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

net/publication/264837887

Application of goal programming to manage agility in manufacturing systems

Article  in  International Journal of Agile Systems and Management · January 2011

DOI: 10.1504/IJASM.2011.040516

CITATIONS READS

6 149

2 authors, including:

P.G. Saleeshya
Amrita Vishwa Vidyapeetham
45 PUBLICATIONS   308 CITATIONS   

SEE PROFILE

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

Sustainable manufacturing systems, Productivity improvement through lean initiative,Lean inventory management View project

All content following this page was uploaded by P.G. Saleeshya on 16 May 2015.

The user has requested enhancement of the downloaded file.

222 Int. J. Agile Systems and Management, Vol. 4, No. 3, 2011

Application of goal programming to manage agility in

manufacturing systems

P.G. Saleeshya*
Department of Mechanical Engineering,
Amrita School of Engineering,
Amrita University,
Coimbatore, 641105, India
Fax: +091-422-2656274 E-mail: pg_saleeshya@cb.amrita.edu
*Corresponding author

A. Subash Babu
Department of Mechanical Engineering,
Indian Institute of Technology,
Bombay, Powai, Mumbai, 400076, India
E-mail: subash@me.iitb.ac.in

Abstract: This paper is related to a real life study carried out in Indian
manufacturing organisations to assess the extent to which the agile
manufacturing paradigm is recognised and deployed, as this paradigm has been
a major objective of many companies. A multi-level framework developed by
the authors along with a questionnaire was used to identify various enabling
factors of agility in Indian manufacturing industries. The data collected from
the industries was analysed by the model called objectivated agility realisation
model (OARM) developed by the authors. This model helps to determine the
measure of agility called ‘effectiveness index’ (EI) at various hierarchical
levels of an organisation. The authors also proposed an approach to measure
the agility of manufacturing systems by combining the analytical hierarchy
process (AHP) and the data envelopment analysis (DEA) methodologies. The
approaches of AHP, OARM and DEA are essentially in the category of
diagnostic methods. Therefore, a need was felt to resort to an appropriate
method, which may provide proper guidelines for deploying the enablers and
also use a suitable method of optimisation. Goal programming method has been
identified suitable to deal with this problem and the salient details of which is
presented in this paper.
Keywords: analytic hierarchy process; AHP; data envelopment analysis; DEA;
effectiveness index; EI; key performance area; KPA; objectivated agility
realisation model; OARM; goal programming; GP; agility.
Reference to this paper should be made as follows: Saleeshya, P.G. and
Babu, A.S. (2011) ‘Application of goal programming to manage agility in
manufacturing systems’, Int. J. Agile Systems and Management, Vol. 4, No. 3,
pp.222–237.
Biographical notes: P.G. Saleeshya is a Professor of Mechanical Engineering
at the Amrita University, Coimbatore, India. She received her PhD in Industrial
Engineering and Operations Research (Agile Manufacturing) from Indian
Institute of Technology, Bombay, India. Her research is mainly focused on the
area of operations strategy: agile manufacturing, lean manufacturing, Lean Six
Sigma, green design and manufacturing, and responsive supply chain.

Copyright © 2011 Inderscience Enterprises Ltd.

 Application of goal programming to manage agility in manufacturing systems 223

A. Subash Babu is currently a Professor of Industrial Engineering and

Operations Research at the Department of Mechanical Engineering,
Indian Institute of Technology Bombay. His areas of interest in teaching
and research include design and management of decision supports for
production, project, supply chain and service systems focusing on issues
related to asset management, inventory management, scheduling, productivity
management, quality management, reliability and maintenance management,
entrepreneurship, and technology-based enterprise solutions. He has guided
20 PhDs at IIT Bombay in these areas and has handled a number of sponsored
and consultancy projects. He is an alumnus of IIT Delhi. His papers have
appeared in many international journals. He is in the editorial board of many
international journals. He has also guest edited special issues of few journals.

1 Introduction

Agility as the term used herein is the ability of an enterprise to respond quickly and
successfully to change. For a company to be agile it must be capable of operating
profitably in a competitive environment of continuous and unpredictable change
(Goldman et al., 1995). Agile manufacturing is a strategy that can create flexible or
virtual organisations to meet increasing customer expectations (Huang et al., 2004). In
modern business environments, agility is advocated as the fundamental characteristics for
competitiveness (Bottani, 2009). Agility is not only the outcome of technological
achievement, advanced organisational and managerial structure and practice, but also a
product of human abilities, skills and motivations. A knowledgeable and highly
motivated workforce organised in a manner to facilitate rapid realignment in anticipation
of or as a response to change is a critical ingredient in agility. The technology employed
by organisations must support the activities and tasks required in and between firms to
continuously meet changing customer needs (Steenhuis and de Boer, 2003).
Implementation of agile manufacturing paradigm has been a major objective of many
companies. To achieve the status of an agile manufacturer, organisations need to clearly
understand the concept of agility relative to their industrial and business circumstances
and then justify and acquire the appropriate characteristics. Critical to successful
accomplishment of agile manufacturing is the ability to combine a number of enablers. It
is important to know and assess the degree of agility of an organisation process along
with the companies’ progress towards becoming an agile system (Gunasekaran, 1998).
A study of Indian manufacturing industries covering 26 major engineering companies
categorised under five groups such as automotive (G1), small products (G2), medium size
products (G3), Small and medium machinery manufacturing (G4) and Large machinery
manufacturing (G5) was conducted with the help of a questionnaire, the framework
developed (Saleeshya and Babu, 2003) and AHP (Saaty, 1980).
The AHP-based study identified different good management practices, which the
organisations are following to enhance their agility. This study helped establishing the
importance that different organisations assign to various factors identified as the enablers
of agility. This factors of importance calculated with the help of AHP is termed as
‘importance index’ of enablers of agility in the present study. The results of AHP-based
analysis shows that in a manufacturing organisation, a number of agility enablers exist
and the ranks obtained for each of these enablers further convey that these enablers
224 P.G. Saleeshya and A.S. Babu

