See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/274254762

AHP-IRP: An Integrated Approach for Decision Making

Conference Paper · March 2015

CITATIONS READS

2 172

2 authors:

Ravi Kumar Surya Prakash Singh

NTPC School of Business Indian Institute of Technology Delhi
10 PUBLICATIONS   18 CITATIONS    121 PUBLICATIONS   1,013 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Create new project "Facility Layout, Decision making, mathematical modelling, cellular manufacturing" View project

Decision making View project

All content following this page was uploaded by Surya Prakash Singh on 17 August 2016.

The user has requested enhancement of the downloaded file.

 ISBN: 978-93-84935-18-4

AHP-IRP: An Integrated Approach

for Decision Making
Ravi Kumar% S.P. Singh2
1'2Department of Management Studies, Indian Institute of Technology Delhi, India
1 ravikumar.iitdu@
gmaiLcom, 2surya.singh@gmaiLcorn

Abstract: This paper proposes an integrated approach for Multi- practice PLD problem is a multi-objective problem and
Criteria Decision Making (MCDM) process using Analytic layout designer should also consider qualitative factors such
Hierarchy Process (AHP) and Interpretive Ranking Process as flexibility, maintenance, and robustness by taking experts'
(IRP). The proposed approach is a novel attempt to integrate AHP
opinion.
and IRP, where weights of criteria computed from AHP is
supplied to IRP. Thus the integrated approach provides a new In the proposed integrated AHP-IRP approach, weights of
dimension for a decision making process, which can be used such
as ranking of alternatives. The working methodology of the both qualitative and quantitative criteria are computed by
proposed integrated approach is demonstrated through an AHP. After this, the weight matrix of AHP is used to create a
illustrative example based on Plant Layout Design (PLD) dominance matrix for IRP where net dominance is calculated
problem. In addition, the comparative analysis of rankings by taking the difference of total dominating and dominated
generated from IRP-AHP, IRP, and AHP is also carried out. factors for all alternatives. The final ranking of alternatives is
Lastly a sensitivity analysis is conducted on IRP-AHP. It is found given on the basis of ascending order of dominating factors.
that IRP AHP, IRP, and AHP provide the almost same ranking in In this approach equal importance to all criteria is given. In
top two positions. this paper weighted AHP-IRP has been also introduced
where weights of criteria computed from AHP is used to
Keywords: AHP, IRP, PLD, Sensitivity Analysis calculate the weighed dominance matrix. Then, weighted net
dominance factor is calculated by taking the difference of
1. INTRODUCTION total dominating and dominated factors for all alternatives. In
Over the last several years, multi-criteria decision making addition, a comparison is also shown and sensitivity analysis
approaches have gain popularity in decision making in is carried out on best three alternatives to show the influence
different application areas. Among all multi-criteria decision of criteria on the layouts. This paper is organized as follows,
making (MCDM) approaches Analytic Hierarchy Process literature review is provided in section 2. Section 3 briefly
(AHP) (Saaty, 1980) is the most popular approach for discusses AHP, and provides overview on IRP, and section 4
decision making (Pohekar and Ramachandran, 2004). This explains proposed integrated AHP-IRP methodology.
paper proposes an integrated AHP-IRP approach for multi- Empirical illustration is given in section 5; section 6 shows
criteria decision making. The proposed approach also sensitivity analysis and conclusion.
analyzes its applicability on Plant Layout Design (PLD)
2. LITERATURE REVIEW
problems. In the integrated approach weights of criteria are
computed by AHP which is further supplied as inputs for Decision making is a branch of operations research, deals in
Interpretive Ranking Process (IRP). IRP (Sushil, 2009) is a the situations where multiple decision criteria taking parts in
decision making approach which uses interpretive matrix as decision making environment. These decision techniques are
basic tool, and also takes care of limitations of intuitive broadly called as multi-criteria decision making (MCDM),
process and rational process. The IRP process is based on ch is further divided into two sub categories, multi-
equal importance of criteria, however unequal importance of attribute decision making (MADM) and multi-objective
criteria can also be considered, and these weights can be decision making (MODM). For detailed study on multi
found out by using decision making approach. PLD problems criteria analysis refer the work by Climaco (1997) and Gal
are concerned with dividing a planar area into different work and Hanne (1999). In all decision making problems, due to
cells or facilities, so that material handling cost of conflicting criteria and incomparable units of criteria, it is
manufacturing can be reduced. Till now, many approaches hard to rank the given alternatives. There are number of
have been used to solve PLD problems such as procedural, decision making techniques and also huge literature is
approximated, algorithmic, and optimization approaches. In available, however this paper only focus on the literature of
PLD problems, much of the work has been done by multi-criteria decision making techniques on plant layout
considering quantitative objective such as material handling design problems. There is a large applicability of this multi-
cost, adjacency score, and shape ratio. However, in real criteria decision making techniques such as supply chain

