1 s2.0 S0305048315001334 Main

Omega 57 (2015) 521
Contents lists available at ScienceDirect
Omega
journal homepage: www.elsevier.com/locate/omega
A multiple objective optimization based QFD approach for efcient

resilient strategies to mitigate supply chain
vulnerabilities: The case of garment industry of Bangladesh,$
Md. Maruf Hossan Chowdhury, Mohammed A. Quaddus n
School of Marketing, Curtin University, Perth, WA, Australia
art ic l e i nf o
a b s t r a c t
Article history:
Received 1 January 2014
Accepted 22 May 2015
Available online 25 June 2015
With the upsurge of frequent disruptive events, organizations have become more vulnerable to the
consequences of these disruptive events. As a result, the need for more resilient supply chain (SC) to
mitigate the vulnerabilities has become paramount. Supply chain resilience (SCR) has been discussed in
the literature and resilience index has been developed, but developing and selecting a portfolio of supply
chain resilience capabilities in order to mitigate the vulnerabilities have not been studied. In this
research we develop a 01 multi-objective optimization model based on QFD methodology. Our multiobjective method is interactive and interacts with the decision makers to choose the most satisfactory
efcient portfolio of supply chain resilience strategies. We apply our methodology to three large readymade garment (RMG) companies of Bangladesh. Results show that lack of materials (high dependence
on imported materials), disruptions in utility supply, increased competition (and hence competitive
pressure), impact of economic recession, and reputation loss are the top most vulnerabilities of
Bangladesh RMG industry. The most preferred resilience strategies to mitigate the vulnerabilities are:
back-up capacity, building relation with buyers and suppliers, quality control, skill and efciency
development, ICT adoption, demand forecasting, responsiveness to customers, and security system
improvement. Theoretical and managerial implications of our study are included.
& 2015 Elsevier Ltd. All rights reserved.
Keywords:
Supply chain
Resilience-efciency
Vulnerability
Multi-objective model
AHP
QFD
Introduction
Ready Made Garment (RMG) industry contributes hugely to
Bangladeshs economy. It creates more than four million direct
employment and several millions of indirect employment and accounts for 78.6 percent of countries export earnings [5]. RMG sector also
immensely contributes in reducing the high rate of women unemployment in the country as 80 percent of the garments workers are
women [5]. Thanks to the RMG sector, Bangladesh is also the second
largest apparel exporter in the world.
Despite its huge potentials the industry is struggling with numerous Supply Chain (SC) disruptions [49,42]. The consequences of the
disruptions are huge, for example, RMG industry of Bangladesh loses
$26.15 million per day due to problems in SC functions caused by
political instability [1]. Moreover, the preferential access in U.S. market
is cancelled because of the poor safety standard in production plants
as building collapse in garment factory caused the death of more than
It is an equally authored paper. Authors names are listed in alphabetical order.
This manuscript was processed by Associate Editor B. Lev.

Corresponding author.
E-mail addresses: marufhossan@gmail.com (Md.M.H. Chowdhury),
m.quaddus@curtin.edu.au (M.A. Quaddus).
n
http://dx.doi.org/10.1016/j.omega.2015.05.016
0305-0483/& 2015 Elsevier Ltd. All rights reserved.
eleven hundred workers [37]. These disruptions have chain effect to

all the members in SC network including the international buyers
(retail chains) and suppliers. In the wake of such a critical state in
RMG supply chain, developing resilience capabilities is vital, which is the
primary objective of this study.
Resilience has been dened by a number of authors in a related
manner. Vugrin et al. [99] dene system resilience and resilience in
general. The authors highlight that resilience is the ability of a systems
to respond to a disruption due to an event or set of events. Along the
same vein Christopher and Peck [24], Ponomarov and Holcomb [81]
and Jttner and Maklan [53] dene supply chain resilience as the
capability of the supply chain to responds to disruptions and recover
from them. On the other hand Pettit et al. [79,80] developed a supply
chain resilience framework by identifying seven categories of vulnerabilities and creating supply chain capabilities along 14 areas (sourcing, order fullment, capacity development; among others). The
authors surmise that current level of vulnerabilities and capabilities
must be assessed in order to ascertain the current level of resilience.
Literature emphasizes that developing resilience capability is vital for
organizations. It enables organizations to improve system performance [80,99], achieve sustainable competitive advantage [81], gain
market share in competitive environments [90], and decreases
vulnerabilities [53,79,80]. However current literature lacks in
Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521
proposing the ways and means to achieve supply chain resilience

capabilities. In this research we stress that resilience capabilities of RMG
supply chain of Bangladesh must be developed to mitigate organizations
vulnerabilities. In pursuing this research objective we introduce the
concept of supply chain resilience efciency (SCREF) which has
signicant pedagogical importance.
It has been established that supply chain resilience capabilities has
multiple dimensions (objectives). For example Pettit et al. [79,80] in
their framework highlight 14 areas of supply chain resilience capabilities to be developed from order fullment, capacity development
to nancial strengths. From multiple objectives decision making
(MODM) perspectives the supply chain resilience capabilities must
be efcient to mitigate vulnerabilities. While literature on efciency
approach in MODM and its applications in production/operations
management area are plentiful (for example see, [63]; among many
others), the notion of supply chain resilience efciency (SCREF) from
multiple objectives perspective is novel.
In this research we dene SCREF as follows:
present study is highly relevant to the supply chain aspect of

enterprise risk management. There are applications of enterprise risk
management (ERM) in supply chain. For example, Wu and Olson [105]
developed a DEA based value at risk (VaR) model to manage supply
risks, specically vendor selection problem. Olson and Wu [75]
presented a review of ERM in supply chain. Jiang et al. [50] developed
a LOGIT model of job satisfaction to reduce supply chain risks.
However, it has been mentioned before that we developed a new
approach to nd an optimal portfolio of efcient resilience capabilities
to mitigate supply chain vulnerabilities.
In the next several sections we present the state of affairs of
RMG industry of Bangladesh, relevant literature, methodology
(QFD based mixed qualitativequantitative approach), application
in RMG industry, followed by the results. The paper concludes
with the discussions & implications and conclusions.
(i) resilience capability must be resource efcient (e.g. minimum

cost of implementation), and
(ii) portfolio of chosen resilience capabilities must be efcient (or
non-dominated) from multiple objectives perspective [63].
Bangladesh is one of the leading exporters of Ready Made

Garments (RMG) in the world. RMG industry is an economic propeller
of Bangladesh and apparel exports stood-up at 19.90 billion US dollar
in 2011 and marked Bangladesh as the second largest apparel exporter
in the world [5]. Because of enormous economic importance of RMG
in the economy of Bangladesh, smooth and efcient functioning of
supply chain activities is crucial. But, the RMG supply chain is facing a
climax situation owing to numerous challenges, such as, labour unrest
for violation of human rights, poor wages, poor and hazardous
working environment, political instability, interruption in utility
supply especially power shortage, inefciency in customs and port
management, exchange rate uctuation, disruption in timely supply of
fabrics and other accessories, increased competition, inefciency in
operations, intensive competitive pressure, strict compliance code
regarding social and environmental issues; among many others
[49,22,42]. Furthermore, increased lead time and cost due to disruptions in procurement and shipment of goods, lack of linkages and coordination among related industries in the value chain, dependence
on imported inputs, limited variety of nished products [42], fall of
order because of global economic downturn are also issues of high
concern for the RMG supply chain of Bangladesh. As a result of these
disruptions the growth of RMG export from Bangladesh has fallen
from 23% in 20052006 to 15% in 2008-2009 [22]. In such a situation
it is crucial to nd ways and means to make RMG supply chain
resilient and sustainable. Previous researchers focused mainly on RMG
competitiveness, the existing problems and challenges of the industry.
Table 1 summarizes these briey. However, the issue of making RMG
supply chain resilient and efcient has not been investigated yet. This
study aims to ll this gap in the literature.
We shall elaborate on (i) and (ii) later. It is observed that a number

of logistics and SC related capabilities are discussed in the literature
(for example; [80,38,81,90,24]; among many others) to develop SC
resilience but most of those are conceptual studies and fall short of
introducing the notion of resilience efciency. Furthermore, in a state
of uncertainty, dynamic changes and resource limitation, selection of
optimal and efcient portfolio of resilience capabilities has not yet
been addressed adequately in the existing SC literature.
While a number of approaches could be undertaken to achieve the
research objective, this study has adopted Quality Function Deployment (QFD) [77,102] as a methodology to develop the resilience
capabilities of the RMG supply chain of Bangladesh and nd the
optimal efcient portfolio of the resilience capabilities using a nonlinear 01 programming approach. Literature on QFD approach is
plentiful, which will be reviewed briey in a later section. It is sufce
to say that QFD enables organizations to be proactive to vulnerabilities
mitigation rather than reactive and it is a proven technique for
designing supply chain mitigation capabilities in such situations
[34]. However we shall use Analytical Hierarchy Process (AHP) [86]
within QFD for the analysis of data. It is important to note that
methodologically our contribution lies as follows: we dene supply
chain resilience efciency (SCREF) and nd portfolio of efcient
resilience capabilities for implementation using multiple objectives
based non-linear 01 mathematical program.
It is noted that overall domain of our present study is Enterprise
Risk management (ERM). Enterprise Risk management (ERM) has
been dened in many different ways. However, one common theme
of ERM is that it takes a holistic and strategic approach to manage all
risks that an organization faces [32,74]. A recent literature review [20]
has found that Desheng Wu and David Olson are two of the most
dominant contributors on ERM and various aspects of enterprise risk.
One of their highly cited works is the application of ERM to assess
credit worthiness in bank [106]. Wu and Olson [104], Wu et al.
[107,108] have edited special issues of various journals on various
aspects of ERM, ranging from risk methods and tools in operations,
enterprise risk management in operations and business intelligence in
risk management. Various other applications of ERM are available
elsewhere [20] and hence will not be repeated here.
Our present research focusses on resilience capabilities of RMG
supply chain of Bangladesh to mitigate supply chain vulnerabilities.
Literature suggests that vulnerability is an exposure to serious
disturbance arising from risks within supply chain [78]. Hence our
Background
Literature review
Supply chain vulnerabilities
Maintaining an effective Supply Chain (SC) has become challenging
and difcult as the supply chains are inherently complex and in recent
times are overwhelmed with disruptive events. These disruptive
events make a supply chain vulnerable, as supply chain vulnerability
is the susceptibility of the supply chains to the consequences of
disruptive events [8,53]. Wagner and Neshat [101] posit that supply
chain vulnerability is determined by the vulnerability drivers arising
from demand side, supply side and supply chain design issues.
Similarly, supply chain vulnerability may also arise from a number
of factors such as, delay during transportation, port stoppages,
frequent occurrence of natural disasters, weak communication, supply
shortages, demand volatility, quality problem, operational issues and
Table 1
Vulnerability factors and the mitigation capabilities in RMG industry of Bangladesh.
Khondker et al. [60]
Deals with important issues and challenges facing the RMG industry during the post-MFA regime. It also discusses the competitiveness issue in
relation to productivity, working environment and stakeholders.
Choudhury and Hossain The authors have taken step to dene the challenges and opportunities in the post-MFA period in the RMG sector and indicated the
[21]
government response to combat post-MFA challenges.
Karim [55]
Discusses how the Bangladeshi RMG sector coped startegically with the post-MFA challenges.
Tewari [97]
Emphasized on timely supply, short lead times, low inventories, innovation and the ability to contribute to design and full package supply.
Ferdousi and Ahmed
Investigated the improvement of manufacturing performance through lean practice that helps to reduce lead time and cost and to improve
[36]
productivity.
Hossan et al. [47]
Discuss the recent unrest in the RMG industry which indicates political action in this industry.
Islam et al. [48]
Operational disturbances, manufacturers are facing competition with respect to quality, cost and time to market.
Nuruzzaman et al. [73] Described how to create competitive advantage through SCM.
Chowdhury et al. [23]
Addressed supply chain disruptions and mitigation processes
Dealt with upstream supply chain barriers and mitigation processes
Haider [42]
Presents the challenges and the surface-level and deep-level competitiveness of the Bangladeshi RMG industry.
terrorism; among many others [25,61,7]. Schmitt et al. [88] study the
impact of supply disruptions on both centralized and the decentralized multi-location system. Sawik [87] presents a bi-objective (minimum cost and maximum service level) model to study the impact of
supplier disruptions Mizgier et al. [71] show the far-reaching effect of
disruptions in a SC network and the impact on the performance of
overall system. The studies by Hendricks and Singhal [45] show that
announcement of SC disruptions, such as, operational issue or delay in
shipment cause decrease in shareholder value signicantly. Kleindorfer and Saad [61] identify three main sources of SC vulnerability:
rstly, operational factors which include equipment malfunctions and
systemic failures, abrupt discontinuity of supply, labour strikes, among
others; secondly, natural hazards which include earthquakes, hurricanes, storms; and thirdly, terrorism or political instability, among
others. Blos et al. [8] suggest four major sources of SC vulnerabilities
such as, nancial vulnerability, strategic vulnerability, hazard vulnerability, and operations vulnerability. Similarly, a number of researchers
(such as [79,24,90,7,61]; among others) discussed SC vulnerability
factors which are summarized in Table 2 in terms of various vulnerability factors. However, these studies did not deal with strategies and
capabilities to mitigate SC vulnerabilities. Furthermore, most of the
studies are conceptual rather than empirical in nature. The research
reported in this paper prioritizes the existing vulnerabilities and nds
the efcient mitigation strategies and capabilities in the context of
RMG supply chain of Bangladesh by using AHP integrated QFD
approach. It thus addresses a specic gap in the existing literature.
Supply chain resilience capability
Resilience is a multidisciplinary concept. Holling [46] was one
of the pioneers to conceptualize resilience as the ability of system
to absorb changes. Since then many authors echoed the concept of
resilience as systems ability to recover and get back to the original
state [70,81,24]. Heckmann et al. [44] however mention that
supply chain resilience must have the ability to overcome supply
chain vulnerability and to reduce supply chain risk. In line with
extant literature [81,44], in this paper, we dene supply chain
resilience (SCR) as the capability of a supply chain to reduce the
impact of vulnerabilities (due to disruptions) through developing
required level of readiness, quick response and recovery ability.
Vulnerabilities in the SC are sometimes beyond the direct control
of SC managers. However, SC managers need to be proactive to
predict the vulnerability factors in advance and develop resilience
capacity for mitigating the vulnerabilities [53]. Otherwise, the
consequence will be the discontinuity of SC operations which will
adversely affect both revenue and cost of the whole chain [81].
Researchers in supply chain management (SCM) emphasized on
capabilities, such as, adaptability, pro-activeness, diversity, exibility,
efciency, reserve capacity, integration, market development,
cohesion, control, connectedness to measure resilience [80,81,98].

