International Journal of Systems Science

Vol. 39, No. 12, December 2008, 1193–1202

Evaluating flexibility on order quantity and delivery lead time for a supply chain system
Yi-Chi Wang*
Department of Industrial Engineering and Systems Management, Feng Chia University, Taichung, Taiwan
(Received 9 November 2006; final version received 12 June 2008)

Most of the previous literature on production flexibility is centred on the flexibility of manufacturing systems.
However, the manufacturing system is just one of several key components of a supply chain. A supply chain is
a network involving all of the activities within individual organisations that link material suppliers,
manufacturing factories, distributors, warehouses, retailers and customers. Research into the flexibility of
a supply chain therefore extends from the intra-organisational flexibilities to the inter-organisational flexibilities.
This article provides a study of examining two aspects of supply chain flexibility: order quantity flexibility and
lead time flexibility, which have been clarified as the two most common changes which occur in supply chains.
Order quantity flexibility refers to the ability to provide proper order quantity for customer needs. Lead time
flexibility allows customers to set the order due date depending on their needs. A simulation model is built to
evaluate the performance on different flexibility levels of a supply chain. The experimental results provide
interesting insights and can be applied in selecting suppliers with order quantity flexibility and delivery lead time
flexibility.
Keywords: supply chain flexibility; volume flexibility; delivery lead time flexibility

1. Introduction chain flexibility has been conspicuous by its absence

Given the increased level of global competition, (Barad and Sapir 2003; Aprile, Garavelli, and
manufacturing enterprises must be capable of making Giannoccaro 2005; Sánchez and Pérez 2005). Any
customised high-quality products with shorter life individual firm in a supply chain system must be able
cycles under the circumstances of revised regulations to respond rapidly to any change from its partners or
and rapid market changes. Flexibility, which made it its own customers. Such capability has become
possible for organisations to survive and remain a requirement for surviving in today’s competitive
competitive advantages by implementing the new environment. Flexibility has been defined as an
computerised technology, has often been proposed as adaptive response to unpredictable situations (Gupta
the solution. Since the first flexible manufacturing and Goyal 1989). A more comprehensive definition
system was successfully installed in the 1960s (Groover might be the ability to change or react to unpredictable
2001), abundant researches have been focussed on the events with little penalty in time, effort, cost or
flexibility of manufacturing systems (Sethi and Sethi performance (Beach, Muhlemann, Price, Paterson,
1990; DeToni and Tonchia 1998). However, with and Sharp 2000). Flexibility in a supply chain therefore
a global perspective, improving flexibility only on the is a good representation of the potential capability to
production systems of a manufacturing enterprise may improve the company’s efficiency, and can be
not lead to competitiveness. Instead, appropriately a significant measure for evaluating a supply chain.
implementing flexibility in supply chain logistics and This study examines two aspects of supply chain
management may well represent an important source flexibility: order quantity flexibility and delivery lead
of competitive advantage. A supply chain can be time flexibility. Order quantity flexibility refers to the
viewed as a system which involves all of the activities ability of providing proper order quantity for customer
within individual firms that link material suppliers, needs. Delivery lead time flexibility allows customers
manufacturing factories, distributors, retailers and to set the order due date depending on their needs.
customers. The manufacturing system is just one of Order quantity and lead time are frequently found to
the several elements in the chain. However, contrary be the cause of supplier-buyer grievance (Das and
to manufacturing flexibility, which has been widely Abdel-Malek 2003). They likely impact customers’
studied, it is quite noteworthy that research on supply perceptions and cause out-of-stock conditions for

*Email: [email protected]