International Conference on Evidence Based Management 2015 (ICEBM2015) 605

 AHP-IRP: An Integrated Approach for Decision Making

management (Zaeri, et al., 2011) manufacturing systems considered by this decision making technique. Sushil (2009)
(Putrus, 1990) etcetera . Yang and Kuo (2003) proposed a proposed a novel interpretive ranking process (IRP), a
hierarchical AHP/ DEA methodology for plant layout design decision making technique. In this work, IRP is used in
problems. They used a computer aided tool to generate integration with AHP.
considerable alternatives and quantitative criteria, also
qualitative criteria are weighted by AHP, and then to 3. AHP AND IRP METHODOLOGIES
compare and rank the different decision making units 3.1 AHP Methodology
(DMUs), they used DEA. Yang and Hung (2007) also
applied TOPSIS and fuzzy TOPSIS methodology on the MCDM techniques are most common to rank the different
same problem used in Yang and Kuo (2003), and they found alternatives based on some set of conflicting criteria. AHP is
these methodologies are viable approach in solving plant a most commonly used MCDM technique, which converted
layout design problems. An integrated AHP-NLP technique the problem in a hierarchical structure, and then rank the
was proposed by Vencheh and Mohamadghasemi (2013)7 alternatives to find the best alternative among them based on
Also Altuntas et al. (2014) proposed a fuzzy DEMA'IL the experts' judgments. The hierarchy of general AHP model
based solution approach for facility layout problems, they is shown below in Fig. 1.
explained quantitative factors that are linguistic can also be

Objective

Criteria m
Criteria 1 Criteria 2

Alternative 1 Alternative 2 Alternative 3 Alternative n

Fig. 1. general hierarchy AHP

In general hierarchy of AHP, the first level is to set an 1. The consistency ratio is calculated for each decision
objective, i.e. to choose a best alternative among available matrix by using the formula given below in equation (1)
alternatives. Level 2 is to define the criteria. In next level,
alternatives are mentioned. Next step is to develop pair-wise CR = ((a._ ax - n)/((n — 1) ))/R1 Eq (1)
comparison matrix by taking experts' opinion for each level.
Further, from the comparison matrix, priority vector is RI is random index used to check the inconsistency of a
compute, by finding normalized Eigen vector of the decision matrix, and its value is determined by size of the
comparison matrix. This priority vector shows relative matrix, n is the size of matrix. Consistency ratio should be
weights. Also, consistency index has been calculated. Saaty less than or equal to 10% for a decision matrix to be
1-9 scale is used for pair-wise comparison is given in Table consistent. For detailed study of AHP refer to Saaty (1990).

TABLE!: Scale of relative importance

Intensity of Importance 1 3 5 7 9 2, 4, 6, 8
Definition Equal Moderate Strong Very Strong Extreme Compromise

3.2 IRP For detailed study and explanation of steps, paper by Sushil
(2009) can be referred.
In any MCDM technique, the main aim is to rank the
alternatives depending on the experts' opinion. In many of 4. PROPOSED INTEGRATED AHP-IRP
the cases, experts' opinion is intuitive and rational. Both
4.1 AHP-IRP Approach
intuitive and rational process has limitations (Sushil, 2009).
Intuitive process does not include all possible interactions. Step!. Identification of Variables: IRP, first define two sets
of variables, one set consists of ranking variables