command varying importance in the organisation. The very presence of variation offers
challenges to pursue this subject further to find out how best an organisation’s efforts can
be channelised to minimise the impact of such variations, that the agility of an
organisation is not just left to chance factors alone, but manage effectively to achieve
superior agility. With the help of this study it is not possible to assess the contribution of
different sub-sections of an organisation to enhance the overall agility of that
organisation. For this, a model called objectivated agility realisation model (OARM),
(Mali, 1978; Vrat et al., 1998) was identified suitable and applied in this study. The
OARM was developed to assess the consequences of how the hierarchy of performance
objectives can be achieved in coordinated manner by the sub-systems concerned. This
model facilitates identification of sub-systems, key performance areas (KPAs) in each of
the sub-systems, and their performance objectives, ranking and weighting of sub-systems
KPAs and performance objectives, calculation of effectiveness index (EI) for agility and
also the identification of sub-systems and the corresponding KPAs with low performance.
In the OARM approach, the EI for the system is built up in stages, involving the
sub-systems and the KPAs of the sub-systems. This model uses the systems approach and
considers an organisation in totality with the sub-systems interacting with each other.
This model also enabled an organisation to identify its weaker areas and helped to
develop suitable strategies to improve the performances of the weaker areas. These two
approaches are mainly procedure-based and do not make use of any optimisation
concepts. In order to fill this gap, two different optimisation approaches have been
identified applicable in this study. These are DEA (Charnes and Cooper, 1961) and GP.
These two approaches have been found to be very useful in dealing with various issues
related to agility.
The overall objective of any organisation is to maximise its agility. The agility of any
organisation will depend upon, how well each sub-systems and each KPA of a
sub-system perform both individually and collectively. Therefore the DEA model used
here has been built to reflect this feature. For this the optimum deployment of its input
resources and hence to maximise the output is essential. To deploy the resources
optimally we should know the (importance) weightages which each input and output
variable should receive. With the help of either AHP or OARM model it could not be
possible to obtain these weightages. The results obtained using AHP in terms of the
Importance Index of agility and helped identify the relative importance of each factor in
the view point of resources deployment (input). The results obtained using OARM in
terms of EI for agility helped assessing the relative importance of each enabler in the
view point of effectiveness (output) of the same. In order to attain 100% efficiency,
organisations should extract maximum possible output from the most effectively
deployed amount of input resources of each of its subunits. For this, various subunits
should be prioritised based on the attainable efficiency. However, the AHP and OARM
models when used jointly in the DEA-based model, (Yang and Kuo, 2002) helped to
obtain optimum weights for multiple input and output variables used in this study. The
results obtained using DEA clearly indicate where significant attention should be paid in
terms of input initiatives and output initiatives to realise the desired agility.
The results obtained using AHP, OARM and DEA were primarily used to compare
the 26 organisations in terms of their understanding as to what enables agility and also in
terms of their appreciation of to what extent they might have been able to realise agility.
These three methods rely very much upon the data/information that may be made
available. Besides, these three approaches are essentially in the category of diagnostic
 Application of goal programming to manage agility in manufacturing systems 225

methods. Therefore a need was felt to resort to an appropriate method, which may
provide proper guidelines for deploying the resources and also use a suitable method of
optimisation. As the objective of this paper is the systematic assessment of agility in
manufacturing systems, it has been found necessary to assess the degree of focus of
various enablers of AMS, which are being identified as significant by the AHP, OARM
and DEA models. The goal programming (GP)-model (Ignizio, 1976) was found useful in
this regard. The GP-based model incorporates the results obtained by the other three
models and tries to find out the degree of focus of those most significant enablers in our
companies with the available limited resources. The salient detail of this model is
presented in this paper.

2 Literature review

Agility is the successful exploration of competitive bases such as speed, flexibility,

innovation, quality, and profitability through the integration of reconfigurable resources
and best practices in a knowledge-rich environment to provide customer-driven
products and services in a fast changing market environment. A firm’s ability to
respond to competitive challenges and to sustain its competitive advantage is a key
element of success in today’s global marketplace (Porter, 1991; Prahalad and Hamel,
1990; Teece et al., 1997). The dynamic nature of market environments explains why
agility is an essential of a firm’s long-term success and survival. Being responsive is an
increasingly important skill for firms in today’s global economy; thus firms must be
agile. Ganguly et al. (2009), propose three matrices such as: market share matrices,
responsiveness metrics, and cost-effectiveness metrics for determining the enterprise
agility.
Many practical instances have proved that enterprise must have collaborative
capability with other enterprises within supply chain in order to succeed. The agile supply
chain is an operational strategy focused on inducing velocity and flexibility in a supply
chain. A key characteristic of an agile organisation is flexibility. In their study, Song
et al. (2007) tries to differentiate between agile supply chain and general supply chain.
According to them there are two differences between the agile supply chain and general
supply chain namely speed and flexibility. The speed emphasises the ability to respond
quickly to changing customers requirements, while flexibility emphasises the ability of
reconfiguring quickly according to the change. In order to achieve customer satisfaction
all physical and logical events within the supply chain to be enacted swiftly, accurately
and effectively. Today’s marketplace is characterised by intense competitive pressure as
well as high levels of turbulence and uncertainty. Organisations require agility in their
supply chains to provide superior value as well as to manage disruption risks and ensure
uninterrupted service to customers. Naturally, it follows that an organisation’s agility
depends on its supply chain being agile.
According to Swafford et al. (2008) achieving supply chain agility is a function of
other abilities within the organisation; specifically supply chain flexibility and
information technology (IT) integration. The results of their study indicate that
IT integration enables a firm to tap its supply chain flexibility which in turn results in
higher supply chain agility and ultimately higher competitive business performance.
The authors further argue that firms should invest first in IT for integrating information
226 P.G. Saleeshya and A.S. Babu