ISSN 0020–7721 print/ISSN 1464–5319 online

ß 2008 Taylor & Francis
DOI: 10.1080/00207720802298939
http://www.informaworld.com
1194 Y.-C. Wang

products that are suddenly in high demand. In Das and for managers to evaluate and change the flexibility of
Abdel-Maleks’ model, the authors considered a make- their operations. Koste and Malhotra (1999) outlined
to-order case in which the supplier does not have ready 10 dimensions of manufacturing flexibility and identi-
stock for shipping and inventory cost incurred is fied four elements of flexibility, and proposed
ignored. This article is centred at a make-to-stock case a hierarchical relationship for the flexibility dimen-
in which the upstream supplier is also a manufacturer sions. Their proposed framework appears to be
who produces products and stocks them as inventory a promising approach for developing measures for
until they are sold. A discrete-event simulation model other flexibility dimensions. Their framework was later
in this study is employed to evaluate the performance empirically tested and validated by Koste, Malhotra,
of a supply chain under different levels of these two and Sharma (2004). Scala, Purdy, and Safayeni (2006)
flexibilities. Results indicate that order quantity flex- presented a conceptual model in which cybernetics
ibility may offer great cost savings when the holding concepts were applied to manufacturing flexibility.
cost dominates the total cost, and delivery lead time His model can be used to clarify certain aspects of
flexibility is able to improve the service level and system flexibility and to measure variety regarding the
effectively reduces the possibility of shortages transformation of technology flexibility into manufac-
occurring. turing flexibility. In those studies, the researchers have
The purpose of this article is to provide a detailed made a great effort to propose or construct conceptual
quantitative analysis on two key dimensions of models or frameworks for clarifying, implementing,
flexibility in a supply chain. The remainder of this and measuring certain manufacturing flexibilities.
article is organised as follows. Section 2 reviews the On the other hand, some researchers have provided
literature on flexibility in manufacturing systems. well-organised surveys for the research of manufactur-
Section 3 extends this discussion and addresses the ing flexibility. Gupta and Goyal (1989) reviewed
key elements of flexibility in a supply chain. Section 4 around 30 research studies on flexibility in manufac-
presents the simulation model employed in this study turing systems. They narrowed down the studies
and describes the experiments. Section 5 provides carried out to date to eight types of flexibility, and
results and discussion. Finally, in Section 6 we draw classified flexibility measures as qualitative and quan-
our conclusions as to the findings of this study and titative, and theoretical and non-theoretical. Sethi and
discuss directions for further research. Sethi (1990) conducted a comprehensive survey for the
issue of manufacturing flexibility. They discussed 11
different flexibilities and analysed the linkages between
2. Flexibility in manufacturing systems the various flexibilities. They also reviewed several
A considerable amount of research studies have been analytical models and empirical studies regarding these
carried out in the area of manufacturing flexibility for flexibilities. Another excellent reference regarding
the past several years, but they all seem to be quite flexibility of manufacturing systems was provided by
fragmented. Each individual work might be focused on Das (1996). He proposed a theoretical basis for
certain aspects of manufacturing flexibility, with some measuring the flexibility of manufacturing systems
overlapping the work of others and some being and introduced multiple levels for each type of
aggregates of others. Upton (1994) presented flexibility. De Toni and Tonchia (1998) provided
a framework which has been applied in reality by a comprehensive literature review and proposed
several companies seeking ways to improve their a scheme for classifying six aspects of manufacturing
flexibility. His framework allows managers to deter- flexibility: definition of flexibility, request for flex-
mine and distinguish what is required of manufactur- ibility, classification in dimensions of flexibility,
ing operations, and provides a common base of measurement of flexibility, choices for flexibility and
understanding, so that all involved know what is to interpretation of flexibility. Anand and Ward (2004)
be improved. Suarez, Cusumano, and Fine (1996) surveyed 101 manufacturing firms and concluded
conducted an empirical study of manufacturing flex- that flexibility is a stronger predictor of performance
ibility using data from 31 printed circuit-board plants in more dynamic environments. They focused on
in Europe, Japan and the United States. They also dynamic environments and two specific aspects of
proposed a framework focussing on the measurement, dynamism: unpredictability and volatility.
implementation, and evaluation of three flexibility
types: mix, volume and new-product flexibility. Gerwin
(1993) proposed a conceptual model serving as 3. Flexibility in a supply chain
a foundation for work on the strategic aspects of Vickery, Calantone, and Droge (1999) conducted an
manufacturing flexibility. His model provides a means empirical study focused on supply chain flexibility.
 International Journal of Systems Science 1195