International Conference on Evidence Based Management 2015 (ICEBM2015) 606

 AHP-IRP: An Integrated Approach for Decision Making

(alternatives) which are to be ranked, and other set consists used for further analysis. This decision matrix is shown in
of reference variables (criteria) on which alternatives are to table 3. Based on the given decision matrix, a dominating
be ranked matrix is developed by pair-wise comparison of alternatives
for each criterion. To develop this pair-wise comparison
Step2. Identification of interactions of ranking variables matrix, if weight of an alternative for a particular criterion is
(alternatives) with reference variables (criteria)
higher than other alternative, then this alternative is
Step3. Pair-wise comparison of ranking variables dominating the other alternative. This pair-wise comparison
(alternatives), with respect to different reference variables matrix for net dominance is given in table 4.
(criteria) to identify the dominating interactions The next stage is to develop a matrix of dominating
Step4. Making a dominance matrix by finding net dominance interactions by giving each dominating criteria value 1, and
for each ranking variable (alternatives) considering equal adding all the dominating criteria value for each pair of
importance/weight of all reference variables alternatives. For example, alternative A2 is dominating over
alternative Al for criteria C2, C3, C4 and C5, i.e. A2 is
Step5. Rank the ranking variables (alternatives) dominating over Al by a dominating factor of 4. Each row of
this matrix represent the dominance of corresponding
4.2 Weighted IRP alternative over other alternatives, Sum of this row will give
total dominance of corresponding alternative over other
In step4, instead of finding net dominance by considering
equal importance to all criteria, this work used weights of alternatives, and column sum will give how many factors are
dominating the corresponding alternative. We can get an
these criteria, found by using analytic hierarchy process as
explained in section 3. overall dominating factor by subtracting column sum from
row sum. This calculation is given in the table 5.
5. EMPIRICAL ILLUSTRATION
TABLE 2 Available floor space (Yang and Hung, 2007)
This paper uses the same problem of IC packaging company 99.25 m (width) * 27.00 m (length)
used by Yang and Kuo (2003), Company wants to determine
present layout is efficient or not for future expansion. For No.1 Department name Size (M.
analysis, Yang & Kuo (2003) divide the IC packaging
process into 10 departments. The table 2 defines total 1 Wafer sawing 89.21
available area of the plant, available area of each department, ° 2 Die bond 181.51
and name of each department. They consider 18 alternatives 3 Wire bond 577.38
for evaluation, 17 were generated by using a commercial
software SPIRAL (Goetschalckx, 1999) and 18th layout was 4 Molding 599.57
the original layout. These layouts were evaluated on six 5 Dejunk/trimming and curing 183.71
criteria namely distance, adjacency, shape ratio, flexibility, 6 Electro deflash/solder platting 500.13
accessibility, and maintenance. In this paper, the criteria used
7 Marking 199.94
in Yang and Kuo (2003) of the same problem of IC
packaging company are used. Out of this six criteria, 8 Forming and singulation 186.40
distance, adjacency, and shape ration are quantitative criteria, 9 Lead scanning/inspection 110.78
and flexibility, accessibility, and maintenance are qualitative
criteria that is Cl, C2, C3, C4, C5 and C6 respectively. In 10 Packaging 51.09
this paper, decision matrix given in Yang and Hung (2007) is
TABLE 3 Decision matrix (Yang and Hung, 2007)
Distance Adjacency Shape ratio Flexibility Accessibility Maintenance
Al 185.95 8 8.28 0.0119 0.026 0.069
A2 207.37 9 3.75 0.0595 0.026 0.0575
A3 206.38 8 7.85 0.0714 0.0519 0.0345
A4 189.66 8 8.28 0.0714 0.0779 0.046
A5 211.46 8 _ 7.71 0.0714 0.039 0.046
A6 264.07 5 2.07 0.0357 0.0519 0.069
A7 228 8 14 0.0476 0.039 0.023
A8 185.59 9 6.25 0.0476 0.013 0.0575
A9 185.85 9 7.85 0.0357 0.026 0.0575
A10 236.15 8 7.85 0.0595 0.0779 0.069
All 183.18 8 2 0.0952 0.1169 0.092

