.0511 0.3 0.2 0.5 0.4 0.1 0 0.9 0.8 0.9 0.8 0.1 0 0 0.9
D 0.0171 0.125 0 0.625 0.5 0.625 0.5 0.125 0 0.625 0.5 0.125 0 0 0.625
The next step is calculating the separation measure of the positive and negative ideal
alternatives. Using Eqs. (3.18, 3.19) and, d+ and d- are found. Finally the calculated d+ and
d- are used to find the relative closeness C+. The calculated relative closeness values are
shown in Table 7.4.
Table 7.4: Separation measures and the relative closeness of each alternative for RP process selection
Alternative d+ d- C+ Rank
SLA 0.5745 0.5132 0.4719 4
SLS 0.4946 0.4939 0.4996 3
FDM 0.4396 0.5447 0.5534 1
LOM 0.5573 0.4109 0.4244 6
3DP 0.4953 0.6086 0.5510 2
SGC 0.6626 0.5014 0.4308 5
According to the results of Table 7.4, the priority of the alternatives are determined as
FDM-SLS-3DP-SLA-LOM-SGC.The calculated results showed that the FDM is the best
option for this RP selection problem followed by SLS which shows that FDM is the best
choice followed by SLS. The worst option was SGC. From the initial decision matrix of
Table 7.1, it is observed that FDM and SLS outperforms most of the other alternative with
respect to higher values of (PV) and (L). On the other hand, the main reason behind the
underperformance of SGC is its very low value of (PR), (F) and (L) criteria. It is also
identified that this SGC has very high values of non-beneficial criteria such as (C) and (D).
7.5 Rapid prototyping Process Selection using Grey ARAS method
From the weighted normalized decision matrix, as given in Table 7.5, and using Eq. (3.24,
3.25), the optimality function (Si) for each of the RP Process is calculated. Then the
corresponding values of the utility degree (Ui) are determined for all the alternatives.
56
Table 7.5: Weighted normalized decision matrix in ARAS method for RP process selection
Opt.
max max max max min min min
direction
A0 0.061 0.053 0.060 0.052 0.052 0.045 0.035 0.031 0.023 0.035 0.014 0.020 0.008 0.009
SLA 0.030 0.032 0.054 0.052 0.005 0.009 0.011 0.012 0.023 0.017 0.002 0.003 0.001 0.001
SLS 0.043 0.043 0.030 0.031 0.016 0.018 0.025 0.025 0.023 0.017 0.003 0.003 0.002 0.002
FDM 0.043 0.043 0.030 0.031 0.026 0.027 0.025 0.025 0.023 0.017 0.002 0.002 0.002 0.002
LOM 0.018 0.021 0.018 0.021 0.036 0.036 0.018 0.019 0.004 0.006 0.014 0.010 0.001 0.001
3DP 0.006 0.011 0.042 0.041 0.047 0.045 0.032 0.031 0.002 0.003 0.014 0.010 0.002 0.002
SGC 0.055 0.053 0.006 0.010 0.005 0.009 0.004 0.006 0.004 0.006 0.002 0.002 0.001 0.001
The values of Si and Ui, and the ranking achieved by the RP Process are exhibited in Table
7.6. It is revealed from this table that FDM is the best chosen alternative and SGC obtains
the last rank. 3DP has the second rank.
Table7.6: Si and Ui values in ARAS method for RP process selection
Alternatives Si Ui Rank Macbeth
A0 0.249 1
SLA 0.126 0.507 4 4
SLS 0.140 0.563 3 3
FDM 0.148 0.595 1 1
LOM 0.111 0.446 5 5
3DP 0.144 0.579 2 2
SGC 0.082 0.330 6 6
57
Table 7.7: Comparison table for Rapid prototyping process selection
SLA 4 4 4
SLS 3 3 3
FDM 1 1 1
LOM 5 6 5
3DP 2 2 2
SGC 6 5 6
Using Spearmans rank correlation coefficient (rs) value, the similarity between two sets of
rankings can be measured. Usually, its value lies between 1 and +1, where the value of +1
denotes a perfect match between two rank orderings. Table 7.8 shows the Spearmans rank
correlation coefficients when the rankings of Rapid Prototyping process selection
alternatives as obtained using all the three preference ranking methods are compared
between themselves. It is observed that the rs value ranges between 0.9429 to 1.
Table 7.8: Comparative study on ranking performance for RP process selection
The similarity of rankings obtained by these methods is also measured using Kendalls
coefficient of concordance (z). Its value lies between 0 and 1, where a value of 1 results in
a perfect match. In this case, the value of z is computed as 0.9746, which suggests that
there is an almost perfect agreement between the considered methods. A high z value
signifies the suitability of these methods to solve the considered Rapid Prototyping process
selection problem.