before investing in flexible processes. By considering the implications of agile

manufacturing and supply chain management, Gunasekaran et al. (2008), developed a
framework for responsive supply chain with the objective of achieving speed,
flexibility, and cost reduction. Goldman et al. (1995) and Yusuf et al. (2004), provide
insights into the relationship between IT integration, supply chain flexibility, supply
chain agility, and competitive business performance in a comprehensive manner.
Braunscheidel and Suresh (2009) in their study, supply chain agility, being external
facing, are regarded as a capability. In their view, cultivation of agility is a risk
management initiative that enables the firm to respond rapidly to market place
changes, as well as to potential and actual disruptions in the supply chain. Thus, agility is
of value for both risk mitigation and response. According to the authors flexibility
alone (Zhang et al., 2002; Swafford et al., 2006) is not enough for agility in the
supply chain. Internal and external integration are also necessary to ensure connected
and coordinated response to meet unforeseen changes. The impact of two
organisational characteristics, market and learning orientation that stem from
organisational culture is also explored in their research. Besides its effectiveness, supply
chain management is a complex process because of the stochastic and dynamic nature,
multi-criterion and ever-increasing complexity of supply chains. Further more companies
have realised that agility is essential for their survivel and competitiveness. Consequently
there is no generally accepted method by researchers or practitioners for designing,
operating and evaluating agile supply chains. Jain et al. (2008) developed a new approach
based on fuzzy association rule mining to support the decision makers by enhancing the
flexibility in making decisions for evaluating agility with both tangible and intangible
attributes.
At present, competence of manufacture is more than just in quality, price, date of
delivery and service of product; what is even more important is that the enterprises
response to the markets at speed. Accordingly, agile manufacturing has been one of the
major strategies of modern enterprise. Yang and Li (2002) argue that, in the environment
of market continuing to vary with high speed, mass customisation enterprise is required
to strengthen its ability of controlling future markets by improving its agility of product
manufacturing.
Reconfigurable manufacturing systems facilitate the demand for mass producing
custom products, which fall within the scope of agile manufacturing (Boxing et al.,
2006). Agile manufacturing provides the ability to accomplish rapid changeover
between manufacturing processes for different assemblies. Vazques-Bustelo and
Avella (2006), identified the most relevant factors for replacing traditional manufacturing
systems with agile production practices. According to the authors, all the attempts at
improving agility in the factories focused on four main areas: strategies, technologies,
organisation, and human resources. Correct integration of all of these reveals the
essence of the agile manufacturing model and becomes the most important challenge
for management this century. Agile manufacturing provides the ability to accomplish
rapid changeover between manufacturing processes for different assemblies. Yauch
(2007) analyses the team attributes necessary to facilitate agile manufacturing. Teams
operating within the context of agile manufacturing are characterised as multifunctional,
dynamic, cooperative, and virtual. The author also explores the opportunity exists to
improve the understanding of how teams should be designed and operated in order to
achieve manufacturing agility, including issues related to job satisfaction, team structure,
division of labour, inter-group relations, and inter-organisational relations. Workforce
 Application of goal programming to manage agility in manufacturing systems 227

agility is commonly described as a strategy that facilitates profitability in rapidly

changing and uncertain production environments. A lack of workforce agility has
been reported as one of the main reasons that some enterprises have difficulty in keeping
pace with markets and technical changes. Qin and Nembhard (2010) illustrate that the
real option (RO)-based workforce agility inherits the risk management capability of
options, and thus it reduces substantially the sensitivity of production quantity to market
risks, allowing manufacturers’ to rapidly and economically adapt to the unexpected
changes in market. They also demonstrate the advantages of RO-based agility such as
asymmetric attributes towards risk exposure and profitability, and the robustness to high
uncertainty.
It is quite evident from the study that researchers and practitioners are devoting
significant time, effort and resources in attempts to determine how enterprises of all
types can become sufficiently agile to compete in dynamic global markets. Some of
the important aspects which have not been given due importance in the published
literature are perhaps, method to form effective partnerships in an agile manufacturing
system (AMS) and method to evaluate the selection and performance of partners, and
also it is noticed that researchers and practitioners have not made significant attempts to
quantify the agility of manufacturing systems so far. To fill this gap suitable
mathematical models need to be developed and the authors have made an attempt in their
study.