International Conference on Evidence Based Management 2015 (ICEBM2015) 607

 AHP-IRP: An Integrated Approach for Decision Making

Al2 204.18 8 13.3 0.0357 0.039 0.0575

A13 225.26 8 8.14 0.0714 0.039 0.0345
A14 202.82 8 8 0.0357 0.0779 0.0575
Al5 170.14 9 8.28 _ 0.0952 0.1169 0.092
A16 216.38 9 7.71 0.0476 0.0519 0.069
A17 179.8 8 10.3 0.0476 0.0779 0.0345
A18 185.75 10 10.16 0.0595 0.0519 0.0345

TABLE 4: Pair-wise comparison matrix

Al A2 A3 A4 A5 A6 A7 A8
Al - Cl, C6 Cl, C6 Cl, C6 Cl, C6 Cl, C2 Cl, C3, C6 C5, C6
C2, C3, C4,Cl, C2, C3, Cl,C2, C3, C4,
A2 - C2, C3, C6 C2, C3, C6 Cl, C2, C4
C5 C6 C3, C4, C5
C6
A3 C3, C4, C5 Cl, C4, C5 - C3 , ---„ CI, C5 Cl,C3, C4, C5,
Cl, C2, C4 C4, C5
----&-------
- C6
A4 C4, C5 Cl, C4, C5 Cl, C5, C6 Cl, C2, C4, Cl, C3, C4, C5,
Cl, C5 C4, C5
CS C6
A5 C3, C4, C5 C5 C3, C6 C3 - Cl, C2, C4 Cl, C3, C4, C6 C4, C5
A6 C3, C4, C5 C3, C5, C6 C3, C6 C3, C6 C3, C5, C6 - C3, C5, C6 C3, C5, C6
A7 C4, C5 C5 - - Cl, C2, C4 - C5
Cl, C2, C3, Cl, C2, C3, Cl, C2, C3, Cl,C2, C3,
A8 Cl Cl, C2, C4 CI, C2, C3, C6
C4 C6 C6 C6 -
Cl, C2, C3, Cl,C2, C3,
A9 Cl Cl, C2, C6 Cl, C2, C6 Cl, C2 Cl, C2, C3, C6
C4 C6 C5
A10 C3, C4, C5 C5, C6 C5, c6 C ' C2, C4,
C3, C6 C5, C6 C1 C3, C4, C5, C6 C4, C5, C6
C5
Cl, C3, C4, Cl, C3, C4, Cl, C3, C4, Cl, C3, C4, Cl, C3, C4, Cl, C2, C3, Cl, C3, C4, C5, Cl, C3, C4,
All
C5, C6 C5, C6 C5, C6 C5, C6 C5, C6 ' C4, C5, C6 C6 CS, C6
Al2 C4, C5 Cl, C5 Cl, C6 C6 Cl, C6 C6 Cl, C6 Cl, C2
A13 C3, C4, C5 C4, C5 - C3 - Cl, C2, C4 Cl, C3, C4, C6 C4, C5
A14 C3, C4, C5 Cl, C5 Cl, C5, C6 C3, C6 Cl, C5, C6 Cl, C2, C5 CI, C5, C6 C5
Cl, C2, C4, Cl, C4, C5, Cl, C2, C4, Cl, C2, C4, Cl, C2, C4, Cl, C2, C4, Cl, C2, C3, C4, CI, C4, C5,
A15
C5, C6 C6 C5, C6 C5, C6 C5, C6 C5, C6 C5, C6 C6
C2, C3, C4, Cl, C2, C3, C5,
A16 CS C5, C6 C2, C3, C6 Ci., C3, C6 C2, C5, C6 Cl, C2, C4 C5, C6
C6
A17 Cl, C4, C5 Cl, C5 Cl, C5 Cl, C2, C4,
Cl CI, C5 Cl, C3, C5, C6 Cl, C5
C5
Cl, C2, C4, Cl, C2, C3, C4,
Alb Cl, C2, C5 Cl, C2 Cl, C2 Cl, C2, C4 C l,C2, C4 C2, C4, C5
C5 C5, C6