Table 3 summarizes these capabilities. It is noted that two fundamental organizational resilience capabilities are buffering and bridging
[9,35]. All others are related to these two fundamental strategies.
Buffering is external to a current relationship with a supply chain
partner and acts as shock absorbers to mitigate the detrimental
consequences [9]. For example the capability factors of exibility,
reserve/backup capacity of Table 3 fall in this category. On the other
hand bridging is internal to a current relationship and refers to
strengthening the current relationship via formal structure [9]. The
capability factors of integration, efciency of Table 3 fall in this
category. The resilience capability needed by a system depends on
context, extent and type of vulnerabilities [16]. Therefore, in order to
deal with resilience it is important to identify the vulnerability factors
of the specic SC and the corresponding mitigation capabilities.
Along with developing resilience it is also important to measure
resilience to ensure a better resilience outcome. In the literature,
resilience is measured in a number of ways: (i) based on the extent of
systems departure from desired state [46,68], (ii) based on recovery
time after disaster or disruptions [92,90], (iii) based on reduction of
impact and consequences [85,67], (iv) based on time to respond [90]
and (v) based on cost of recovery [69,99].
Once the resilience capabilities are designed and measured, it is
also important to determine the efciency of the resilience capability
for reducing the impact of vulnerabilities [99]. Literature lacks signicantly in conceptualizing resilience efciency. In this paper, therefore, we offer a unique and elegant operationalization of resilience
efciency in the methodology section.
Quality function deployment (QFD)
QFD is a systematic process used by cross-functional teams to
identify and resolve the issues involved in providing products,
processes, services, and strategies that enhance customer satisfaction
[41]. The benets of QFD model have been highlighted by various
researchers. For example Chan and Wu [17] in their review of QFD
theory and applications have noted wide range successful applications
of QFD from product development, customer needs analyses to
decision making. Carnevalli and Miguel [15] in another review of
QFD highlight QFDs ability in adapting into various research methods
from modelling, theoretical-conceptual to action-research, experimental. Because of its wide applicability QFD has been used in various
elds, such as, determining customer needs [94], developing priorities
[43], manufacturing strategies [29,52], logistics and SCM [10,4]. QFD
has also been applied successfully for supply chain risk identication
and mitigation [82,34]. In line with previous literature we have
adapted QFD methodology to identify supply chain vulnerabilities
and mitigate the vulnerabilities with optimal and efcient resilience
capabilities.
Table 2
Supply chain vulnerability factors.
Vulnerability factors
Specifc vulnerabilities
References
Hazard vulnerability
Natural disaster (ood, cyclone)
Christopher and Peck [24]; Shef and Rice [90]; Kleindorfer and Saad [61]; Wu et al.
[109]; Blackhurst et al. [7]
Kleindorfer and Saad [61]; Wu et al. [109]; Blackhurst et al. [7]; Blos et al. [8]
Blos et al. [8]
Shef and Rice [90]; Kleindorfer and Saad [62]; Wu et al. [109]; Blackhurst et al. [7]
Political instability
Fire and other accidental damage
Labour unrest
Strategic vulnerability
Increased competition
Non-compliance of social and environmental
factors
Problem of relation with buyer (switching of buyer)
Problem of integration and real-time information
Problem of relation with supplier
Plant location problem
Haider [42]; Blos et al. [8]

Islam and Deegan [49]
Financial vulnerability
Currency uctuation
Economic recession
Raw material price uctuation
High bank interest & fund shortage
Bankruptcy or credit default of any supply chain
member
Blos
Blos
Blos
Blos
Blos
Operational
vulnerability
Shortage of skilled worker

Switching and absenteeism of workers
Fault in production planning and inventory
management
Failure of IT system and machineries
Disruption in utility supply
Product quality defection (poor quality)
Illiteracy of workers and supervisors
Haider [42]
Chowdhury et al. [22].
Chowdhury et al. [22]; Wu et al. [109]
Blos et al. [8]
Blos et al. [8]
Blos et al. [8]
Infrastructure
vulnerability
Delay in custom clearance

Delay for Congestion and inefciency in port
Strike by port workers
Delay in transportation for Poor infrastructure
Colicchia [25]
Colicchia [25]; Blackhurst et al. [7]
Colicchia [25]; Blos et al. [8]
Blackhurst et al. [7]
Demand & supply

vulnerability
Suppliers delay
Dependence on imported material and lack of
backward linkage
Lack of alternative for some critical items
Defection or nonconformity of material
Opportunism of buyers (expect discount)
Demand uctuation/uncertainaty
Suppliers opportunism

Haider [42]; Nuruzzaman and Raq [73]; Craighead et al. [26]
Blos et al. [8]

Gaudenzi and Borghesi [39]
Blos et al. [8]
Field study
m
X
i1
wi Rij
8 j ; j 1; ; n
al.
al.
al.
al.
al.

[8]
[8]
[8]
Craighead et al. [26]

Ponomarov and Holcomb [81]
Wu et al. [109]
Ponomarov and Holcomb [81]
In QFD modelling, customer requirements or existing problems of

the organizations (for example vulnerabilities) are referred to as
WHATs and how to full the customers requirements or organizational problems are referred to as HOWs (resilience capabilities). The
basic QFD framework is shown in Fig. 1 where CRi and DRj are the
HOWs and WHATS respectively. The process of using appropriate
HOWs to meet the given WHATs is represented in the relationship
matrix (Rij in Fig. 1). Different researchers build different QFD models
involving various elements but the most widely used QFD model
contains at least the requirements/problems (WHATs) and their
relative importance (Wi in Fig. 1), technical measures or design
requirements (HOWs) and their relationships with the WHATs, and
the correlation between the HOWs (see Fig. 1).
In our case row elements CRi (WHATs) of Fig. 1 represent the
vulnerabilities that RMG supply chains are facing currently. The
column elements DRj (HOWs) are the resilience strategies or capabilities to mitigate the vulnerabilities. We dene the elements of the
relationship matrix Rij (see Fig. 1) as the extent of mitigating the
specic vulnerabilities by specic resiliency strategies. In line with
QFD literature [17,82,34] we measure Rij using the scale of 9 (strong
mitigation), 3 (moderate mitigation), 1 (little mitigation) and 0 (no
mitigation). The AI and RI in Fig. 1 are the absolute and relative
importances of the HOWs (resilience strategies) which are found as
follows [77]:
AI j
et
et
et
et
et
where AIj is the absolute importance of jth design requirement (DR)

(or, resilience strategy), wi is the weight of the ith supply chain
vulnerability. Rij is the relationship value; extent of mitigating ith
vulnerability by jth resilience strategy (9, 3, 1, or 0), n is the number of
design requirements (resilience strategies); m is the number of supply
chain vulnerabilities.
It is noted that in our case AIj is interpreted as total resilience
of the jth resilience strategy to mitigate the vulnerabilities.
The relative importance (resilience) of the resilience strategy j
is
AIj
RIj Pn
j1
AI j
The correlation between the DRjs (HOWs) (see Fig. 1) plays a

signicant part in many QFD applications including ours. It represents
the extent of correlation (similarities) when two HOWs are implemented. Literature suggests that there is some degree of dependencies
among the HOWs in real applications [103,77]. If HOWi and HOWj are
correlated then there is cost savings sij in their implementation [77].
These sijs need to be estimated from the decision makers. We argue
that in most business, management and social science applications
(including our present case) some HOWs will be highly correlated.
Integration with Analytical Hierarchy Process (AHP)

In basic QFD three fundamental data are needed: (i) relative
importance of WHATs, (ii) the relationships between the WHATs
Table 3
Supply chain resilience capabilities.
Capability Factors
Specic capabilities
References
Flexibility
Flexibility in production (different volume of order, exible production schedule)
Duclos et al. [33]; Braunscheidel and

Suresh [11]
Braunscheidel and Suresh [11]
Duclos et al. [33]
Duclos et al. [33]
Ability to modify a wide varity of product as per buyer requirement (mix exibility)
Flexibility in contract with SC partners (Partial order and payment, partial shipment)
Efcient and effective logistics and supply chain functions (e.g. sourcing, producing,
distribution)
Ability to respond to additional order or sudden demand
Jttner and Maklan [53]
Ability to supply new and different products to different customer groups (mix exibility) Braunscheidel and Suresh [11]
Reserve/backup capacity
Alternative and reserve capacity (logistical options)

Buffer stock
Backup energy source
Integration
Sharing information with supply chain partners
Efciency
Waste elimination (efcient use of resource)
Pettit et al. [79]

Pettit et al. [79]
Pettit et al. [79]
Braunscheidel and Suresh [11]; Blackhurst

et al. [7]
Communication and information ow with different departments (e.g. supply chain and Braunscheidel and Suresh [11]
other departments)
Joint or collaborative planning (e.g product development)
Communication with supply chain partners
ICT supported planning and integration
Narasimhan and Kim [72]
Pettit et al. [79]; Fiksel [38]; Shef and
Rice [90]
Efcient and hardworking employees

Quality control and less defection
Pettit et al. [79]; Kleindorfer and Saad [61]
Customer satisfaction & market

position
Buyer satisfaction (product quality and service)

Customer loyalty and Preference
Good relation with buyers and suppliers
Pettit et al. [79]

Pettit et al. [79]
Zsidisin et al. [111]; Pettit et al. [79]
Financial strength
Fund availability
Protability
Insurance
Pettit et al. [79]; Tang [95]