3 Goal programming

GP usually results in a ‘compromise’ since the solution provides ‘satisfactory’ levels

of performance in terms of conflicting objectives, rather than identifying the ‘optimal’
solution with respect to a single objective. The GP model suggested here provides
flexibility to the decision maker in evaluating different alternatives. Famugiwa et al.
(2008) presented an integrated method for relating the compatibility of potential
partners in a strategic alliance. This method utilises a model based on fuzzy logic/GP
to analyse the vague, imprecise, and usually subjective information regarding
compatibility of potential suppliers that is available during the early formation of a
strategic partnership. Oddoge et al. (2009) and Arenas et al. (2002), explore the
link between GP and simulation for resource planning in various healthcare delivery
systems.
The studies related to the assembly line balancing problem have been a task of
interest to the industrial engineers and operational researchers for a number of years. A
GP approach would seem to be a natural modeling tool and a more realistic approach for
the assembly line balancing problems. Gokcen and Agpak (2006) suggested a GP model
for simple U-line balancing problem. Gokcen and Erel (1997) developed a binary GP
model to a mixed model assembly line balancing problem.
Badri (2001) presented a model which demonstrates how a combination of AHP and
GP has been used to identify the best strategies to manage quality of source of a super
market. In the present study a somewhat similar approach has been used to identify the
most important enablers of AMS. For this, the data collected was subjected to analysis
using three different methods viz. AHP, OARM and DEA and the results obtained are
then incorporated into the GP model.
228 P.G. Saleeshya and A.S. Babu

4 Application of GP

As reported earlier, the framework developed consists of various agility enablers in a

hierarchical form, and was used to collect data from five groups of industries, with the
help of pairwise comparison matrices. The data thus collected was subjected to analyses
using three different methods, viz. AHP, OARM and DEA. The results obtained helped
identifying the relative importance of each enabler at different levels, as indicated by the
respective measures. Using these measures, it is possible to rank each enabler at its
respective level. It may be noticed from Table 1 that the enablers with the top four ranks
only are considered in this study. The first column of this table indicates the levels of the
hierarchical framework, and the remaining columns contain enablers with respective
ranks identified by the four methods.
GP is a procedure for handling multiple-objective situations within the general
framework of linear programming (LP) as stated by Ignizio (1976). Each objective is
viewed as a goal. Then, given the usual resource limitations or constraints, the decision-
maker can attempt develop decisions that provide the best solution in terms of coming as
close as possible to reaching all goals. In the present study in order to improve agility, the
organisation needs to manage the enablers appropriately. The method GP has been
identified suitable to deal with this problem, for which the results obtained using AHP,
OARM and DEA are used. Each enabler would require various levels of resources and
would also result in different effectiveness. As shown in Table 1, four different sets of
factors have been obtained using AHP, OARM, DEA (input) and DEA (output). The
importance, existence and influence of these enablers in contributing to the agility of
manufacturing systems are indicated by the weights, indices and the ranks indicated
therein.
Since each method is associated with different principles, as expected there are also
variations in the importance with which the factors are identified with. For example, in
top level, the order of importance of enablers identified by AHP method is different from
that of OARM method. AHP identifies ‘SM’ as the most important enabler while OARM
gives least importance to SM. In the second level also the enablers identified by these two
methods show variations in their preferences. For example, AHP model has identified
leadership as the most important enabler under SM whereas OARM model does not
identify this factor among the top four ranks. The results obtained by OARM identify
‘integration’ as the most important factor. The DEA model does not identify does not
identify ‘integration’ among the four top ranks. In order to assess how important each of
these enablers is taking into account all the four aspects considered in Table 1, it was felt
necessary to identify each enabler with an overall score. For this purpose, appropriate
notional values were assigned to each rank position, a value of 20 to rank 1, 15 to rank 2,
10 to rank 3 and 5 to rank 4. Using the individual scores assigned to each enabler under
each ordered list, the total score with reference to each enabler was found and presented
in Table 2.
From Table 2, it can be noticed that each enabler has been identified with a score that
is supposed to indicate the achievement potential of each enabler in the point of view of
agility. Then, the important enablers identified by each of the four methods have been
individually assigned scores as shown in columns 3 to 6 of Table 2. These scores are used
to obtain a score representing the overall importance for each enabler as shown in
column 7.
 Application of goal programming to manage agility in manufacturing systems 229

Table 1 Most important enablers with respective ranks

Methods used
Level
AHP OARM DEA-input DEA-output
I 1 Strategic 1 OM 1 SM 1 SM
management (SM)
2 Tactical 2 TM 2 TM 2 TM
management (TM)
3 Operational 3 SM 3 OM 2 OM
management (OM)
SM-II 1 Leadership 1 Integration 1 Leadership 1 Leadership
2 Organisational 2 OS 2 OS 2 Integration
structure OS)
3 Customer centred 3 New tech 3 Customer 3 OS
paradigm absorption centred
paradigm
4 New technology 4 Customer 4 New tech 3 C.C. paradigm
absorption centred absorption 4 New Tech
paradigm absorption
TM-II 1 People 1 People 1 Partnership 1 People management
management management
2 Partnership 2 Partnership 2 People 2 Partnership
management
3 Value 3 LT 2 Response to 1 Value management
management, management competitive
lead-time Pressure
management
4 Response to 4 IT-based 3 Assurance 1 LT-management
competitive technology management 1 Response
pressure competitive
pressure
1 IT-based
technology
OM-II 1 Source 1 Source 1 Source 1 Process
management management management management
2 Process 2 Response to 2 Response to 1 Response to market
management market market conditions
conditions conditions
3 Response to 3 Process 3 Process 1 Response to
market conditions management management environmental
conditions
4 Delivery 4 Response to 4 Delivery 2 Source management
management environment management 2 Service
al conditions management
1 Delivery
management
230 P.G. Saleeshya and A.S. Babu