A9 A10 All Al2 A13 A14 A15 A16 A17 A18

Al C6 C1 - Cl, C3, C6 Cl, C6 Cl, C6 - Cl C6 C6
A2 C3, C4 Cl, C2, C3 C2 Cl, C2, C3, CI, C3 ,
C2, C3, C4 C2, C3, C4 C3 C2, C3, C4, C6 C3, C6
C6 C4 _
A3 C4, C5 C1, C4 C3, C4, C5 Cl, C3, C5 C3, C4 C3 CI, C4 C3, C4 C3, C4
A4 C4, C5 CI, C4 Cl, C3, C4, C5 CI, C5, C6 CI, C4 CI, C4, C3, C4, C5,
C5 C3, C4, C6
C6 _
C3, C4,
A5 CI, C3, C4 C3, C4 CI, C3, C6 C3, C4 C3 Cl, C4 C3, C4, C6 C3, C4, C6
C5
C3, C5,
A6 C3 - C3, C5, C6 C3, C5, C6 C3, C6 C3 C3 C3, C6 C3, C6
C6
_
A7 C4, C5 C1 - C4 - C4 - - -
A8 Cl, C3, Cl, C2, C3, CI, C2, C3, Cl, C3,
C4 CI, C2, C3 C2 Cl, C2, C3, C4 C3 C2, C3, C6 Cl, C3, C6
C6 C4 C6
A9 - Cl, C2 C2 Cl, C2, C3 ,
Cl, C2, C3 Cl, C2, C3 C3 Cl C2, C3, C6 C3, C6
C6
, C5,
A10 - - C3, C4, C5, C6 C3, C5, C6 C3, C4, C6 C3 C4, C5 C3, C4, C6 C3, C5, C6
C4 C6

International Conference on Evidence Based Management 2015 (ICEBM2015) 608

 AHP-IRP: An Integrated Approach for Decision Making

A9 A10 All Al2 A13 A14 A15 A16 A17 A18

Cl, C3, Cl, C3, C4,
C1, C3, C3, C4, C5,
Cl, C3, C4, Cl, C3, C4,
All C4, C5, CI, C3, C4, C5, C6 C3 C4, C5, C3, C4, C5, C6
C5, C6 C5, C6 C5, C6 C6
C6 C6
Al2 C5 C1 - - C1, C6 Cl - C6
A13 C4, C5 CI, C4 C3, C4 C4 C3 C4 C3, C4 C3, C4
Cl, C3, C5,
A14 C5 Cl - Cl, C5 - C3 CI, C5 C3, C6 C3, C5, C6
C6
C4, C5, Cl, C2, Cl, Cl, C2, C3, Cl, C2, C4, C1, C2, C4, Cl, C4, Cl, C2, C3, Cl, C3, C4,
A15 C4, C5, C6 C2 C4, C5, C6 C5, C6 C5, C6 C5, C6 C4, C5, C6 C5, C6
C6
C3, C4, C2, C3, C4, Cl,C2, C3, C2, C3, C4,
Al6 Cl, C2, C3 C2 C3 C2, C3, C6 C3, C6
C5, C6 C5, C6 C5, C6 C6
Cl C5C4
, ,
A17 Cl Cl CI, C3, C4, C5 Cl, C5 Cl, C4 - Cl, C5 - CI, C5