They surveyed 65 firms in the furniture industry and indicated that a high degree of flexibility may not be
examined the relationships between five identified better than a suitable but limited flexibility. Das and
flexibility types and the environmental uncertainty. Abdel-Malek (2003) defined the supply chain flexibility
In their study, they defined quantity flexibility, the as the elasticity of a buyer–supplier relationship under
same as commonly defined in the literature for changing conditions. They utilised the supply chain
manufacturing systems—the ability to effectively flexibility to measure the suppliers and further devel-
increase or decrease aggregate production in response oped a supplier selection model. Tables 1 and 2 present
to customer demand. Sánchez and Pérez (2005) the types of supply chain flexibilities and their related
explored the relationship between the dimensions of literature, respectively. Among these types of flexibil-
supply chain flexibility and company performance in ity, order quantity flexibility and lead time flexibility,
the automotive industry. They clarified 10 dimensions the most common needs in a supply chain (Das and
in supply chain flexibility and found that these Abdel-Malek 2003) are analysed in the present study.
dimensions are not equally important for company The kinds of benefits a supplier with various levels of
performance. They also pointed out that companies such flexibility can offer are measured as well.
normally focus too much on basic flexibility capabil-
ities (at the manufacturing system level), and neglected
the impact of aggregate flexibility capabilities (at the 3.1. Order quantity flexibility
supply chain level). Their study, however, reveals that Quantity flexibility offered by the upstream supplier
the aggregate flexibility capabilities are more positively usually refers to their ability to adjust internal
related to company performance than the basic production or replenishment policy to meet the down-
flexibility capabilities. stream buyer’s needs. However, flexibility on order
Barad and Sapir (2003) compared the substantial quantity may mean different things. Some suppliers
amount of literature on manufacturing flexibility and may request a minimum quantity for an order or set
suggested a framework for flexibility types in logistic a certain quantity no matter the amount their customer
systems. They quantitatively analysed the effect of the really needs, while others may allow the buyers to
trans-routing flexibility factor on logistics dependabil- change their order quantity within a certain period. Das
ity, a new performance measure proposed for a logistic and Abdel-Malek (2003) defined a measure by which
system. Garavelli (2003) evaluated nine supply chain the buyer is able to estimate the flexibility of the
configurations using a simulation model. He concluded suppliers. In their study, the relationships between
that the concept of limited flexibility, a particular buyers and suppliers are governed by contracts in which
supply chain configuration, can yield many benefits the buyer guarantees the supplier a minimum order
without increasing the flexibility costs. His findings quantity. Based on such an order commitment, when an

Table 1. Types of supply chain flexibility.

No. Definitions

1 The ability to produce or deliver highly customised products.

2 The ability to adjust production quantity in response to customer demand.
3 The ability to rapidly introduce new products to the market.
4 The ability to provide widespread or intensive distribution coverage.
5 The ability to respond to the needs or changes of the target markets.
6 The ability to find another supplier for each specific component or raw material.
7 The ability to adjust lead times to meet customer requirements.
8 The ability to process a part through varying routes by using flexible transporting networks.
9 The ability to move stock between the demand locations and the supply locations without incurring high costs.
10 The ability of keeping products in their generic form as long as possible to incorporate the customer’s product
requirements in the later stages.
11 The ability to configure assets and operations to react to emerging customer trends (product changes, volume,
mix) at each node of the supply chain.
12 The ability to reconfigure the supply chain, altering the supply of product in line with customer demand.
13 The ability to align labour force skills to meet customer service/demand requirements.
14 The ability to align information system architectures.