Table 2 Agility scores nos. in brackets stands for weights corresponding to ranks
Methods used
Level Factors AHP rank OARM rank DEA-input DEA-output
Total
and wt and wt rank and wt rank and wt
I Strategic management 1(20) 3(10) 1(20) 1(20) 70
Tactical management 2(15) 2(15) 2(15) 2(15) 60
Operational management 3(10) 1(20) 3(10) 2(15) 55
SM-II Leadership 2(15) 1(20) 1(20) 60
Integration 1(20) 2(15) 35
Organisation structure 2(15) 2(15) 2(15) 3(10) 55
Customer 3(10) 4(5) 3(10) 3(10 35
centred – paradigm
New tech. absorption 4(5) 3(10) 4(5) 3(10) 30
TM-II People management 1(20) 1(20) 2(15) 1(20) 75
Partnership 2(15) 2(15) 1(20) 2(15) 65
Value management 3(10) 1(20) 30
L-T management 3(10) 3(10) 1(20) 70
Res. to comp. pressure 4(5) 2(15) 1(20) 70
IT-based tech 4(5) 1(20) 25
Assurance management 3(10) 1(20) 30
OM-II Source management 1(20) 1(20) 1(20) 2(15) 75
Process management 2(15) 3(10) 3(10) 1(20) 55
Res. market conditions 3(10) 2(15) 2(15) 1(20) 60
Delivery management 4(5) 4(5) 1(20) 30
Res. environmental 4(5) 1(20) 25
conditions
Service management 1(20) 20
Note: *Agility score of SM = 20 + 10 + 20 + 20 = 70.
An organisation may find it more suitable to identify the extent to which each of the top
ranked enablers may have to be deployed in order to achieve a stipulated level of agility
and within the given level of resources available. The scores shown in Table 2 also act as
a good guideline in identifying appropriate levels of performance. Besides, the GP model
has the ability to indicate the extent to which the goals are either underachieved or
overachieved. This information will be quite useful for the organisation to carry out
follow up activities unlike in the case of LP, wherein, the only information available will
be the extent to which each enabler can be deployed to maximise agility. The sensitivity
analysis on GP will be much more useful to a manager than that on a LP model. The
details of how the GP model was used in the present study are given below.

5 The GP model

Using these scores in a GP model, it is possible to identify the most important enablers at
each level. For the GP-based model developed in the present study, only the top 12
enablers at level 2 shown in Table 3 have been considered. An attempt has been made to
develop a GP model to link the overall agility scores shown in Table 3 and the resource
constraints. In order to select the most significant enablers of AMS and to obtain the level
 Application of goal programming to manage agility in manufacturing systems 231

of focus of each of these enablers primary considerations have been given to four factors
such as the cost incurred, management hours to spend, the employee hours to spend and
the agility scores possessed by these enablers. Corresponding to these four factors the
following four goals have been established.
1 available cost Rs.7281
2 available management hour is 1,758 hrs
3 available employee hours is 1,050 hrs
4 the agility score attainable is 406.
Table 3 The top 12 enablers

Level Factors Agility score

I Strategic management 70
Tactical management 60
Operational management 55
SM (II) Leadership 60
Integration 35
Organisation structure 55
Customer centred paradigm 35
TM (II) People management 75
Partnership 65
Response to competitive pressure 70
Lead-time management 40
OM (II) Source management 75
Response to market conditions 60
Process management 55
Delivery management 30

Table 4 Variables used in GP model

Variable Factor
x1 Leadership
x2 Integration
x3 Organisation structure
x4 Customer centred paradigm
x5 People management
x6 Partnership
x7 Response to competitive pressure
x8 Lead-time management
x9 Source management
x10 Response to market conditions
x11 Process management
x12 Delivery management
232 P.G. Saleeshya and A.S. Babu

Table 5 Resource data – a sample

Resource Decision alternatives (significant enablers of AMS) Total

items budgeted
x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 resources
Estimated 1,000 700 980 700 1,300 1,100 750 750 1,300 1,000 980 650 7,286
cost (Rs.)
Management 280 150 220 150 300 275 200 200 300 280 220 130 1,758
hour (hr)
Employee 150 700 120 70 250 170 100 100 250 150 120 65 1,050
hours (hr)
Agility 60 35 55 35 75 65 40 40 75 60 55 30 406
scores

There are 12 variables x1, x2 … and x12 (representing significant enablers) shown in
Table 4 and a specific set of resources with constraints (goals) as shown in Table 5. The
resource items considered here are cost incurred, available management hours and
employee hours. GP permits resource limitations such as budgetary limitations, limited
hours of management and labour. For example, the unit cost required by these variables
x1, x2 … x12 respectively are 1,000, 700 … 650 and the total available cost (constraint) is
7,281. Similar consideration holds good for the other two constraints viz., management
hours and employee hours also. The values used for the agility scores and constraint are
also shown in this Table. Agility scores obtained by the variables x1, x2 … x12 are
respectively 60, 35 … 30, and the constraint (goals) value of agility score is taken as 406.
The significance of the unit costs given in respect of the variables x1 to x12 can be
explained as follows. If the value of x1 is one, then the cost incurred is Rs.1000. This
means that if the degree of focus of x1 is 1 then the system would incur a cost of Rs.1000,
implying that the enabler x1 is deployed to an extent equivalent to index value 1.
Similarly the other unit costs related to x2 to x12. Similar explanation holds goods for the
other two resource constraints also. However, some what a different explanation is in
order as far as the fourth constraint is concerned. If variable x1 takes values 1, then the
contribution of the corresponding enabler to the overall agility is 60. Similar explanation
holds good for the other variables x2 to x12 also.
It must be emphasised that all the values presented in Table 2 do not have any
significance to the organisations considered. These are merely random values used for the
sake of explaining the model.