Pettit et al. [79]
Pettit et al. [79]
and HOWs, and (iii) the correlation between the HOWs. In nding the
relative importance of WHATs (i.e.wis; see Fig. 1) we plan to use AHP.
AHP was originally developed by Saaty [86] which is an wellestablished multi-criteria decision making approach that employs a
unique method of hierarchical structuring of a problem and subsequent ranking of alternative solutions by a paired comparison
technique. For brevity full description of AHP process will not be
presented in this paper, which is available elsewhere in the literature
[86]. AHP is frequently used in QFD process, for instance see, Kamvysi
et al. [54], Park and Kim [77], Bhattacharya et al. [6], Chan and Wu
[18]; among others. Methodologically QFD has been frequently
combined with other tools to increase its robustness and applicability.
For example Ramanathan and Yunfeng [84] combined QFD with Data
Envelopment Analysis (DEA) and applied it to design security fasteners in a Chinese company. Lin et al. [66] combined QFD with DEA and
AHP and applied it to evaluate the economic performance of local
governments in China.
QFD optimization
It is noted that in QFD an optimization method is always needed to
determine the most desirable HOWs to satsify the WHATs [31,57]
under certain constraints. For example, Park and Kim [77] optimize
total absolute importance of the HOWs by formulating a 01 linear
and quadratic program to nd the most desirable HOWs under budget
constraint. Wasserman [103] formulates a linear program to nd the
most desirable design requirements, HOWs, under budget constraints.
Zhou [110] proposes a mixed integer linear program that maximizes
an utility function under budget, technological feasibiity and competition constraints to determine the HOWs.
Multiple objective optimization approaches [62,30] have also
been used in determining the HOWs. For example, Karsak et al.
[59] formulated a 01 goal program combining analytical network
process in QFD application in product planning. Delice and Gngr
[31] proposes a mixed integer goal program where the objectives
functions (customer satisfaction, cost and technicial difculties)

are converted into goals. Other multiple objectives optimization
based QFD applications also include various variations of goal
program, for example see Karsak and zogul [58], Chen and Weng
[19], Lee et al. [65], Bykzkan and Berkol [13]; among others.
Karsak [56,57] on the other hand has applied multiple objective
optimization in QFD and has determined the non-dominated
(efcient) solutions of QFD design requirements (HOWs). The
author however has not gone far enough and developed any
procedure of obtaining efcient design requirements in QFD
applications. In our research we contribute in this aspect and
develop a procedure to obtain efcient design requirements
(efcient resilience capabilities in our case).
Proposed methodology to determine efcient resilience

capabilities in QFD
The concept of efciency and generating efcient solutions is
prevalent in multiple objective decision making domain [62,30]. A
general multiple objective decision problem is represented as
follows:
)
MaxMin f i X C i X; i 1p
3
Subject to : g j X r bj ; j 1q
where X(x1, x2, , xn) are n-dimensional decision variables; fi (.)
represents p linear conicting objective functions and gj(.) are q
different constraints inequalities. A feasible solution Xn to problem
(3) is said to be efcient (for a maximizing problem) if there does
not exist any other feasible solution X such that for all i1, , p,
fi(X)Zfi(Xn), and fi(X)4fi(Xn) for at least one i. In other words Xn is
not dominated by any other solution in terms of achievement in
the objective function. As will be shown in a later section in our
application we have three objective functions in QFD optimization
10
formulated as follows:
P
Max f 1 X j A n REj xj
P
Max f 2 X k A n;k a j REk xk
Max f p X
Subject to :
xAX
Fig. 1. QFD Framework. Note: CRi customer requirements; Wi degree of importance of CRis; DRj design requirements; Rij relationship matrix (i.e. degree to
which CRi is met by DRj). A.I. absolute importance of DRjs; R.I. relative importance of DRjs.
problem as: maximize procurement strategies, maximize processing strategies, and maximize distribution strategies, to mitigate
the vulnerabilities.
As presented earlier we dene supply chain resilience efciency (SCREF) as follows:
(i) resilience capability must be resource efcient (e.g. minimum
cost of implementation), and
(ii) portfolio of chosen resilience capabilities must be efcient (or
non-dominated) from multiple objectives perspective [63].
Vugrin et al. [99] touched on the efciency of system resilience.

According to the authors system resilience must use lowest
possible amount of resources to be efcient. Inline with
Vugrin

et al. [99] we dene Resilience Efciency REj as AIj =C j , where C j
is the cost of implementing jth resilience strategy. It is noted that
parameters in the objective functions in our QFD optimization
problem are the REj. This satises our condition (i) of SCREF.
To satisfy condition (ii) of SCREF we ascertain that the solution
of the QFD multiple objective optimization problem is efcient
(non-dominated). We develop a multiple objective 01 optimization problem that will nd the efcient portfolio of resilience
strategies to mitigate the vulnerabilities. Pettit et al. [79] recently
have mentioned portfolio approach to resilience capabilities.
However, the authors did not offer any methodology of achieving
that. Our 01 QFD multiple objective optimization problem is
9
>
>
>
>
>
>
>
>
>
>
>
>
=
l A n;l a k a j RE l xl
>
>
n X
>
X
>
n
>
>
c j xj
s
x
x
r
B
>
ij
i
j
>
j4i
>
>
j1
i1
>
n
X
>
;
where n is the number of resilience startegies; REj is the resilience

efciency (described earlier) of resilience strategy j, xj is one or
zero depending on if the corresponding resilience strategy j is
selected or not; cj is the expected cost of implementing resilience
strategy j; sij is the savings if resilience strategies i and j are
implemented together; B is the available budget.
It is noted that our constraints are non-linear and similar to
Park and Kim [77]. But depending on the application the constraint set could be formulated in an extensive way including the
budget, technological feasibility and competition constraints [110].
We argue that there are p different conciting objectives among
the REjs which need to be optimized simultaneously, hence
satisfactory effcient solution of problem (4) need to be found out
by interacting with the decision maker. It is noted that any
solution to problem (4) will offer a portfolio of resilience strategies
to mitigate the vulnerabilities. To nd the efcient portfolios of
strategies we need to reformulate problem (4) as follows:
9
p
X
>
>
>
Max
i f i X
>
>
>
>
i1
=
n
n X
n
X
X
5
Subject to :
c j xj
sij xi xj r B >
>
>
>
>
j1
i 1j 4 i
>
>
;
x AX
where i (i1,,p) are positive numbers representing the weights
(importance) attached to the objective function f i X by the decision
maker. Theorems from multiobjective optimization domain suggest
that any solution of problem (5) above is an efcient (non-dominated)
solution to problem (4) [63]. It is noted that the important weights i
are only needed to nd the rst efcient solution to problem (4). It is
noted that Karsak [56,57] has also determined efcient solution for
multiple objective optimization in QFD. However our multi-objective
model formulation and solution approach are different from Karsak
[56,57] in the following ways: (i) Karasak deals with fuzzy multiple
objectives, while we deal with non-fuzzy multiple objective formulation, (ii) Karasak nds fuzzy priorities of the objectives and use those
priorities to nd a single efcient solution, while we provide an
interactive method which nds an initial efcient solution then
explores other efcient solutions by changing the weights as par the
likings of the decision makers. It is noted that in the domain of
multiple objective decision making interactive approaches are preferred than the non-interactive approaches [91, p. 98].
We now offer an interactive procedure to nd satisfactory portfolio
of efcient resilience strategies to mitigate the vulnerabilities.
Step 1: Optimize each objective function of problem (4). There
will be p such solutions. Offer them to the decision maker.
These will act as maximum goal of each individual objective.
Any efcient solution will be a compromise solution from
these goals.
Step 2: Formulate problem (5) where each i 1 (i 1,,p).
Solve problem (5). The solution will be efcient (non-dominated) for problem (4). Offer it to the decision maker.
Step 3: If the decision maker is satised with this solution (after
comparing it with the solutions found in step 1), Stop. This will
be the satisfactory portfolio of resilience strategies to mitigate

the vulnerabilities.
If the decision maker is not satised, go to step 4.
Step 4: Interact with the decision maker to nd new values of
is which represent his/her preferences for the objective
functions.
Step 5: Formulate and solve problem (5) with the new values of
is. Offer it to the decision maker. Go to Step 3.
Application in RMG industry of Bangladesh

Methodology
There are fundamentally two research paradigms: positivist and
interpretivist [76]. Positivist paradigm is associated with the quantitative research method based on specic research questions and
hypotheses testing [51,27]. Whereas the interpretivist paradigm relies
on the qualitative method and there is subjective interpretation of the
researcher involved [27]. However, in recent times research based on
mixed methods, a combination of qualitative and quantitative methods, has gained popularity [12], because it assists in increasing the
quality, accuracy, validity and reliability of data [27,2].
It is noted that the primary objective of this study is to develop
efcient resilience capabilities of RMG supply chain of Bangladesh to
mitigate organizations vulnerabilities. To effectively conduct this study
we have adopted mixed methods of qualitative and quantitative
approaches under positivist paradigm. This uniquely ts with the
Quality Function Deployment (QFD) as a research methodology which
has aspects of both qualitative and quantitative methods [77,102] and
which we have embraced in our current study.
Our applications in RMG industry of Bangladesh are conducted
in three studies and in three stages as follows.
Study 1
Our study 1 is conducted on one of the largest apparel manufacturer in Bangladesh. Since its establishment in 1984, the company has
been attaining specic experience in designing and manufacturing
different types of apparels. It specializes in high quality apparel
production, and is one of the leading apparel exporters in Bangladesh.
It has 30,000 employees in 28 apparel production units in different
parts of Bangladesh as well in Cambodia and Vietnam. It produces
Bottom, Shirt, Sportswear, Polo knit and Sweater compliant with
various quality requirements. It exports its products to North America,
South America and Western Europe. The total export volume of the
11
company is 120,000 dozen/month which accounts for an aggregate

turnover of US$100 million. Sears, C & A, PVH, GAP, Wal-mart, JCPanny
and H&M are the major buyers of its products. Paired with the apparel
production, it has developed the backward linkage facilities as it has
its own textile (waiving, cotton yarn spinning) with dying facilities of
cotton & synthetic, poly, label, button, zipper, thread and carton
factories. The apparel export of the company is increasing every year
however, it is still bogged with multiple disruptive events like labour
unrest, political instability, interruption in utility supply especially
power shortage; among others. These disruptive events expose the
company to various types of vulnerabilities. Although the conglomerate has 30 years of experience in combatting various vulnerabilities, a
formal resilient approach to mitigate the vulnerabilities is needed.
Consequently, the management of this group of companies has agreed
to take part in QFD approach proposed in this paper to explicate the
vulnerabilities and resilience strategies to overcome the vulnerabilities.
Stage 1
In stage 1 the supply chain vulnerabilities (WHATs) and corresponding resilience strategies (HOWs) to mitigate these vulnerabilities
are found. To accomplish this, data have been collected from three
RMG manufacturing companies and two accessory production companies of the parent Group. Although under the same conglomerate,
these ve companies have unique features, and face and try to resolve
problems independently. It also allows us to collect data from multiple
sources (ve decision makers) and thus, enhances the reliability of our
data [3]. The data have been collected via semi-structured detailed
interviews from ve respondents. Table 4 provides the proles of the
ve companies and the respondents. Each interview lasted between
60 and 80 min.
Table 5 presents the explicated supply chain vulnerabilities and
resilience strategies. There are 26 vulnerabilities. Out of these, 17
vulnerabilities have been supported by majority of the respondents.
These 17 vulnerabilities are considered for further analysis. Corresponding to the 17 vulnerabilities the respondents identied 13
resilience strategies to mitigate the vulnerabilities. It is observed that
most of these vulnerabilities and resilience strategies are consistent
with the literature (as per Tables 2 and 3). This adds further validity to
our collected data [3].
Stage 2
In stage 2 we collect the quantitative data (wi, Rij) to nd the AIj
and RIj (see Table 2 and Eqs. (1) and (2)). We also collect the data on
costs (cj) of implementing the resilience strategies in order to nd the
resilience efciency REj (see earlier discussion) and the savings sijs
when two resilience strategies i and j are implemented together. As
Table 4
Prole of ve companies and Participants.
Participant Position
Company type
Product(s)
Company size (number of Age of

employees)
company
Production and
sales
D1
Manager
Merchandising
Bottom item (shorts, trousers)
More than 5000
10 years
6080,000 dozens/
year
D2
Deputy general
manager
Manager
Merchandising
RMG
manufacturing
unit
Accessory
supplying unit
RMG
manufacturing
unit
RMG
manufacturing
unit
Accessory
supplying unit
Cartoon and poly bags which are used for packing

products during shipment.
Main product is shirt.
Less than 1000
5 years
20003000
10 years
141,500,000
cartoons each year
2530,000 dozens/
year
Main product is shirt.
More than 5000
25 years
4050,000 dozen/
year
15 years
405,000,000 yard/
year
D3
D4
Supply chain
manager
D5
Deputy General
manager
Process raw cotton to yarn and then spin the yarn for Less than 1000
knitting.
12
Table 5
Vulnerabilities and corresponding resilience capabilities.
Vulnerability factors
(WHATs)
Specic vulnerabilities
Enterprises
AHP Weights
Adjusted weights
1 2 3 4 5
Hazard (.132)
Natural disaster (ood, cyclone) (HV1)

Political instability and labour unrest (HV2)
Fire and other accidental damage (HV3)
Sabotage
Piracy and theft
y y
y
y y y y y
y y
y
y
y
y
.169
.534
.297
.022
.071
.039
Strategic (.223)
Increased competition (SV1)

Failure to comply with social-environmental factors (SV2)
Problem of relation with buyer & suppliers (SV3)
Problem of integration
y y y y y
y y y y
y y
y
y
.445
.383
.172
.099
.085
.04
Financial (.154)
Currency & raw material price uctuation (FV1)