CI, C2, Cl, C2, C3, Cl, C2,

A18 Cl, C2 C2 Cl, C2, C5 Cl, C2, C4 C2 C2, C3, C4 -
C4, C5 C4, C5 C4

TABLE 5 Net dominance matrix

A A A A A A A A A Al Al Al Al Al Al Al Al Al
P P-D Rank
1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8
Al - 2 2 2 2 2 3 2 1 1 - 3 2 2 - 1 1 1 27 -30 15
A2 4 - 3 3 4 3 5 3 2 3 1 3 4 3 1 3 4 2 51 12 5
A3 3 3 - 1 2 3 5 2 2 2 - 3 3 2 1 2 2 2 38 -5 10
A4 2 3 3 - 2 4 5 2 2 2 - 4 3 2 - 3 3 4 44 5 8
A5 3 1 2 1 - 3 4 2 3 3 - 2 3 2 1 2 3 3 38 -7 11
A6 3 3 2 2 3 - 3 3 3 1 3 3 2 1 1 2 2 37 -18 13
A
A7 2 1 - - - 3 - 1 2 1 1 - 1 - •- - - 12 -60 17
A8 4 1 4 4 4 3 4 - 3 3 1 4 4 4 1 3 3 3 53 13 4
A9 4 1 3 4 3 2 4 1 - 2 1 3 4 3 1 1 3 2 42 -2 9
A10 3 2 , 2 2 2 4 4 3 3 - 4 3 3 1 2 3 3 44 6 7
All" 5 5 5 5 5 6 5 5 5 5 1 5 5 5 1 5 4 4 81 72 1
Al2 2 2 2 1 2 1 2 2 1 1 - - 2 - - 1 - 1 20 -39 16
A13 3 2 - 1 - 3 4 2 2 2 - 2 - 1 1 1 2 2 28 -27 14
A14 3 2 3 2 3 3 3 1 1 1 2 4 - 1 2 2 3 36 -8 12
A15 5 4 5 5 5 5 6 4 3 5 2 6 5 5 - 4 6 5 80 68 2
A16 4 2 3 3 3 3 5 2 4 3 1 5 5 4 1 - 3 2 53 17 3
A17 3 2 2 1 2 4 4 2 3 1 1 4 2 2 - 2 - 2 37 -7 11
A18 4 3 2 2 3 3 6 3 4 2 1 5 3 3 1 3 3 - 51 10

D 38 9 59 55 44 12 36 44 41 772
7 9 3 9 5 5 2 0 4

Alternative All is having the largest dominating factor the created by using Saaty's scale of relative importance given in
is 72, so it is ranked as number 1 alternative, followed by Table 1, with the help of experts of the related area. Further
alternative 15, and similarly, ranked other alternatives, the AHP analysis is done to find out the weights of each criterion
descending order is given below. among themselves. The table 6 shows weights, and
A 1 1>A1 5>A16>A8>A2>A18>A10>A4>A9>A3>AS>A17> consistency ration which is below 0.1.
A14>A6>A13>A1>Al2>A7
TABLE 6: Final weights of criteria
The above ranking is done by considering equal importance Cl 0.136 Maximum Eigen Value =
to all six criteria, but in practical which may not be true. So, C2 0.313 6.25364
this work also proposed a ranking with considering unequal C3 0.050
weights of criteria. The weights of criteria are finding out by Consistency Index = 0.05
C4 0.179
taking experts' opinion and using approach of analytic C5 0.256 weighted
Further, a
hierarchy process (AHP). A pair-wise comparison matrix is C6 0.066 dominance matrix is created

International Conference on Evidence Based Management 2015 (ICEBM2015) 609

 AHP-I RP: An Integrated Approach for Decision Making

by considering the weights of criteria from table 6, instead of sum. The ranking for weighted AHP-IRP from the table 7 is
giving equal weight of 1 to all criteria. The table 7, below given as
shows the weighted dominance matrix, and ranks the A 1 5>A11>A18>A16>A5>A2>A17>A8>A10>A3>A9>A14
alternatives by taking difference of row sum and column >A5>A13>Al2>A1>A7>A6.
TABLE 7: Net dominance matrix for AHP-IRP
Al A2 A3 A4 A5 A6 A7 A8
Al - 0.202 0.202 0.202 0.202 0.449 0.251 0.322
A2 0.798 - 0.429 0.429 0.565 0.629 0.744 0.485 _
A3 0.485 0.571 - 0.050 0.392 0.629 0.687 0.435
A4 0.435 0.571 0.458 _
- 0.392 0.885 0.687 0.435
A5 0.485 0.256 0.115 _
0.050 - 0.629 0.430 0.435
A6 0.485 0.371 0.115 _
0.115 0.371 - 0.371 0.371
A7 0.435 0.256 - - - 0.629 - 0.256
A8 0.678 0.136 0.565 S0.565 0.565 0.629 0.565 -
A9 0.678 0.136 0.515 \ 0.565 0.515 0.449 0.565 0.256
A10 0.485 0.322 0.322 - 0.115 0.322 0.885 0.551 0.501
All 0.687 0.687 0.687 0.687 . 0.687 1.000 0.687 0.687
Al2 0.435 0.392 0.202 0.066 0.202 0.066 0.202 0.449
Al3 0.485 0.435 - 0.050 - 0.629 0.430 0.435
A14 0.485 0.392 0.458 0.115 0.458 0.706 0.458 0.256
Al5 0.950 0.637 0.950 0.950 0.950 0.950 1.000 0.637
A16 0.798 0.322 0.429 0.429 0.635 0.629 0.821 0.322
Al7 0.571 0.392 0.392 0.136 0.392 0.885 0.507 0.392
Al8 0.885 0.706 0.449 0.449 0.629 0.629 1.000 0.749
D 10.262 6.784 6.286 4.971 7.276 11.304 9.954 7.424