6 Formulation

The basic approach of GP is to establish a specific numeric goal for each of the
objectives, formulate an objective function for each objective, and then seek a solution
that minimises the weighted sum of deviations of these objective functions from their
respective goals (Hillier and Lieberman, 2001).
The problem is to find x1, x2, x3, … x12, so as to minimise

( ) ( ) ( ) (
z = P1 d1− − d1+ + P2 d 2− d 2+ + P3 d3− d3+ + P4 d 4− d 4+ ) (1)

subject to constraints
 Application of goal programming to manage agility in manufacturing systems 233

1, 000 x1 + 700 x2 + 6,980 x3 + 700 x4 + 1,300 x5 + 1,100 x6 + 750 x7

(2)
+750 x7 + 750 x8 + 1,300 x9 + 1, 000 x10 + 980 x11 + 650 x12 + d1− − d1+ = 7, 286

280 x1 + 150 x2 + 220 x3 + 150 x4 + 300 x5 + 275 x6 + 200 x7 + 200 x8 + 300 x9

(3)
+280 x10 + 220 x11 + 130 x12 + d 2− − d 2+ = 1, 758

150 x1 + 70 x2 + 120 x3 + 70 x4 + 250 x5 + 170 x6 + 100 x7 + 100 x8 + 250 x9

(4)
+150 x10 + 120 x11 + 65 x12 + d3− − d3+ = 1, 050

60 x1 + 35 x2 + 55 x3 + 35 x4 + 75 x5 + 65 x6 + 40 x7 + 40 x8 + 705 x9
(5)
+60 x10 + 55 x11 + 30 x12 + d 4− − d 4+ = 406

xi ≥ 0, di − ≥ 0, di + ≥ 0 (6)

The objective function given by equation (1) attempts to minimise the overall deviations
in each of the goal constraints. There are five constraints as given by equations (2) to (6)
of which the first four are related to the goal constraints discussed above. d1, d2, d3 and d4
are respectively the deviation variables, associated with these constraints. The fifth
constraint stands for non-negativity whereas, P1, P2, P3 and P4 are the penalty weights
associated with the constraints one to four, for missing or not achieving the targeted goal.
For example, the first goal constraint given by equation (2) stands for the cost incurred.
The maximum cost that can be incurred is limited to Rs.7286. Suppose the solution of
this problem gives a value for cost incurred, which is more than Rs.7,286, then we could
say that we miss the goal for cost incurred. For missing this goal ‘P1’ is the penalty
weight, which is given in the objective function. Also the difference between the higher
value obtained for cost incurred and the limit kept (i.e., Rs.7,286) is referred to as the
deviation variable (d1) associated with the first constraint. Since this deviation is over
achievement, we are trying to reduce it. Similar consideration can be given to other
constraints also.

7 Solution methodology and results obtained

To link the agility score of the most important enablers with the resource limitations such
as estimated cost, available management hours and available employee hours and hence
to identify the degree of effectiveness of these factors in making an organisation agile, we
have used the GP model. The solution of this problem is generated using M.S. Solver
package.
For the sample data used in Table 5 the results obtained are presented in Table 6. The
12 variables used x1, x2 etc. and the values obtained for these variables are shown in
columns 1 and 2 respectively. The value for these variables is essentially an indication of
the degree of focus enjoyed by this variable in the respective organisations. The values in
column 2 are ranked as shown in column 3. The variable source management (x9) is
identified with the maximum degree of focus 0.801 and the variable delivery
management (x12) with the minimum value 0.477. In column 4 the mean rank values
obtained for the factors under strategic, tactical and OM are presented. This mean rank
value (5.5) indicates that the OM practices have achieved the maximum degree of focus
234 P.G. Saleeshya and A.S. Babu

in the respective organisations followed by tactical (6) and strategic (6.5) management
practices. Results presented also show the values obtained for all goal constraints V1, V2,
V3 and V4, the deviation associated with each constraint d1, d2, d3 and d4 and the
respective values of penalty P1, P2, P3 and P4. For example the total available value for
the cost constraint is Rs.7,286 and the amount utilised is obtained as Rs.7,021.61 only.
Therefore the value of deviation variable d1– or under achievement is 7,286–7,021.61,
which is equal to 264. Similarly the values of negative deviations (under achievement)
for constraints 2 to 4 are obtained as 58.96, 0 and 10.623 respectively. In case of penalty
weights P1, P2, P3 and P4, we have given penalty weight 8 for missing the cost constraint,
i.e., in case of over achievement and penalty weight 1, in case of under achievement.
Since the result obtained shows the under achievement of the goal and therefore the value
of penalty P1 is 1. Similar explanation can be given for the values of penalty weights P2,
P3 and P4. The value of objective function (z), which is the weighted sum of deviations
from their respective goals, is obtained as 585.274. Though we missed two goals (2 and
4) out of four goals, the deviation variables achieved with the missed goals are found to
be very less and therefore we can assume that the value of z obtained is optimum.
Table 6 Results obtained