Economic recession (FV2)
Bankruptcy of supply chain members (FV3)
Higher rate of bank intest
y y
y y
y y y y y
y
y
y
.253
.602
.145
.039
.093
.022
Operational (.195)
Worker (skill, absenteeism, illeteracy) (OV1)

Disruption in utility supply (OV2)
Product quality defection (poor quality) (OV3)
Machinery breakdown and failure
IT system failure
Production planning problem
y y y y
y y y y y
y
y
y
y
y
.282
.543
.175
.055
.106
.034
Infrastructural (.115)
Delay in port and customs (IV1)

Delay in transportation for poor infrastructure and port facilities (IV2)
Strike by port workers
y y y
y y
y
.578
.422
.067
.049
Demand & supply (.181)
Suppliers disruptions (DSV1)

y y y
Lack of material (dependence on imported material and lack of backward linkage) y y y y y
(DSV2)
Buyers disruptions (expect discount) (DSV3)
y y
y
Unpredictability of demand
y
.242
.597
.043
.108
.161
.029
Resience strategies
(HOWs)
Product differentiation and customization (St1)

Multiple sources of supply (St2)
Channel rerouting, reconguration (St3)
Back up capacity (St4)
Quality control and reducing defection (St5)
Skill and efciency development through traing and counselling (St6)
Product and process improvement for efciency and waste reduction (St7)
Forecasting and predictive analysis (St8)
Forward and backward linkage (St9)
Responsiveness to customer (St10)
Compliance of social and environmental issues (St11)
ICT adoption and information intergartion (St12)
Cooperation, communication and building relation with buyers and suppliers
(St13)
y
y y
y y y y
y
y y y
y y
y
y y y y
y y y
y y y
y
y y
y y y y
y y y y
y
y y
y y y y
y
y
y
y
y
y
y
y
y
y
y
Note: Factors with low responses have not been considered for importance rating.
these data are extremely demanding to collect, we select one RMG

manufacturing company from the parent group of companies (D1 in
Table 4). Table 4 shows that it is a large company both in terms of
number of employees and sales volume. It also produces more than
one product type and hence susceptible to more vulnerabilities.
Barratt et al. [3] mention that single case company allows to collect
much deeper data, which is the situation in our study. However, we
collect data from multiple sources (three decision makers) of the case
company which enhances reliability of the collected data [3].
To nd the wi we use Saatys [86] AHP method in a hierarchical
setting. The wis are averaged for three respondents and are shown
in the last column of Table 5. To nd Rijs we ask the respondents to
indicate (in their opinions) the extent of mitigating the vulnerability i by resiliency strategy j using the widely used scale of 9, 3,
1 and 0 [17,82,34]. The Rijs are also averaged for three respondents.
Fig. 2 shows the wiRij values ( in the main body of the matrix) and
the AIjs and the RIjs for different resilence strategies. It is observed
that resilience strategies 13 (building relations with with buyers
and suppliers) and 4 (back-up capacity) have the highest AIs of 4.11
and 3.74 respectively.
The cost (cj) of implementing the resilience strategies is found

in a more elaborate way. Each respondent is asked to give their
most likely, optimistic and pessimistic estimates of cj. Then the
expected cost is found by the formula Ce (4Cm Co Cp)/6, where
Ce, Cm, Co and Cp are the expected, most likely, optimistic and
pessimistic cost estimates. These costs are then averaged for three
respondents. Fig. 2 shows the costs cjs and the resilience efciencies REjs. It is noted that resilience strategy 13 (building relations
with buyers and suppliers) has the highest RE of .19 followed by
resilience strategy 8 (forecasting and prediction) of .15. The cost
gures are in millions of Taka.1 To nd the savings sij the
respondents were asked to indicate which resilience strategies
could be implemented simultaneously and what could be the
estimated savings. The roof of Fig. 2 shows these savings data. For
example, resilience strategies 1 and 10 can be implemented simultaneously and the estimated savings would be Taka 3.6 million.
1
Taka is Bangladeshi currency. At the time of this study the exchange rate was
1 US$ 77 Taka.
13
6.4
3.6
7
5.6
3.3
8
9.7
8.3
4
St1
0
0
0
0.891
0
0.067
0
0.279
0
0
0
0.034
0
St2
0.066
0.165
0.039
0.165
4.6
5.2
4.3
2.9
5.5
7.2
SCVs Weights
HV1
0.022
HV2
0.071
HV3
0.039
SV1
0.099
SV2
0.085
SV3
0.04
FV1
0.039
FV2
0.093
FV3
0.023
OV1
0.055
OV2
0.106
OV3
0.034
IV1
0.067
IV2
0.049
DSV1
0.043
DSV2
0.108
DSV3
0.029
A.I
Cost
RE
5.8
3.7
St3
0.066
7.9
5.6
4.8
St4
St5
St6
St7
St8
St9
St10
St11
St12
St13
0.98
0
0
0
0.037
0.022
0
0
0.051
0
0.497
0
0
0.071
0.071
0
0
0.497
0.165
0
0.117
0.351
0
0
0
0
0
0
0.351
0.117
0
0
0.099
0.693
0.693
0.891
0.099
0.231
0.693
0.495
0.231
0.891
0
0
0.255
0
0.198
0
0
0.085
0.595
0
0
0.12
0
0.12
0.04
0
0.2
0
0.12
0.28
0.04
0.04
0.36
0.117
0
0.195
0
0
0
0.351
0.039
0
0
0
0.195
0.279
0
0
0.22
0.651
0.837
0.837
0.155
0.279
0.093
0.22
0.837
0.161
0
0.023
0
0
0
0.069
0.038
0.023
0
0.38
0.54
0
0
0
0
0.495
0.055
0
0
0
0.165
0
0
0
0
0.247
0.106
0.106
0.106
0.106
0
0.106
0
0.106
0.247
0.08
0
0
0.306
0.238
0.306
0
0.067
0.102
0.057
0.034
0.102
0.56
0.18
0.112
0
0
0
0.18
0.18
0.067
0
0.067
0.112
0.114
0
0.147
0
0
0
0
0
0
0
0
0
0
0.387
0.72
0.129
0.043
0
0
0.1
0.215
0.129
0
0.129
0.387
0.324
0.972
0.252
0.756
0
0.108
0.18
0.252
0.756
0
0
0.108
0.18
0.029
0
0.029
0.087
0.203
0.261
0.203
0.029
0.087
0.068
0.145
0.087
0.261
1.624
3.225
1.364
3.743
1.866
2.552
3.047
2.131
1.91
1.832
2.438
1.735
4.112
51.6
36.4
25.2
27.3
22.7
18.8
37.2
14.3
71.8
22.4
56.7
15.6
21.6
0.031473 0.088599 0.054127 0.137106 0.082203 0.135745 0.081909 0.149021 0.026602 0.081786 0.042998 0.111218 0.19037
A.I
1.222
1.466
0.975
6.072
1.133
1.387
0.897
4.687
1.234
0.715
1.13
1.326
1.458
0.261
2.239
3.888
1.489
Fig. 2. Supply Chain resilience model: Study 1. Note: A.I.absolute importance; Stj resilience strategy j; HV, SV, FV, OV, IV, DSV various vulnerabilities, RE resilience
efciency.
Stage 3
In stage 3 we develop the 01 multiple objective problem (as in
(4)) and apply the stepwise procedure to nd the satisfactory
portfolio of efcient resilience strategies. In order to nd the
multiple objectives among the resilience strategies we interacted
with the three decision makers of the case company and came up
with three objectives to be maximized as follows:
f1(X) Procurement related strategies which includes strategies ST2, ST4, ST9 and ST13.
f2(X) Processing related strategies which includes ST5, ST6,
ST7, ST11 and ST12.
f3(X) Distribution related strategies which includes ST3, ST8,
ST1, and ST10.
Problem (4) now becomes:
Max f1(X) RE2x2 RE4x4 RE9x9 RE13x13
Max f2(X) RE5x5 RE6x6 RE7x7 RE11x11 RE12x12
Max f3(X) RE3x3 RE8x8 RE1x1 RE10x10
Subject to:
c1 x1 c2 x2 c3 x3 c4 x4 c5 x5 c6 x6 c7 x7
c8 x8 c9 x9 c10 x10 c11 x11 c12 x12
c13 x13 S1;6 x1 x6 S1;10 x1 x10 S2;4 x2 x4
S3;8 x3 x8 S3;9 x3 x9 S3;10 x3 x10 S4;8 x4 x8
S4;10 x4 x10 S4;12 x4 x12 S4;13 x4 x13 S5;6 x5 x6
S5;7 x5 x7 S6;7 x6 x7 S7;10 x7 x10
S7;11 x7 x11 S7;13 x7 x13 S8;13 x8 x13 S9;13 x9 x13 S10;11 x10 x11
S10;12 x10 x12 S11;13 x11 x13 rB where xj 0 or 1
The values of REj and cj are obtained from Fig. 2. According to

the decision makers the budget can be set aside as 110 million
Taka. Hence the budget B is set at 110 million Taka.
We now follow the stepwise procedure to nd the satisfactory
portfolio of efcient resilience strategies. As per step 1 of the stepwise
procedure, each objective is optimized separately. We use EXCEL
Solver as the optimization software. The optimal solutions and
corresponding portfolio of resilience strategies are shown in Table 6.
It is observed that optimal values of the resilience efciencies (REs) of
the procurement (f1), processing (f2) and distribution (f3) strategies are
.416, .411 and .316 respectively. If f1 is optimized alone then multiple
sources of supply (St2), back up capacity (St4) and cooperation and
communication with buyers and suppliers (St13) should be implemented (see Table 5) for a total cost of 85.3 million Taka leaving 24.7
million Taka unspent. However, since (St2, St4) and (St4, St13) are
implemented together there is potential savings of 13.6 million Taka
(see roof of Fig. 2). Thus, there is still 38.3 million Taka available for
implementing other strategies. Since forward and backward linkages
(St9; another component of f1) costs 71.8 million Taka to implement
the management can look into other strategies to be implemented for
the remaining budget of 38.3 million Taka. Other optimized solution
for f2 and f3 can be analysed similarly.
In our case the three decision makers wanted to explore more
efcient solutions in order to obtain a compromise among the
three objectives. We then applied step 3 of the stepwise procedure
using equal weighting for the three objectives. This produces an
efcient solution as shown in Table 7 (rst row). It is noted that
the objective function values (resilience efciencies) are .327 (for
f1; a deviation of 21.65% from the optimal value of f1), .329 (for f2; a
deviation of 19.95% from the optimal value of f2) and .231 (for f3; a
deviation of 26.89% from the optimal value of f3). The decision
makers now wanted to weight the objectives according to their
preferences and explore further compromise and efcient solutions. The steps 3 and 4 come into action now. After some
14
Table 6
Aspiration levels of the objectives for study 1.
Objective Function
f1
f2
f3
Aspiration level
.416
.411
.316
Procurement strategy (f1)
Processing strategy (f2)
Distribution strategy (f3)
Budget
X2
X4
X9
X13
X5
X6
X7
X11
X12
X1
X3
X8
X10
1
0
0
1
0
0
0
0
0
1
0
0
0
1
0
0
1
0
0
1
0
0
0
0
0
1
0
0
0
1
0
0
1
0
0
1
0
0
1
110
Note: X2, X4, X9, X13 procurement strategy (f1); X5, X6, X7, X11, X12 processing strategy (f2); X1, X3, X8 and X10 distribution strategy (f3).
Table 7
Efcient resilient portfolios of study 1.
Objective function
Weighted (1,1,1)
Weighted (.2, .5, .3)
Procurement strategies
Processing strategies
Distribution strategies
Budget
f1
f2
f3
X2
X4
X9
X13
X5
X6
X7
X11
X12
X1
X3
X8
X10
.327
.327
.329
.411
.231
.149
0
0
1
1
0
0
1
1
1
1
1
1
0
1
0
0
1
1
0
0
0
0
1
1
1
0
110
Note: .2 weight for f1; .5 weight for f2; .3 weight for f3.
deliberations the weights of (.2, .5, and .3) were settled with for
objectives f1, f2 and f3 respectively. It shows that the decision
makers prefer processing objective (f2) more compared to other
two objectives. Applying step 5 we now obtain a second efcient
solution as shown in Table 7 (second row). The objective function
values of this solution are .327 (for f1; a deviation of 21.65% from
the optimal value of f1), .411 (for f2; a deviation of 0% from the
optimal value of f2), and .149 (for f3; a deviation of 52.8% from the
optimal value of f3). It is interesting to note that this solution
produces optimal value for objective f2 (processing), as more
weight (.5) was given to this objective. However, the deviation
from the optimal solution for f3 was on the high side. The decision
maker therefore chose the earlier (equal weighting) solution with
objective function values of .327, .329, and .231. It is observed from
Table 7 that this solution selects strategies St4 (back-up capacity),
St13 (building relation with buyers and suppliers), St5 (quality
control), St6 (skill and efciency development), St12 (ICT adoption), St8 (forecasting), and St10 (responsiveness to customers) to
be implemented which require a total budget of 142.7 million
Taka. However there is a savings of 41.1 million Taka as (St4, St8),
(St4, St10), (St4, St12), (St4, St13), (St5, St6), (St8, St12), and (St10,
St12) are implemented together. Thus net required budget is 101.6
million Taka which is well within the budget constraint of 110
million Taka.
Study 2
The study 2 is conducted on one of the leading manufacturer of
jeans in Bangladesh to ensure the transferability of our ndings and
applicability of our method in another setting. From its modest
beginning in 1984 the company has come a long way. It has modern
research and development facility. Within its six production units the
company now employs over 22,000 employees, produces over 30
million jeans per year and exports to more than 25 countries. We
conducted the case study on one of the units of the company which
produces casual wears for both men and women. Two executives of
the unit took part in our study.
Stages 1 and 2
Tashakkori and Teddlie [96] mention that external validity of
quantitative research and transferability of qualitative research are
similar in nature, which refer to the degree to which results of one
context can be applicable to other situation. Since our study 2 is also
on apparel industry, we offered the vulnerabilities of study 1 to the