Sometimes, the decision maker may be interested to select the best Rapid
Prototyping process as the single choice. So, another test is performed based on the
agreement between the top three ranked FMS alternatives. Here, a result of (1,2,3) means
the first, second and third ranks match, (1,2,#) means the first and second ranks
match, (1,#,#) means only the first ranks match, and (#,#,#) means no match. Table 4.8
shows the results of this test, which indicates that Macbeth, and GARAS method has the
maximum number of matches with respect to the ranking of the top three Rapid
Prototyping process alternatives. It is also quite interesting to note that for all the methods,
the top ranked alternative is FMS7.The last test is performed with respect to the number of
ranks matched, expressed as the percentage of the number of alternatives considered.
These results are also shown in Table 7.8. It is observed that all these methods are quite
capable to deal with both the cardinal and ordinal data, and can provide the total ranking of
the considered alternatives, although they have different mathematical treatments and
operational approaches.
58
References
[1] Byun, H.S., Lee, K.S. (2004) A decision support system for the selection of a rapid prototyping
process using the modified TOPSIS method, International Journal of Advance Manufacturing
Technology, 26, pp. 13381347.
[2] Rao, R. V., & Padmanabhan, K. K. (2007). Rapid prototyping process selection using graph theory
and matrix approach. Journal of Materials Processing Technology, vol. 193 no. (1-3), pp. 81-88.
[3] Rao, R.V. and Patel, B.K. (2009) Decision making in the manufacturing environment using an
improved PROMETHEE method, International Journal of Production Research, vol.48, pp.4665
4682.
[4] Chakraborty, S. (2011)Applications of the MOORA method for decision making in manufacturing
environment, International Journal of Advance Manufacturing Technology, vol. 54, pp.11551166
[5] V. Shende, P. Kulkarni, (2014) Decision Support System for Rapid Prototyping Process
Selection, International Journal of Scientific and Research Publications, Volume 4, no.1. pp. 452-
461
[6] B. N. Panda, B. B. Biswal, B. B. L. V. Deepak, Integrated AHP and fuzzy TOPSIS Approach for
the Selection of a Rapid Prototyping Process under Multi-Criteria Perspective, 5th International &
26th All India Manufacturing Technology, Design and Research Conference (AIMTDR 2014)
December 12th14th, 2014, IIT Guwahati, Assam, India.
[7] M. Shahrabi, M. Javadi, (2014) Selection of Rapid Prototyping Process Using Combined AHP and
TOPSIS Methodology, International Journal of Information Science and System, vol.3 Issue 1, ,
pp.15-22
[8] M. A. Makhesana, (2015) "Application of improved complex proportional assessment (COPRAS)
method for rapid prototyping system selection," Rapid Prototyping Journal, Vol. 21, Issue 6, pp. 671
674.
59
60
CHAPTER
8
Conclusion and future scope
8.1 Conclusion
Different Multi criteria decision making methods have already been proposed by the past
researchers to address the issue of Advance Manufacturing Systems evaluation and
selection, it is still not clear which MCDM method is the best for a given AMS selection
problem. Researchers are challenged to provide guidance for choosing the method that is
both theoretically well-founded and practically operational to solve AMS selection
problems.
The present work considers three potential ranking methods (MACBETH,
GTOPSIS, GARAS) and compares their ranking performance while considering four real
time AMS selection problems having both qualitative and quantitative attributes. Four
performance tests are conducted for this ranking performance comparison and also for
measuring the degree of agreement between the rankings derived by the considered
methods. The uniqueness of the proposed methodology is that it offers a general procedure
that can be applied to a diverse selection of problems encountered in the manufacturing
environment that incorporate vagueness and a number of selection attributes. The
methodology is logical, simple, and convenient to implement when compared with the
other MCDM methods.
In all the cases, it is observed that the top-ranked alternatives exactly match with
those derived by the past researchers. There are slight discrepancies between the
intermediate rankings of the alternatives which may be attributed due to the normalization
method taken. Different aggregation procedures and different normalization procedures
leads to the selection of different most acceptable alternatives. At the same time, it is also
seen that different relative weights of criteria, used in the decision making model, have a
significant impact on the selection of most appropriate alternatives, as well as ranking
orders.
61
8.2 Future scope
In future studies, the number of criteria and alternatives may be changed for the same
selection problem. The weights and normalization technique of the criteria may be derived
from different methods. The ranking of the alternatives may be performed with other
MCDM methods and the obtained results may be compared. Application of this method
could be used for conducting similar studies in several sectors and in a wider range of
selection problems in some other real-time manufacturing environment remains as a future
research scope of the present work.
62
List of publications
1. Bablu Kumar Mandal and Saikat Ranjan Maity A decision making Approach for
selection of Rapid Prototyping process using MACBETH method Proceedings of
International Conference on Evaluations in Manufacturing: Technologies and
Business Strategies for Global Competitiveness, BITS Mesra Ranchi, 12th-13th Nov.
2016, 173-178.
63