(a) Decision variables

Variables used Values Ranks Mean rank
1 2 3 4
Leadership x1 .746 3 6.5
Integration x2 .492 8 Strategic management
Organisation structure x3 .558 7 practices
Customer centred paradigm x4 .492 8
People management x5 .789 2 6
Partnership x6 .639 4 Tactical management
Response to competitive pressure x7 .482 9 practices
Lead-time management x8 .482 9
Source management x9 .801 1 5.5
Response to market conditions x10 .621 6 Operational
Process management x11 .622 5 management practices
Delivery management x12 .477 10
(b) Deviations in resource limitation constraints
Constraints Usage Total available Deviation, d
Estimated cost (V1) 7,021.61 7,286 264 d1–
Management hours (V2) 1,699.03 1,758 58.96 d2–
Employee hours (V3) 1,050 1,050 0 d3
(c) Agility score constraints
Agility score (V4) Obtained Target Under achievement
395.37 406 10.623 d4–
(d) Penalties given
(8, 1) (10,4) (10,5) (8,1)
P1 = 1 P2 = 4 P3 = 5 P4– = 8
(e) Objective function z = 585.274
 Application of goal programming to manage agility in manufacturing systems 235

From the sample results the following interesting observations can be made:
a the OM practices have achieved the maximum degree of focus in AMS as it
possesses the maximum value of mean rank, followed by tactical and SM practices
b as the sub-system OM is identified as highly focused and hence the sub-factors of
OM can be identified as most significant enablers of an agile organisation
c under TM the degree of focus of the enablers identified are in the order of people
management, partnership, response to competitive pressure and lead-time
management
d under SM the degree of focus of the enablers identified are in the order of leadership,
organisation structure, integration and customer centred paradigm.

8 Sensitivity analysis

To identify the suitability and to check the sensitivity of the model, similar exercises have
been carried out with different sets of data.
Mainly two experiments have been carried out. In the first experiment the model is
tested by changing the penalty weights. In experiment one, the values of goals for one of
the cases, V1, V2, V3 and V4 obtained are 7,286, 1,730.08, 1,050 and 402.54 respectively.
Out of the four goals, the goals corresponding to cost and employee hours are achieved.
‘The goals’ which we missed are identified with very minimum deviation and therefore
the value of z achieved (139) is considered as optimum. Somewhat similar observations
can be made for the other cases also.
In the second experiment the sensitivity of the model is tested by changing the values
of goal constraints only. For constraints one to four the RHS values have been increased
by 30% of the base value and also decreased by 30%. These changes have been made
independently and hence for eight different cases the results were obtained. Analyses of
the results revealed many interesting facts. As per the changes in the values of goal
constraints changes in the values of all the variables achieved appear to be logical and
there is reason to believe that the GP model exhibits plausible behaviour.

9 Conclusions

In the present study the framework developed consists of various agility enablers in a
hierarchical form and was used to collect data from 26 manufacturing organisations. The
data collected was subjected to analyses using three different methods, viz. AHP, OARM
and DEA. These results on further analysis revealed that there is scope to apply a GP
model with the help of these three sets of results. Each enabler would require various
levels of resources and would result in different effectiveness. As shown in Table 1 four
different sets of factors have been obtained using AHP, OARM, DEA (input) and DEA
(output). In order to assess how important each of these enablers in taking into account all
the four aspects considered in Table 1 and the weights and ranks shown in Table 2, each
of the enablers is identified with an overall agility score. In order to select the most
significant enablers of AMS, these overall agility scores have been incorporated in the
236 P.G. Saleeshya and A.S. Babu

GP model the details of which is presented in this paper. The result obtained is an
indication of the degree of focus of these enablers in the respective organisations.