two executives of study 2 and asked them to delve into them and
come up with vulnerabilities specic to their situation. In the end the
executives came up with 25 vulnerabilities of which 20 are similar to
studies 1 and 5 are new. Table 8 shows these vulnerabilities. The
bolded ones are the new vulnerabilities. It is noted that for the next
stage of the analysis (i.e. developing the supply chain resilience
model) the highest three weighted vulnerabilities from each group
were considered. For example, for hazard vulnerability group HV1,
HV2, and HV3 were selected which had the highest weights among
the ve hazard vulnerabilities (see Fig. 3).
Next, resilience strategies to mitigate these vulnerabilities were
sought. Again the strategies of study 1 were given to the executives.
They explored these and at the end came up with 14 strategies of
which 13 are similar to study 1 and one is new (see Table 8). This
acceptability of the outcomes from study 1 to study 2 ensures the
transferability of our results.
In the next stage of the analysis, the quantitative data (wi, Rij)
were collected from the two executives. The data on cost cj and
savings sij were also estimated. Finally the supply chain resilience
model was developed as shown in Fig. 3. Unlike study 1 the
resilience strategy 9 (Demand forecasting) has the highest RE of
.108 followed by resilience strategy 5 (Maintaining reserve capacity). However it is noted that demand forecasting was ranked 2 in
study 1. This highlights the need for accurate demand forecasting
in the apparel industry of Bangladesh to mitigate the demand
supply related vulnerabilities.
Stage 3
Like study 1 in stage 3 we develop and solve the 01 multiple
objective model and nd the satisfactory portfolio of efcient
resilience strategies. After interacting with the two executives the
following three objectives were formulated:
f1(X)Procurement related strategies comprising ST3, ST5,
ST10 and ST14.
f2(X)Processing related strategies comprising ST1, ST6, ST7,
ST8, ST12, ST13.
f3(X)Distribution related strategies comprising ST4, ST9, ST2
and ST11.
For brevity the multi-objective model is not presented here which
is similar to study 1. The stepwise procedure is now followed to nd
15
Table 8
Vulnerabilities and strategies for study 2.
Vulenarabilities
Strategies
HV1 Sabotage
HV2 Political instability
HV3 Factory re
HV4 Natural disaster (ood, cyclone)
HV5 Foreign government policy change
SV1Competitive pressure
SV2Reputation loss
SV3Problem of integration with supply chain members
SV4Failure to comply with socio-environmental issues
FV1 Increasing raw material price
FV2 Impact of Economic recession
FV3 Cost of Finance
FV4 Exchnage rate uctuation
OV1 Shortage of skilled worker
OV2 Impact of Power crisis
OV3Concentrated production location
OV4 Machine breakdown & failure
OV5 IT system failure
OV6 Production planning problem
IV1 Delay in port
IV2 Poor transportation infrastructure
IV3 Delay due to export import document processing
DSV1 Suppliers disruption
DSV2 High dependence on imported materials
DSV3 Buyers disruption
St1 Security system improvement

St2Product customization
St3Alternative sources of supply
St4Alternative transportation routing
St5Maintaining reserve capacity
St6Quality control
St7Skill development training
St8Improving process technology
St9Demand forecasting
St10 Backward linkage development
St11 Quick response to customers requirements
St12 Improving social and environmental performance
St13 Information integration
St14 Cooperation and collaboration with suppliers.
4.5
9
9.5
4.5
7
3.5
7.5
8
5
SCVs Weights
HV1
0.02
HV2
0.063
HV3
0.034
SV1
0.082
SV2
0.089
SV3
0.07
FV1
0.041
FV2
0.081
FV3
0.023
OV1
0.048
OV2
0.104
OV3
0.033
IV1
0.061
IV2
0.045
IV3
0.036
DSV1
0.046
DSV2
0.106
DSV3
0.025
AI
1.007
Cost
RE
St1
0.27
0
0.238
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.508
11.5
0.044
St2
3.7
8.5
St3
0
0.147
0.039
0.165
0
St4
St5
St6
St7
St8
St9
St10
St11
St12
St13
St14
0
0.1
0
0
0
0
0
0
0
0
0
0
0
0.567
0
0
0
0.119
0
0
0
0.119
0.084
0
0
0.273
0
0
0
0
0
0
0.351
0.065
0.102
0.891
0
0
0.693
0.891
0.693
0.297
0.231
0.891
0.231
0.231
0.297
0
0
0
0.623
0
0
0
0
0.801
0
0.41
0.067
1.2
0.093
0.04
0
0.12
0
0.12
0.28
0
0.04
0.36
0
0.091
0
0.117
0
0
0
0.351
0.039
0
0
0
0.117
0.465
0.279
0
0
0.093
0.837
0.651
0.837
0.093
0.217
0.093
0.093
0.837
0
0.207
0
0.054
0
0
0
0.069
0.023
0.054
0
0.054
0.069
0
0
0
0
0
0.495
0.55
0
0
0
0.09
0
0
0
0
0
0.954
0
0
0
0.177
0
0
0
0
0.106
0
0
0
0.231
0
0
0
0
0
0
0
0
0
0
0.335
0.469
0.18
0
0
0
0.18
0.156
0
0
0
0.18
0.08
0
0.245
0
0
0
0
0
0
0
0
0
0
0
0
0.18
0
0
0
0.08
0.108
0
0
0
0
0
0.387
0
0.215
0
0
0
0.215
0.245
0.129
0
0.072
0.129
0
0.972
0
0.54
0
0.252
0
0.324
0.972
0
0
0
0.54
0.075
0
0.175
0.058
0.175
0.058
0.125
0.087
0.068
0.225
0.125
0.058
0
1.498
2.702
1.844
3.807
1.624
2.533
2.139
2.736
2.055
1.796
1.691
0.732
3.231
82.2
45.7
37.4
41.6
43.6
36.3
75.7
25.2
120.4
46.8
102.3
33.6
45.4
0.018224 0.059125 0.049305 0.091514 0.037248 0.06978 0.028256 0.108571 0.017068 0.038376 0.01653 0.021786 0.071167
AI
0.37
1.036
1.068
5.511
1.834
2.32
0.715
4.495
0.53
1.135
1.237
0.231
1.5
0.325
0.368
1.392
3.6
1.229
Fig. 3. Supply Chain resilience model: Study 2.
the satisfactory efcient portfolio for study 2. First, each objective

function is optimized separately. The optimal solutions and corresponding portfolio of resilience strategies are shown in Table 9. It is
observed that optimal values of the resilience efciencies (REs) of the
procurement (f1), processing (f2) and distribution (f3) strategies are
.163, .173 and 0. 196 respectively. It is noted that the optimal values of
the objectives are quite different from that of study 1. When f1 is
optimized alone maintaining reserve capacity (St5), and cooperation
and collaboration with suppliers (St14) should be implemented. It is

noted that these two strategies were also selected in study 1.
Having seen the aspiration levels of three optimized objectives
the two executives wanted to explore further efcient portfolios of
resilience strategies. Thus step 3 of our stepwise procedure came
into action at this stage. We rst found an efcient solution by
equal weighting of the objectives. This solution is shown in
Table 10 (rst row). Next the executives wanted to weight the
16
Table 9
Objective function
f1
f2
f3
Aspiration level
.163
.173
.196
Budget
X3
X5
X10
X14
X1
X6
X7
X8
X12
X13
X2
X4
X9
X11
0
0
0
1
0
0
0
0
0
1
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
1
0
0
1
130
Note: X3, X5, X10, X14 procurement strategy (f1); X1, X6, X7, X8, X12, X13 processing strategy (f2); X2, X4, X9, and X11 distribution strategy (f3).
Table 10
Efcient resilient portfolios of study 2.
Objective Function
Weighted (1,1,1)
Weighted (.25, .45, .3)
Budget
f1
f2
f3
X3
X5
X10
X14
X1
X6
X7
X8
X12
X13
X2
X4
X9
X11
.092
.092
.114
.151
.158
.109
0
0
1
1
0
0
0
0
1
1
0
1
1
1
0
0
0
0
0
0
0
0
1
0
1
1
0
0
130
objectives now and a weighting scheme of (.25, .45, and .3) was
settled with for objectives f1, f2 and f3 respectively. With this, the
second efcient solution is obtained which is shown in second row
of Table 10. Compared to the equal weighting solution this solution
still offers the optimal value for objective f1, with some improvement in objective f2 at the sacrice of objective f3. The two
executives preferred this solution compared to the equal weighting solution. It is observed from Table 10 that this compromise
solution selected ve out of fourteen resilience strategies as
follows: St1 (security system improvement), St5 (maintaining
reserve capacity), St6 (quality control), St7 (skill development
training) and St9 (demand forecasting). The study 2 also selects
one of the resilience strategies security system improvement
(St1) which was not even one of the resilience strategies of study 1.
It is noted that while in study 1 the preferred optimal efcient
solution was spread out among the three objectives (procurement,
processing, and distribution), in study 2 the preferred solution
concentrated more on the processing strategies. This highlights
the difference in management attitude of the two giant garment
companies in Bangladesh.
Study 3
Study 2 was primarily conducted to ensure the transferability
[96] of results and ndings from one context to another context (i.
e. transferring vulnerabilities and resilience strategies from study
1 to study 2). To ensure further external validity [14] of our
method we have conducted study 3 on another garment manufacturer from Bangladesh. It is noted that data from this company
were independently collected without any reference to studies
1 and 2. Two executives from the company took part in the study.
Study 3 company is a family owned business, situated in
Chittagong, Bangladesh and was established in 1983. It has four
manufacturing factories with work force of more than 3000. The
products of the company are casual and dress pants, shirts and
mens shorts. The company is certied by the buyers for complying
with the social and environmental sustainability factors. It markets
its products mainly in the USA, the UK and Australia. It has its own
washing plant, screen printing unit, embroidery unit and in-house
clearing unit. This group also has their own transportation and
logistical services. The company is growing steadily but the
external uncertainties, specically the political and economic
factors, pose threat to the smooth operation and growth of its
business. A resilience approach is therefore essential for the