A9 A10 All Al2 Al3 Al4 Al5 Al6 Al7 A18 P P-D Rank
Al 0.066 0.136 - 0.251 0.202 0.202 - 0.136 0.066 0.066 2.952 -7.30998 16
A2 0.229 0.499 0.313 0.542 0.565 0.542 0.050 0.365 0.608 0.115 7.906 1.121598 6
A3 0.435 0.315 - 0.485 0.442 0.229 0.050 0.315 0.229 0.229 5.977 -0.30912 10
A4 0.435 0.315 - 0.621 0.458 0.315 - 0.571 0.294 0.551 7.424 2.453373 5
A5 0.485 0.365 - 0.229 0.251 0.229 0.050 0.315 0.294 0.294 4.913 -2.36293 13
A6 0.371 0.050 - 0.371 0.371 0.115 0.050 0.050 0.115 0.115 3.809 -7.49516 18
A7 0.435 0.136 - 0.179 - 0.179 -- - - 2.506 -7.44836 17
A8 0.365 0.499 0.313 0.678 0.565 0.678 0.050 0.251 0.429 0.251 7.781 0.356852 8
A9 - 0.449 0.313 0.499 0.565 0.499 0.050 0.136 0.429 0.115 6.735 -0.72958 11
A10 0.501 - - 0.551 0.371 0.294 0.050 0.435 0.294 0.371 6.369 0.295886 9
All 0.687 0.687 0.687 0.687 0.687 0.050 0.687 0.551 0.551 11.076 8.923325 2
Al2 0.256 0.136 - - 0.202 - - 0.136 0.066 2.808 -6.32524 15
Al3 0.435 0.315 - 0.229 - 0.179 0.050 0.179 0.229 0.229 4.309 -3.74408 14
A14 0.256 0.136 - 0.392 0.507 - 0.050 0.392 0.115 0.371 5.547 -1.10364 12 _
A15 0.501 0.950 0.449 1.000 0.950 0.950 - 0.637 1.000 0.687 14.151 13.29157 1
Al6 0.551 0.499 0.313 0.864 0.821 0.608 0.050 - 0.429 0.115 8.633 3.006139 4 _
A17 0.571 0.136 0.136 0.621 0.392 0.315 - 0.392 - 0.392 6.624 1.000875 7 _
Al8 0.885 0.449 0.313 0.934 0.706 0.629 0.313 0.629 0.542 - 10.896 6.378485 3 _
D 7.464 6.073 2.153 9.134 8.053 6.651 0.859 5.627 5.623 4.518 0 _

International Conference on Evidence Based Management 2015 (ICEBM2015) 610

 AHP-IRP: An Integrated Approach for Decision Making

6.1 Sensitivity Analysis sensitivity analysis, it can be predict that A 1 1 is best

alternative. However if All is compare with A15, All is not
In this section, sensitivity analysis is done to validate and as good as All for shape ratio and distance criteria, but it is
verify the results. In decision making, criteria weights good for adjacency criterion. A15 can be a best alternative if
represent the real importance of criteria in different units. adjacency is not a primary concern. From Fig. 2, it can be
According to Traintaphyllou and Sanchez (1997), it is seen that adjacency is the most important criteria, followed
difficult to represent accurate importance of these criteria. In by accessibility in plant layout design problems. The
this work Expert choice (1990), an AHP software package is important of these criteria may vary according to the nature
used for sensitivity analysis. After results of decision making of the facility layout problem. A comparative table of results
technique, we found All, A15 and A18 are best three from different methodologies on the same problem is shown
alternatives, and sensitivity analysis is done on these three in table 8.
alternatives only. From the given graph in Fig. 3 of