References
Arenas, M., Bilbao, A., Caballero, R. and Gomez, T. (2002) ‘Analysis via goal programming of the
minimum achievable stay in surgical weighting lists’, Journal of Operations Research Society,
Vol. 53, pp.387–3762.
Badri, M.A. (2001) ‘A combined AHP-GP model for quality control systems’, International
Journal of Production Economics, Vol. 72, pp.27–40.
Bottani, E. (2009) ‘A fuzzy QFD approach to agility’, International Journal of Production
Economics, Vol. 119, No. 2, pp.380–391.
Boxing, J.E., Bright, G. and Potgieter, J. (2006) ‘Reconfigurable manufacturing system for agile
manufacturing’, IFAC Conference, Proceedings, Vol. 12, pp.487–492.
Braunscheidel, M.J. and Suresh, N.C. (2009) ‘The organizational antecedents of a firm’s supply
chain agility for risk mitigation and response’, Journal of Operations Management, Vol. 27,
No. 2, pp.119–140.
Charnes, A. and Cooper, W.W. (1961) ‘Management models and industrial applications’, Linear
Programming, Vol. 1, II ed., John Wiley and Sons, New York.
Famugiwa, O., Monplaisir, L. and Nepal, B. (2008) ‘An integrated fuzzy-goal programming-based
framework for selecting suppliers in strategic alliance formation’, International Journal of
Production Economics, Vol. 113, pp.862–875.
Ganguly, A., Nilchiani, R. and Farr, J.V. (2009) ‘Evaluating agility in enterprises’, International
Journal of Production Economics, Vol. 118, No. 2, pp.410–423.
Gokcen, H. and Agpak, K. (2006) ‘A goal programming approach to simple u-line balancing
problem’, European Journal of Operational Research, Vol. 171, No. 2, pp.577–585.
Gokcen, H. and Erel, E. (1997) ‘A goal programming approach to mixed-model assembly line
balancing problem’, International Journal of Production Economics, Vol. 48, pp.77–85.
Goldman, S.L., Negel, R.N. and Preiss, K. (1995) Agile Competitions and Virtual Organization:
Strategies for Enriching the Customers, Von Nostrand Reinhold, New York.
Gunasekaran, A. (1998) ‘Agile manufacturing: enablers and an implementation framework’,
International Journal of Production Research, Vol. 96, pp.1223–1247.
Gunasekaran, A., Lai, K-H. and Cheng, E.T.C. (2008) ‘Responsive supply chain: a competitive
strategy in a networked economy’, Omega, Vol. 36, pp.549–564.
Hillier, F.S. and Lieberman, G.J. (2001) Introduction to Operations Research, 7th ed.,
McGraw-Hill, New York.
Huang, X.G., Wong, Y.S. and Wong, J.S. (2004) ‘A two-stage manufacturing partner selection
framework for virtual enterprises’, International Journal of Computer Integrated
Manufacturing, Vol. 17, No. 7, pp.294–304.
Ignizio, J.P. (1976) Goal Programming and Extensions, Lexington Books, D.C. Heath and
Company.
Jain, V., Benyoucel, L. and Deshmukh, S.G. (2008) Engineering Applications of Artificial
Intelligence, Vol.21, No. 3, pp.367–385.
Mali, P. (1978) Improving Total Productivity: MBO Strategies for Business, Government and not-
for-profit organizations, John Wiley and Sons, New York.
Oddoge, J.P., Jones, D.F., Tamiz, M. and Schmidt, P. (2009) ‘Combining simulation and goal
programming for healthcare planning in a medical assessment unit’, European Journal of
Operations Research, Vol. 93, pp.250–261.
Porter, M.E. (1991) ‘Toward a dynamic theory of strategy’, Winter, Strategic Management
Journal, Vol. 12, pp.95–117.
 Application of goal programming to manage agility in manufacturing systems 237

Prahalad. C. and Hamel, C. (1990) ‘The core competence of cooperation’, Harvard Business
Review, May–June, pp.79–91.
Qin, R. and Nembhard, D.A. (2010) ‘Workforce agility for stochastically diffused conditions – a
real option perspective’, International Journal of Production Economics, Vol. 125, No. 2,
pp.324–334.
Saaty, T.L. (1980) The Analytic Hierarchy Process, McGraw Hill, New York.
Saleeshya, P.G. and Babu, A.S. (2003) ‘A multilevel modeling framework to assess the agility in
manufacturing systems’, 20th AIMTDR Conference Proceedings, pp.682–690.
Song, M., Rul-Xue, F.U., Zhu, C. and Xin, Z-H. (2007) ‘Study on the agile supply chain
management based on agent’, The journal of China Universities of Posts and
Telecommunications October 1,, Vol. 14, Supplement.
Steenhuis, H-J. and de Boer, S. (2003) ‘Agile manufacturing and technology transfer to
industrializing countries’, International Journal of Technology Management, Vol. 26, No. 1,
pp.20–27.
Swafford, P.M., Ghosh, S. and Murthy, N. (2008) ‘Achieving supply chain through IT integration
and flexibility’, International Journal of Production Economics, Vol. 116, pp.288–297.
Swafford, P.M., Gosh, S. and Murthy, N. (2006) ‘The antecedence of supply chain agility of a firm:
scale development and model testing’, Journal of Operations Management, Vol. 24, No. 2,
pp.170–188.
Teece, D.J, Pisano, G. and Shuen, A. (1997) ‘Dynamic capabilities and strategic management’,
Strategic Management Journal, Vol. 18, No. 7, pp.509–533.
Vazques-Bustelo, D. and Avella, L. (2006) ‘Agile manufacturing: industrial case studies in Spain’,
Technovation, Vol. 26, pp.1147–1161.
Vrat, P., Sardana, G.D. and Sahag, B.S. (1998) Productivity Management – A Systems Approach,
Narosa Publishing House, New Delhi.
Yang, S.L. and Li, T.F. (2002) ‘Agility evaluation mass customization product manufacturing’,
Journal of Material Processing Technology, Vol. 129, pp.640–644.
Yang, T. and Kuo, C. (2002) ‘A hierarchical AHP/DEA methodology for the facilities layout
design problem’, European Journal of Operations Research, Vol. 141, pp.128–136.
Yauch, C.A. (2007) ‘Team-based work and work system balance in the context of agile
manufacturing’, Applied Ergonomics, Vol. 38, pp.19–27.
Yusuf, Y.Y., Gunasekaran, A., Adeleye, E.O. and Sivayoganatham, K. (2004) ‘Agile supply chain
capabilities: Determinants of competitive objectives’, European Journal of Operation
Research, Vol. 159, pp.379–392.
Zhang, Q., Vonderembs, M.A. and Lim, J. (2002) ‘Value chain flexibility: a dichotomy of
competence and capability’, International Journal of Production Research, Vol. 40, No. 3,
pp.561–583.

View publication stats