survival and long term growth of the company.
Stages 1 and 2
The vulnerabilities and corresponding resilience strategies
were collected rst from the two executives of the company.
Table 11 shows these vulnerabilities and resilience strategies along
with their commonality with studies 1 and 2. There are three
hazard, ve strategic, three nancial, ve operational, two infrastructural, and ve demandsupply vulnerabilities. Among the 23
vulnerabilities of study 3, nine vulnerabilities are common to both
studies 1 and 2. These are: HV1Political unrest, HV2Fire and
other accident, HV3Natural disaster, SV2Increased competition,
SV4Problem of relation with buyers & suppliers, OV1Disruption
of utility, IV1Inefcient port facility, IV2Inefcient customs
process, and DSV1Dependence of imported material. It can be
thus reasonably assumed that these are the most common
vulnerabilities of Bangladesh garment industry. It is noted that
for the next stage of the analysis (i.e. developing the supply chain
model) the highest three weighted vulnerabilities from each group
were considered.
Table 11 also shows the 14 resilience strategies. It is noted that
six of them are common to both studies 1 and 2. These are: St1
Back up capacity, St3Focusing on sustainability practise, St5
Customer relationship development, St10Multiple suppliers,
St12Developing relationship with suppliers, and St13Product
differentiation.
In the next stage of the analysis the supply chain resilience
model was developed as shown in Fig. 4. It is noted that the
weight (wi) of the vulnerabilities, relationship value (Rij), cost (cj)
and savings (sij) were also collected from the two executives of the
company which are shown in Fig. 4. It is observed that resilience
strategy 5 (customer relationship development) has the highest RE
value of .332 followed by resilience strategy 6 (developing new
buyers and markets) and resilience strategy 8 (strict quality
control at different stage). Interestingly these rankings of strategies are quite different from studies 1 and 2.
Stage 3
Like studies 1 and 2 we then develop the 01 multiple
objective model and nd the satisfactory efcient resilient
17
Table 11
Vulnerabilities and strategies for Study 3.
Vulnerabilities
Study 1
Study 2
Strategies
Study 1
Study 2
HV1Political unrest
HV2Fire and other accident
HV3Natural disaster
SV1Reputation risk
SV2Increased competition
SV3Selecting wrong production site
SV4Problem of relation with buyers & suppliers
SV5Lack of sustainability planning & standard
FV1Increased production cost
FV2Lack of order
FV3Loss due to rejection of shipment
OV1Disruption of utility
OV2Switching of workers
OV3lead time failure
OV4Lack of efciency of workers
OV5Fault in Quality control
IV1Inefcient port facility
IV2Inefcient customs process
DSV1Dependence of imported material
DSV2Switching of buyers
DSV3Lack of commitment of suppliers (late delivery, quality )
DSV4Supply shortage
DSV5Buyers' opportunism
St1Back up capacity
St2Risk management team
St3Focusing on sustainability practise
St4Using updated technology
St5Customer relationship development
St6Developing new buyers & markets
St7Flexibility in production
St8Strict quality control at different stage
St9Training & development
St10Multiple suppliers
St11Supplier selection & evaluation
St12Developing relationship with suppliers
St13Product differentiation
St14Alternative transportation (e.g. air shipment)
strategies. As per the two executives of the company there are

three objectives to be optimized as follows:
f1(X) Procurement related strategies comprising ST1, ST10,
ST11 and ST12.
f2(X) Processing related strategies comprising ST2, ST3, ST4,
ST7, ST8, ST9.
f3(X) Distribution related strategies comprising ST5, ST6,
ST13 and ST14.
The stepwise procedure is now followed to nd the satisfactory
efcient portfolio of strategies for study 3. First, each objective is
optimized separately. Table 12 shows the aspiration level of
objectives for the budgetary restriction of 80 million Taka. Steps
2 and 3 now comes into play to interact with the decision makers
for satisfactory efcient solution. The executives settled with a
weighting scheme of (.2, .4, .35) for objectives f1, f2 and f3
respectively. The corresponding satisfactory solution is shown in
Table 13 (2nd row). It is noted that the satisfactory solution is a
compromise solution among the three objectives. Table 13 shows
that the selected portfolio of strategies is: St10 (Multiple suppliers), St11 (Supplier selection and evaluation), St12 (Developing
relationship with suppliers), St2 (Risk management team), St7
(Flexibility in production), St8 (Strict quality control at different
stage), St9 (Training and development), St5 (Customer relationship
development), and St6 (Developing new buyers and markets).
It is noted from the above ndings of study 3 that our method
possesses good external validity. In spite of the fact that data on
study 3 has been collected independently following the structured
procedure of our method, the results have similarities with studies
1 and 2. This was expected as all three companies belong to the
readymade garment industry in Bangladesh and they operate in
similar competitive environment. The differences in results can be
attributed to the differences in management attitudes of three
different companies and their tangible and intangible resources.
Discussions and implications

This research aimed to develop resilience capabilities of Bangladesh readymade garment (RMG) supply chain in order to
mitigate various vulnerabilities that RMG supply chain face. In

doing so, we adopted mixed-method research design with qualitative and quantitative approaches [27]. We dened supply chain
resilience efciency (SCREF) and develop an interactive methodology to determine efcient resilience capabilities based on QFD
approach. The proposed methodology is then applied on three
large RMG companies in Bangladesh. The three studies were
conducted to ensure external validity (and transferability) [14,96]
of our proposed methodology. Our experience indicates that the
proposed methodology can be successfully applied to determine efcient portfolio of resilience strategies to mitigate vulnerabilities.
While three companies face somewhat similar vulnerabilities
and their resilience strategies are also similar, there are some
differences in quantitative results that deserve attention. Table 11
shows the similarities and differences of vulnerabilities and
resilience strategies among the three companies. It is noted that
company 3 has come up with 11 new vulnerabilities which are
different from both companies 1 and 2. In terms of importance of
these vulnerabilities study 1 has identied four top vulnerabilities
as DSV2 (lack of materials), OV2 (disruptions in utility supply), SV1
(increased competition) and FV2 (impact of economic recession)
(see Fig. 3 and Table 6). While top four important vulnerabilities of
study 2 are DSV2 (high dependence on imported materials), OV2
(impact of power crisis), SV2 (reputation loss), SV1 (competitive
pressure), and FV2 (impact of economic recession) with equal
weighting with SV1 (see Fig. 3 and Table 8). It is noted that
reputation loss is a signicant vulnerability of study 2, while this
was not even considered as a possible vulnerability by study 1.
Both companies are export oriented. However study 2 company
prides in its R & D and takes quality issue very seriously, which
might explain why this company takes reputation loss very
seriously. The top four vulnerabilities of study 3 are FV2 (lack of
order), OV1 (disruption of utility), FV1 (increased production cost),
and HV1 (political unrest) (see Fig. 4 and Table 11). It is noted that
lack of order and increased production cost are signicant
vulnerabilities of study 3, while these two were not even considered as possible vulnerabilities of studies 1 and 2 (see Table 11).
Being family owned, company 3 works in an extremely competitive environment. Hence these two vulnerabilities are very signicant for the company.
18
1.5
2
SCVs
HV1
HV2
HV3
SV1
SV2
SV3
FV1
FV2
FV3
OV1
OV2
OV3
IV1
IV2
DSV1
DSV2
DSV3
A.I
Cost
RE
Weights
0.113
0.028
0.011
0.069
0.021
0.038
0.113
0.225
0.056
0.13
0.041
0.017
0.024
0.012
0.058
0.029
0.015
St1
0.339
0.252
0.099
0
0
0
0
0
0
1.17
0
0
0
0
0.174
0
0.045
2.079
30
0.069
St2
1.017
0.252
0.099
0
0
0.114
0
0
0
0
0
0
0
0
0
0
0
1.482
8
0.185
St3
0
0.252
0
0.207
0
0.342
0.339
0.675
0
0
0.123
0
0
0
0
0.087
0
2.025
40
0.051
5
St4
0
0
0
0.621
0
0
1.107
0
0
0
0
0.051
0
0
0
0.087
0
1.866
35
0.053
St5
0
0
0
0.207
0
0
0
2.025
0.168
0
0
0
0
0
0
0.261
0
2.661
8
0.332
St6
0
0
0
0.621
0
0
0
2.025
0
0
0
0
0
0
0
0
0
2.646
10
0.265
St7
0.339
0
0
0.207
0
0
0.339
0.675
0
0.39
0
0.153
0
0
0
0
0
2.103
15
0.14
St8
0
0
0
0.207
0
0.342
0.339
0.675
0.504
0
0
0
0
0
0
0
0
2.067
10
0.207
St9
0
0.084
0
0.207
0.063
0
1.107
0
0
0
0.123
0.051
0
0
0
0
0
1.635
10
0.164
St10
0
0
0
0
0
0
0.339
0
0
0
0
0
0
0
0.522
0
0.135
0.996
8
0.125
St11
0
0
0
0
0
0.342
0
0
0
0
0
0
0
0
0
0
0.135
0.477
5
0.095
St12
0
0
0
0
0
0.114
0.339
0
0
0
0
0.051
0
0
0
0
0.135
0.639
8
0.08
St13
0
0
0
0.621
0
0
0.339
0.675
0
0
0
0
0
0
0
0.087
0
1.722
25
0.067
St14
0.339
0
0
0
0
0
0
0
0
0
0
0.153
0.072
0
0.174
0
0
0.738
22
0.034
Fig. 4. Supply Chain resilience model: Study 3.

Table 12
Objective function
f1
f2
f3
Aspiration level
.369
.749
.698
Budget
X1
X10
X11
X12
X2
X3
X4
X7
X8
X9
X5
X6
X13
X14
1
0
0
1
0
0
1
0
0
1
0
0
0
1
0
0
0
0
0
1
0
0
1
0
0
1
0
0
1
0
0
0
1
0
0
1
0
0
1
0
0
1
80
Note: x1,x10,x11,x12 procurement strategy (f1); x2,x3,x4,x7,x8, x9 processing strategy (f2); x5,x6,x13 and x14 distribution strategy (f3).
Table 13
Efcient Resilient Portfolios of study 3.
Objective Function
Weighted (1,1,1)
Weighted (.25, .40,.35)
Budget
f1
f2
f3
X1
X10
X11
X12
X2
X3
X4
X7
X8
X9
X5
X6
X13
X14
.30
.30
.696
.696
.597
.597
0
0
1
1
1
1
1
1
1
1
0
0
0
0
1
1
1
1
1
1
1
1
1
1
0
0
0
0
80
As pointed out earlier the three companies have different

management attitudes while mitigating vulnerabilities via resilience strategies. This is reected while nding the satisfactory
efcient resilience strategies via our interactive method. In study
1 the decision makers settled with efcient resilience strategies
which assigned equal weights to the procurement (f1), processing (f2) and distribution (f3) strategies. Their preferred portfolio
of resilience strategies included seven out of possible 13 strategies
with two from procurement objective, three from processing
objective and two from distribution objective. This is a balanced
management approach practised by this company in order to
mitigate the vulnerabilities. On the other hand, in study 2 the

decision makers settled with efcient resilience strategies that
assigned more weight to the processing strategy. Their preferred
solution included ve out of possible 14 strategies with one from
procurement objective, three from processing objective and one
from distribution objective. The management of this company is
more concerned with satisfying the processing objective in order
to mitigate the vulnerabilities. It is observed that in terms of
choosing the portfolio of resilience strategies study 3 is similar to
study 1, where both companies chose a compromise solution
spread out among the three objectives. Study 3 thus chose nine
strategies (among 14 strategies) with three from procurement

objective, four from processing objective and two from distribution objective.
Theoretical implications
The most important theoretical contribution of our research is that
we propose an interactive methodology to obtain satisfactory portfolio
of resilience efcient strategies. This extends traditional QFD based
optimization method to determine the most desirable HOWs [31,57],
which are non-interactive and do not guarantee to nd efcient
solutions. Our method interacts with the decision makers in a
systematic way and generates efcient solutions (portfolios) which
satisfy the decision makers. We therefore take the satiscing
approach to decision making in our proposed methodology [93,89].
It is worthwhile to compare our methodology with existing
methodologies which deal with vulnerabilities in supply chains.
Wagner and Neshat [100] proposed a methodology to mitigate supply
chain vulnerabilities. The authors rst found various vulnerability
drivers and then used graph theory to quantify vulnerability index.
The directed graph shows the relationship among the vulnerability
drivers, the knowledge of which is useful to mitigate vulnerabilities.
The authors mention that supply chain managers can use risk
management methods and implement mitigation strategies to ease
vulnerability drivers. However, the authors do not offer any guide to
explicate the mitigation strategies nor do they show any relationship
between the vulnerabilities and mitigation strategies. In contrast in
our proposed method, although we do not develop any index for
vulnerability, we prioritize the vulnerabilities which are grouped as
hazard, strategic, nancial, operational, infrastructure, and demand
supply vulnerabilities [8]. We also develop specic resilience strategies
to mitigate the vulnerabilities and help the decision makers in an
interactive way to nd the satisfactory efcient portfolio of resilience
strategies. Goh et al. [40] also developed a quantitative stochastic
model and methodology to deal with supply chains and vulnerabilities. However their theoretical model development was for international facility location and distribution problem for a company with
one product. Their primary objective was to nd the optimal open
shut decision of plants and the corresponding shipment quantities
from various plants to various markets in order to maximize prot
and minimize risks. It is noted that although along the same vein
Goh et al.s problem denition and solution methodology is quite
different. It needs rich and elaborate data. We are not yet aware of its
application in any real world problem.
Managerial implications
Organizations need to develop resilience capabilities in order to
mitigate supply chain vulnerabilities [101]. In this vein our research
can help managers, primarily in the RMG sector in Bangladesh,
answer three fundamental questions: (i) what are the supply chain
vulnerabilities that RMG sector currently facing? (ii) What are the
resilience strategies to mitigate these vulnerabilities? (iii) What is the
efcient resilient portfolio of strategies to mitigate vulnerabilities
subject to budget constraint? Our study has identied a number of
vulnerabilities across three studies on three large RMG companies in
Bangladesh. With respect to question (i) it has been shown earlier that
there are lot of similarities among these vulnerabilities of three
companies. Other RMG companies in Bangladesh (and for that matter
elsewhere) can start with these vulnerabilities and contextualize them
for their specic situation. It has been shown in our study that the
vulnerabilities have good external validity. With respect to question
(ii) it has been observed that resilience strategies are also similar for
three companies. Once again, other RMG companies (in Bangladesh
and elsewhere) can start with these strategies and contextualize them
for their own use. For both questions (i) and (ii) starting with the
19
ndings of our study will help the managers of RMG companies to be