TABLE 8: Comparison of proposed methodology with existing methodologies

Ranking
1 2 3

AHP-IRP All A15 Al6

Weighted AHP-IRP A15 All Al 8

Methodology TOPSIS All A15 A10

Fuzzy TOPSIS All Al5 Al8

YK- Model* All A15 A18

-1-

• In YK-model all three alternatives are of same rank

File Options Tools Windo

13.6% DISTANCE

31.3% ADJACENCY

5.0% SHAPE RATIO

17.9% FLEXIBILITY

25.6% ACCESSIBILITY

6.6% MAINTENANCE

. 1 . 1 1.I. I • I • I • I
.5 .6 .7 .8 .9 1 .1 .2 3 .4 .5 .6 .7
.1 .2 .3 .4

ideal Mode .,
Sensitivity vo.r. 1:
2P• ' "Ce.5.44461,

Fig. 2. Weights of criteria using AHP

International Conference on Evidence Based Management 2015 (ICEBM2015) 611

 AHP-IRP: An Integrated Approach for Decision Making

File Options Tools Window

ObjZ 7
.80

.9
.70
.8
.60
.7
- 50
.6 )-

.5
- 40
.4 )_
30
.3 -
20
.2
Al 8
.1 10

.0 .00
ADJACENCY FLEXIBILITY MAINTENANCE
DISTANCE SHAPE RATIO ACCESSIBILIT OVERALL
Sensitivity w.r.t.: (Ideal Mode As

Fig. 3. Sensitivity curve

6. CONCLUSION [6] Gal T, Hanne T, editors. Multicriteria decision making:
Advances in MCDM models, algorjthms, theory, and
This paper proposed an integrated AHP-IRP decision making applications. New York: Kluwer Academic Publishers; 1999.
technique, and the methodology of this proposed approkh is [7] Yang T, Kuo C. A hierarchical AHP/DEA methodology for the
verify and validated by taking an illustrative example of facilities layout design problem. Eur J of Oper Res 2003; 147:
plant layout design problem. Also, this paper explains how 128-136.
critical the role of criteria in any decision making techniques, [8]. Mohamadghasemi A, Hadi-Vencheh A. An integrated AHP-
and there is always a need to do post optimality checks to NLP methodology for facility layout design. Journal of
validate the importance of these criteria. In this paper to Manufacturing Systems' 2013; 32: 40- 45.
check the effect of weights of criteria on top three [9] Altuntas S, Selim H, Dereli T. A fuzzy DEMATEL-based
alternatives, a detailed sensitivity analysis is done. solution approach for facility layout problem: a case study. Int
J Adv Manuf Technol 2014; 73:749-771.
REFERENCES [10] Spiral user's manual 1999. Mark Goetschalckx, Marietta,
[1] Saaty TL. The Analytic Hierarchy Process. McGraw-Hill, New Gorgia.
York 1980. [11] Traintaphyllou E, Sanchez A. A Sensitivity Analysis Approach
[2] Pohekar SD, Ramachandran M. Application of multi-criteria for Some Deterministic Multi-Criteria Decision-Making
decision making to sustainable energy planning- A Review. Methods. Decision Sciences 1997; 28: 151-194.
Renewable and sustainable Energy Reviews 2004; 8: 365-381. [12] Expert Choice [Computer Software]. Pittsburgh, PA: Expert
[3] Sushil. Interpretive Ranking Process. Global Journal of Choice Inc 1990.
Flexible Systems Management 2009;10:1-10. [13] Zaeri MS et al.. Application of multi criteria decision making
[4] Yang T, Hung CC. Multiple-attribute decision making technique to evaluation suppliers in supply chain management.
methods for plant layout design problem. Robotics and African J. of Mathematics and Computer Science Research
Computer-Integrated Manufacturing 2007; 23:126-137. 2011; 4 (3):100-106.
[5] Climaco J, editor. Multicriteria analysis. New York: Springer- [14] Putrus P. Accounting for intangibles in integrated
Verlag; 1997. manufacturing-non-financial justification based on analytical
hierarchy process. Information Strategy 1990; 6:25-30.

International Conference on Evidence Based Management 2015 (ICEBM2015) 612

View publication stats