one-step ahead instead of developing the vulnerabilities and resilience
strategies from clean slate.
For question (iii) we used our methodology to nd the portfolio
of efcient resilient strategies subject to budget constraint. We
found that the three companies had different approaches to
managing the resilient strategies. Our ndings can help the RMG
managers as an eye opener to nd satisfactory efcient resilient
strategies in their own situation. It is highlighted here that our
interactive methodology is targeted for interaction with the
managers (decision makers) where they can see the results when
they input various weight for various objectives.
Conclusions
In this research we address the problem of mitigating vulnerabilities of the ready-made garment industry of Bangladesh. Our basic
methodology is QFD [17,41], which we use to nd the vulnerabilities
and their resilience strategies. We then develop an interactive multiobjective methodology to nd the satisfactory (as per the decision
makers) efcient portfolio of resilience strategies to mitigate the
vulnerabilities. We argue that this approach is novel and has a number
of advantages over existing approaches [31,57]. First, our method
ensures efcient portfolio of strategies and second, it interacts with
the decision makers and thus nds the satisfactory efcient solutions
in an interactive way.
We applied our method to three large garment companies in
Bangladesh. Using the stepwise procedure of QFD we found the
vulnerabilities, and corresponding resilience strategies in study 1. We
observed that these vulnerabilities and resilience strategies are
transferrable to study 2, although study 2 has few specic vulnerabilities and resilience strategies of its own. Data for study 3 has been
collected independently of studies 1 and 2. Even then there are some
similarities among the vulnerabilities and resilient strategies of studies
1, 2 and 3. Combining the top prioritized vulnerabilities of the three
studies we conclude that most important vulnerabilities of the RMG
industry of Bangladesh are: lack of materials (high dependence on
imported materials), disruptions in utility supply, increased competition (and hence competitive pressure), impact of economic recession,
and reputation loss. Similarly, combining the preferred resilience
strategies of the two studies we conclude that the most preferred
strategies to mitigate the vulnerabilities are: back-up capacity, building relation with buyers and suppliers, quality control, skill and
efciency development, ICT adoption, demand forecasting, responsiveness to customers, and security system improvement.
Our research is not free from limitations. First, for practical
application any QFD based method requires rich and detailed data.
Hence data collection is the major limitation. In our case one of the
researchers had good contacts with the RMG industry of Bangladesh.
It helped us to make good connections with three of the largest RMG
companies in Bangladesh. The management of these companies
understood the value of our analysis and were deeply motivated in
participating in the study. Second, in our data collection process we
used qualitative methods to collect various data. Hence reliability of
the data could be an issue. However, we dealt with real decision
makers and as such all the qualitative data (vulnerabilities, resilience
strategies) reect their perception as they see t for the companies.
We surmise that despite the limitations we have offered a detailed
methodology and its applications in RMG industry of Bangladesh in a
systematic manner. The methodology can be effectively applied in
other industrial settings and in other applications elsewhere. However, in any future applications the decision makers of the targeted
company need to be convinced about the value of the study. We have
found that once the value of the research is disseminated well the
decision makers are happy to be involved in the study.
20
Future research can be directed along the following ways. First,

instead of multi-objective optimization multi-attribute methods can
be used based on the resilience efciency values (RE) to nd the
portfolio of resilience strategies [28]. Second, to cater for uncertainties
in input data stochastic methodology can be used instead of fuzzy
methods and thus develop the stochastic efcient resilience strategies
[64]. Third, the concept of dynamics of vulnerability [83] can be
introduced in our methodology and thus develop dynamic portfolio of
resilience strategies. Lastly, a decision support system based on our
methodology can be developed and make it more comprehensive to
include a variety of constraints and objective functions.
Acknowledgements
We are indebted to Prof. Ben Lev, the special issue editor and
three anonymous reviewers for their extensive feedbacks which
greatly improved the paper.
References
[1] Asia News Network, A. Garment sector in B'desh hit hard by instability,
http://www.asianewsnet.net/Garment-sector-in-Bdesh-hit-hard-by-instabil
ity-45844.html; 2013 [accessed 18.06.13].
[2] Babbie E. The practice of social research. 11th ed. Belmont, CA, USA: Thomson
Wadsworth; 2007.
[3] Barratt M, Choi TY, Li M. Qualitative case studies in operations management:
trends, research outcomes, and future research implications. Journal of
Operations Management 2011;29(4):32942.
[4] Bevilacqua M, Ciarapica FE, Giacchetta G. A fuzzy-QFD approach to supplier
selection. Journal of Purchasing and Supply Management 2006;12(1):1427.
[5] BGMEA. Bangladesh apparel and textiles exposition, http://www.bgmea.
com.bd/batexpo/index.htm; 2012 [accessed 04.04.13].
[6] Bhattacharya A, Sarkar B, Mukherjee SK. Integrating AHP with QFD for robot
selection under requirement perspective. International Journal of Production
Research 2005;43(17):367185.
[7] Blackhurst JV, Scheibe KP, Johnson DJ. Supplier risk assessment and monitoring for the automotive industry. International Journal of Physical Distribution & Logistics Management 2008;38(2):14365.
[8] Blos MF, Quaddus M, Wee HM, Watanabe K. Supply chain risk management
(SCRM): a case study on the automotive and electronic industries in Brazil.
Supply Chain Management: An International Journal 2009;14(4):24752.
[9] Bode C, Wagner SM, Petersen KJ, Ellram LM. Understanding responses to
supply chain disruptions: insights from information processing and resource
dependence perspectives. Academy of Management Journal 2011;54
(4):83356.
[10] Bottani E, Rizzi A. Strategic management of logistics service: a fuzzy QFD
approach. International Journal of Production Economics 2006;103
(2):58599.
[11] Braunscheidel MJ, Suresh NC. The organizational antecedents of a rms
supply chain agility for risk mitigation and response. Journal of Operations
Management 2009;27(2):11940.
[12] Bryman A. Integrating quantitative and qualitative research: how is it done?
Qualitative Research 2006;6(1):97113.
[13] Bykzkan G, Berkol C. Designing a sustainable supply chain using an
integrated analytic network process and goal programming approach in
quality function deployment. Expert Systems with Applications 2011;38
(11):1373148.
[14] Calder BJ, Phillips LW, Tybout AM. The concept of external validity. Journal of
Consumer Research 1982;9(3):2404.
[15] Carnevalli JA, Miguel PC. Review, analysis and classication of the literature
on QFD types of research, difculties and benets. International Journal of
Production Economics 2008;114(2):73754.
[16] Carpenter S, Walker B, Anderies JM, Abel N. From metaphor to measurement:
resilience of what to what? Ecosystems 2001;4(8):76581.
[17] Chan LK, Wu ML. Quality function deployment: a literature review. European
Journal of Operational Research 2002;143(3):46397.
[18] Chan LK, Wu ML. Prioritizing the technical measures in Quality Function
Deployment. Quality Engineering 1998;10(3):46779.
[19] Chen LH, Weng MC. An evaluation approach to engineering design in QFD
processes using fuzzy goal programming models. European Journal of
Operational Research 2006;172(1):23048.
[20] Choi Y, Ye X, Zhao L, Luo AC. Optimizing enterprise risk management: a
literature review and critical analysis of the work of Wu and Olson. Annals of
Operations Research 2015:120.
[21] Choudhury SR, Hossain MG. Post-MFA issues and challenges: social dimension. Dhaka, Bangladesh: Ministry of Commerce; 2005.
[22] Chowdhury GH, Sarker MAR, Afroze R. Recent unrest in the RMG sector of
Bangladesh: is this an outcome of poor labour practices? International
Journal of Business and Management 2012;7(3):20618.
[23] Chowdhury MH, Dewan MNA, Quaddus MA. Supply chain resilience to
mitigate disruptions: a QFD approach. In: Pacic Asia Conference on
Information Systems (PACIS). Ho Chi Minh city, Vietnam: AIS; 2012.
[24] Christopher M, Peck H. Building the resilient supply chain. International
Journal of Logistics Management 2004;15(2):113.
[25] Colicchia C, Dallaria F, Melacini M. Increasing supply chain resilience in a
global sourcing context. Production Planning & Control 2010;21(7):68094.
[26] Craighead CW, Blackhurst J, Rungtusanatham MJ, Handeld RB. The severity
of supply chain disruptions: design characteristics and mitigation capabilities. Decision Sciences 2007;38(1):13156.
[27] Creswell JW, Clark VLP. Designing and conducting mixed methods research.
Thousand Oaks, CA: Sage Publications; 2007.
[28] Cristiano JJ, White III CC, Liker JK. Application of multiattribute decision
analysis to quality function deployment for target setting. IEEE Transactions
on Systems, Man, and Cybernetics, Part C: Applications and Reviews 2001;31
(3):36682.
[29] Crowe TJ, Cheng C-C. Using quality function deployment in manufacturing
strategic planning. International Journal of Operations & Production Management 1996;16(4):3548.
[30] Deb K. Multi-objective optimization. In: Burke EK, Kendall G, editors. Search
methodologies: introductory tutorials in optimization and decision support
techniques. New York: Springer; 2014. p. 40349.
[31] Delice EK, Gngr Z. A mixed integer goal programming model for discrete
values of design requirements in QFD. International Journal of Production
Research 2011;49(10):294157.
[32] Dickinson G. Enterprise risk management: its origins and conceptual
foundation. Geneva Papers on Risk and Insurance Issues and Practice
2001;26(3):3606.
[33] Duclos LK, Vokurka RJ, Lummus RR. Delphi study on supply chain exibility.
International Journal of Production Research 2005;43(13):2687708.
[34] Faisal MN. Managing risk in small and medium enterprises (SMEs) supply
chains' using quality function deployment approach. International Journal of
Operations Research and Information Systems 2013;4(1):6483.
[35] Fennell ML, Alexander JA. Organizational boundary spanning in institutionalized environments. Academy of Management Journal 1987;30(3):45676.
[36] Ferdousi F, Ahmed A. An investigation of manufacturing performance
improvement through lean production: a study on Bangladeshi Garment
Firms. International Journal of Business and Management 2009;4(9):10616.
[37] Fibre2fashion News Desk. US suspends GSP facility to Bangladesh, http://
www.bre2fashion.com/news/apparel-news/newsdetails.aspx?news_
id=147897; 2013 [accessed 04.06.13].
[38] Fiksel J. Designing resilient, sustainable systems. Environmental Science &
Technology 2003;37(23):53309.
[39] Gaudenzi B, Borghesi A. Managing risks in the supply chain using the AHP
method. The International Journal of Logistics Management 2006;17
(1):11436.
[40] Goh M, Lim JYS, Meng F. A stochastic model for risk management in global
supply chain networks. European Journal of Operations Research
2007;182:16473.
[41] Gonzlez ME, Quesada G, Picado F, Eckelman CA. Customer satisfaction using
QFD: an e-banking case. Managing Service Quality 2004;14(4):31730.
[42] Haider MZ. Competitiveness of the Bangladesh Ready-made Garment
Industry in major international markets. Asia-Pacic Trade and Investment
Review 2007;3(1):327.
[43] Han SB, Chen SK, Ebrahimpour M, Sodhi MS. A conceptual QFD planning
model. The International Journal of Quality & Reliability Management
2001;18(8):796812.
[44] Heckmann I, Comes T, Nickel S. A critical review on supply chain risk
denition, measure and modeling. Omega 2015;52:11932.
[45] Hendricks KB, Singhal VR. The effect of supply chain glitches on shareholder
wealth. Journal of Operations Management 2003;21:50122.
[46] Holling CS. Resilience and stability of ecological systems. Annual Review of
Ecology and Systematics 1973;4:123.
[47] Hossan CG, Sarker MAR, Afroze R. Recent unrest in the RMG Sector of
Bangladesh: is this an outcome of poor labour practices? International
Journal of Business and Management 2012;7(3):206.
[48] Islam MA, Begum MN, Rashed CAA. Operational disturbances and their
impact on the manufacturing businessan empirical study in the RMG
Sector of Bangladesh. International Journal of Research in Management &
Technology 2012;2(2):18491.
[49] Islam MA, Deegan C. Motivations for an organisation within a developing
country to report social responsibility information: evidence from Bangladesh. Accounting, Auditing & Accountability Journal 2008;21(6):85074.
[50] Jiang B, Baker RC, Frazier GV. An analysis of job dissatisfaction and turnover
to reduce global supply chain risk: evidence from China. Journal of Operations Management 2009;27(2):16984.
[51] Johnson RB, Onwuegbuzie AJ. Mixed methods research: a research paradigm
whose time has come. Educational Researcher 2004;33(7):1426.
[52] Jugulum R, Sek M. Building a robust manufacturing strategy. Computers &
Industrial Engineering 1998;35(1):2258.
[53] Jttner U, Maklan S. Supply chain resilience in the global nancial crisis: an
empirical study. Supply Chain Management: An International Journal
2011;16(4):24659.
[54] Kamvysi K, Gotzamani K, Andronikidis A, Georgiou A. Capturing and

prioritizing students requirements for course design by embedding FuzzyAHP and linear programming in QFD. European Journal of Operational
Research 2014;237:108394.
[55] Karim S. Coping with post-MFA challenges: strategic response for Bangladesh RMG sector. Dhaka: CPD; 2003.
[56] Karsak EE. Fuzzy multiple objective decision making approach to prioritize
design requirements in quality function deployment. International Journal of
Production Research 2004;42(18):395774.
[57] Karsak EE. Fuzzy multiple objective programming framework to prioritize
design requirements in quality function deployment. Computers & Industrial
Engineering 2004;47(2):14963.
[58] Karsak EE, zogul CO. An integrated decision making approach for ERP
system selection. Expert Systems with Applications 2009;36(1):6607.
[59] Karsak EE, Sozer S, Alptekin SE. Product planning in quality function
deployment using a combined analytic network process and goal programming approach. Computers & industrial engineering 2002;44(1):17190.
[60] Khondker BH, Razzak A, Ahmed N. Exports, employment and working
conditions: emerging issues in the post-MFA RMG industry(revised draft).
International Labour Ofce; 2005.
[61] Kleindorfer PR, Saad GH. Managing disruption risks in supply chains.
Production & Operations Management 2005;14(1):5368.
[62] Kksalan MM, Wallenius J, Zionts S. Multiple criteria decision making: from
early history to the 21st century. World Scientic; 2011.
[63] Larbani M, Aouni B. A new approach for generating efcient solutions within
the goal programming model. Journal of the Operational Research Society
2011;62(1):17582.
[64] Lee DE, Lim TK, Arditi D. Automated stochastic quality function deployment
system for measuring the quality performance of design/build contractors.
Automation in Construction 2009;18(3):34856.
[65] Lee AH, Kang HY, Yang CY, Lin CY. An evaluation framework for product
planning using FANP, QFD and multi-choice goal programming. International
Journal of Production Research 2010;48(13):397797.
[66] Lin CM, Lee Y, Ho T. Applying integrated DEA/AHP to evaluate the economic
performance of local governments in China. European Journal of Operational
Research 2011;209:12940.
[67] Lockwood JL, Pimm SL. Biological diversity: species: would any of them be
missed? Current Biology 1994;4(5):4557.
[68] Ludwig D, Walker B, Holling CS. Sustainability, stability and resilience.
Conservation Ecology 1996;1:127.
[69] Martin S. The cost of restoration as a way of dening resilience: a viability
approach applied to a model of lake eutrophication. Ecology and Society
2004;9(2):8.
[70] Mitroff IL, Alpaslan MC. Preparing for evil. Harvard Business Review 2003;81
(4):10915.
[71] Mizgier KJ, Wagner SM, Holyst JA. Modeling defaults of companies in multistage supply chain networks. International Journal of Production Economics
2012;135(1):1423.
[72] Narasimhan R, Kim SW. Information system utilization strategy for supply
chain integration. Journal of Business Logistics 2001;22(2):5175.
[73] Nuruzzaman A Haque, Raq A. Is Bangladeshi RMG Sector t in the Global
Apparel Business? Analyses the supply chain management The South East
Asian Journal of Management 2010;4(1):5372.
[74] Olson DL, Wu D. Enterprise risk management models. Heidelberg: Springer;
2010.
[75] Olson DL, Wu D. A review of enterprise risk management in supply chain.
Kybernetes 2010;39(5):694706.
[76] Onwuegbuzie AJ, Leech NL. On becoming a pragmatic researcher: the
importance of combining quantitative and qualitative research methodologies. International Journal of Social Research Methodology 2005;8(5):
375387.
[77] Park T, Kim KJ. Determination of an optimal set of design requirements using
house of quality. Journal of Operations Management 1998;16(5):56981.
[78] Peck H. Reconciling supply chain vulnerability, risk and supply chain
management. International Journal of Logistics Research and Applications
2006;9(2):12742.
[79] Pettit TJ, Croxton KL, Fiksel J. Ensuring supply chain resilience: development
and implementation of an assessment tool. Journal of Business Logistics
2013;34(1):4676.
[80] Pettit TJ, Fiksel J, Croxton KL. Ensuring supply chain resilience: development
of a conceptual framework. Journal of Business Logistics 2010;31(1):121.
[81] Ponomarov SY, Holcomb MC. Understanding the concept of supply chain
resilience. International Journal of Logistics Management 2009;20(1):
124139.
21
[82] Pujawan IN, Geraldin LH. House of risk: a model for proactive supply chain
risk management. Business Process Management Journal 2009;15
(6):95367.
[83] Raharjo H, Xie M, Brombacher AC. A systematic methodology to deal with
the dynamics of customer needs in Quality Function Deployment. Expert
Systems with Applications 2011;38(4):365362.
[84] Ramanathan R, Yunfeng J. Incorporating cost and environmental factors in
quality function deployment using data envelopment analysis. Omega
2009;37:71123.
[85] Rose A. Dening and measuring economic resilience to disaster. Disaster
Prevention and Management 2004;13(4):30714.
[86] Saaty TL. AHP: the analytic hierarchy process. New York: McGraw-Hill; 1980.
[87] Sawik T. On the fair optimization of cost and customer service level in a
supply chain under disruption risks. Omega 2015;53:5866.
[88] Schmitt A, Sun S, Snyder L, Shen Z. Centralization versus decentralization:
risk pooling, risk diversication, and supply chain disruptions. Omega
2015;52:20112.
[89] Schwartz B, Ben-Haim Y, Dacso C. What makes a good decision? Robust
satiscing as a normative standard of rational decision making Journal for
the Theory of Social Behaviour 2011;41(2):20927.
[90] Shef Y, Rice JB. A supply Chain View of the resilient Enterprise. MIT Sloan
management 2005;47(1):478.
[91] Shin WS, Ravindran A. Interactive multiple objective optimization: Survey I
continuous case. Computers & Operations Research 1991;18(1):97114.
[92] Simchi-Levi D, Schmidt W, Wei Y. From superstorms to factory res
managing unpredictable supply-chain disruptions. Harvard Business Review
2014;92(12):96101.
[93] Simon H. Theories of bounded rationality. In: McGuire CB, Radner R, editors.
Decisions and organizations. North Holland; 1972.
[94] Stratton B. The rened focus of automotive quality. Quality Progress 1989;22
(10):4750.
[95] Tang CS. Robust strategies for mitigating supply chain disruptions. International Journal of Logistics Research and Applications 2006;9(1):3345.
[96] Tashakkori A, C Teddlie C, editors. Sage handbook of mixed methods in social
& behavioral research. Sage; 2010.
[97] Tewari M. Is price and cost competitiveness enough for apparel rms to gain
market share in the world after quotas? A review Global Economy Journal
2006;6(4):146.
[98] Tomlin B. On the value of mitigation and contingency strategies for
managing supply chain disruption risks. Management Science 2006;52
(5):63957.
[99] Vugrin ED, Warren DE, Ehlen MA. A resilience assessment framework for
infrastructure and economic systems: quantitative and qualitative resilience
analysis of petrochemical supply chains to a hurricane. Process Safety
Progress 2011;30(3):28090.
[100] Wagner SM, Neshat N. Assessing the vulnerability of supply chains using
graph theory. International Journal of Production Economics 2010;126
(1):1219.
[101] Wagner SM, Neshat N. A comparison of supply chain vulnerability indices for
different categories of rms. International Journal of Production Research
2012;50(11):287791.
[102] Wang HF, Hong WK. An integrated service strategy by QFD approach: a case
of a telecom company in Taiwan. International Journal of Management and
Decision Making 2007;8(24):25167.
[103] Wasserman GS. On how to prioritize design requirements during the QFD
planning process. IIE Transactions 1993;25(3):5965.
[104] Wu D, Olson DL. On the special issue of risk issues in operations: methods
and tools. Production Planning and Control 2009;20(4):2934.
[105] Wu DD, Olson D. Enterprise risk management: a DEA VaR approach in vendor
selection. International Journal of Production research 2010;48(16):491932.
[106] Wu DD, Olson D. Enterprise risk management: coping with model risk in a
large bank. Journal of the Operational Research Society 2010;61(2):17990.
[107] Wu D, Olson DL, Birge JR. Introduction to special issue on "Enterprise risk
management in operations. International Journal of Production Economics
2011;134:12.
[108] Wu D, Chen SH, Olson DL. Business intelligence in risk management: some
recent progress. Information Sciences 2014;256:17.
[109] Wu T, Blackhurst J, Chidambaram V. A model for inbound supply risk
analysis. Computers in Industry 2006;57(4):35065.
[110] Zhou M. Fuzzy logic and optimization models for implementing QFD.
Computers & Industrial Engineering 1998;35(1):23740.
[111] Zsidisin GA, Melnyk SA, Ragatz GL. An institutional theory perspective of
business continuity planning for purchasing and supply management. International Journal of Production Research 2005;43(16):340120.

1 s2.0 S0305048315001334 Main

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

1 s2.0 S0305048315001334 Main

Transféré par

Droits d'auteur :

Formats disponibles

Omega 57 (2015) 521

Contents lists available at ScienceDirect

A multiple objective optimization based QFD approach for efcient

It is an equally authored paper. Authors names are listed in alphabetical order.

This manuscript was processed by Associate Editor B. Lev.

eleven hundred workers [37]. These disruptions have chain effect to

Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521

proposing the ways and means to achieve supply chain resilience

present study is highly relevant to the supply chain aspect of

(i) resilience capability must be resource efcient (e.g. minimum

Bangladesh is one of the leading exporters of Ready Made

We shall elaborate on (i) and (ii) later. It is observed that a number

Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521

cohesion, control, connectedness to measure resilience [80,81,98].

Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521

Natural disaster (ood, cyclone)

Haider [42]; Blos et al. [8]

Shortage of skilled worker

Delay in custom clearance

Demand & supply

Blackhurst et al. [7]

Blos et al. [8]

[8]; Blackhurst et al. [7]

Craighead et al. [26]

In QFD modelling, customer requirements or existing problems of

where AIj is the absolute importance of jth design requirement (DR)

The correlation between the DRjs (HOWs) (see Fig. 1) plays a

Integration with Analytical Hierarchy Process (AHP)

Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521

Flexibility in production (different volume of order, exible production schedule)

Duclos et al. [33]; Braunscheidel and

Alternative and reserve capacity (logistical options)

Sharing information with supply chain partners

Waste elimination (efcient use of resource)

Pettit et al. [79]

Braunscheidel and Suresh [11]; Blackhurst

Efcient and hardworking employees

Pettit et al. [79]; Kleindorfer and Saad [61]

Customer satisfaction & market

Buyer satisfaction (product quality and service)

Pettit et al. [79]

Pettit et al. [79]; Tang [95]

functions (customer satisfaction, cost and technicial difculties)

Proposed methodology to determine efcient resilience

Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521

Vugrin et al. [99] touched on the efciency of system resilience.

where n is the number of resilience startegies; REj is the resilience

Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521

be the satisfactory portfolio of resilience strategies to mitigate

Application in RMG industry of Bangladesh

company is 120,000 dozen/month which accounts for an aggregate

Company size (number of Age of

Bottom item (shorts, trousers)

More than 5000

Cartoon and poly bags which are used for packing

Less than 1000

Main product is shirt.

More than 5000

Md.M.H. Chowdhury, M.A. Quaddus / Omega 57 (2015) 521

Natural disaster (ood, cyclone) (HV1)

Increased competition (SV1)

Currency & raw material price uctuation (FV1)

Worker (skill, absenteeism, illeteracy) (OV1)