Notes: 1: Product (Mix, Process); 2: Volume (Quantity); 3: Launch; 4: Access (Distribution, Logistics); 5: Responsiveness
or target markets (Market); 6: Sourcing; 7: Delivery (Lead time); 8: Routing; 9: Trans-shipment; 10: Postponement;
11: Operations system (1 þ 2); 12: Supply; 13: Organisational; 14: Information systems.
1196 Y.-C. Wang

Table 2. Literature related to supply chain flexibility.

Supply chain flexibility dimensions

Literature 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1 Garavelli (2003)    
2 Aprile et al. (2005)    
3 Barad and Sapir (2003) 
4 Chan and Chan (2006) 
5 Garavelli (2003)  
6 Das and Abdel-Malek (2003)  
7 Duclos, Vokurka, and Lummus (2003)      
8 Sánchez and Pérez (2005)          
9 Vickery et al. (1999)     

*Represents the supply chain flexibility studied in the literature.

order is below the minimum quantity, then a penalty is Table 3. Contracts for order quantity flexibility.
imposed. Such a contract provides revenue protection
for the supplier. Aprile et al. (2005) carried out an Rules (Intended order quantity ¼ Q)
analysis of comparing different supply chain flexibility QF-A Order quantity ¼ Q
configurations subject to demand and capacity varia- (No batch-sized
bility. They set the market demand with various constraint on
probability distribution functions. Chan and Chan order quantity)
(2006) developed an agent-based supply chain model in QF-B If (Q 5 30)
which quantity flexibility is mutually agreed upon a Order Quantity ¼ 30;
range of quantities between a retailer and a supplier. Else if (30  Q 5 60)
Under this type of contract the supplier can ship Order Quantity ¼ 60;
Else if (60  Q 5 90)
a quantity within the agreed range to the buyer, instead Order Quantity ¼ 90;
of a fixed quantity. In their research, order quantity is Else if (60  Q 5 120)
determined based on a stochastic model with the Order Quantity ¼ 120;
assumption in which the retailer knows its parameter Else if (120  Q 5 150)
of the demand function. The quantity range is set Order Quantity ¼ 150;
Else if (150  Q 5 180)
through a coordination mechanism between the retailer
Order Quantity ¼ 180;
and the supplier. In this study, three ordering rules Else if (Q  180)
representing different flexibility levels of order quantity Order Quantity ¼ 200.
are presented in Table 3. The buyer may intend to order QF-C If (Q 5 50)
Q units. In Policy QF-A, the supplier does not set any Order Quantity ¼ 50;
constraint regarding to order batch size. In other Else if (50  Q 5 100)
words, the supplier takes any-sized order. Contrary to Order Quantity ¼ 100;
the complete flexibility of QF-A, Policies QF-B and Else if (100  Q 5 150)
Order Quantity ¼ 150;
QF-C are both limited in flexibility. For Policy QF-B, Else if (Q  150)
the supplier sets an ordering rule in which the order Order Quantity ¼ 200.
quantity must be a multiple of 30 units when the
desired order quantity is under 200. Whereas as
applying Policy QF-C, the buyer must order in a faster than usual manner, then a price penalty
a multiple of 50 units when the intended order quantity is imposed. In the agent-based model of Chan and
Q is less than 200. Chan (2006), the range of delivery due dates is deter-
mined through a coordination mechanism between
supplier and retailer. In the simulated supply chain
3.2. Delivery lead time flexibility model of this study, we suppose that the lead time for
Like quantity flexibility, flexibility on delivery lead a regular order is 7 days and that the supplier offers
time can be discussed from various aspects. In the options for ‘hot’ orders with a shorter lead time.
model proposed by Das and Abdel-Malek (2003), there The ‘hot’ orders with lead time less than 7 days carry
is a minimum delivery lead time for the supplier to extra cost. The buyer then needs to carefully decide
ship orders. When orders are requested to be delivered how ‘hot’ the order is. Three rules representing various
 International Journal of Systems Science 1197

Table 4. Contracts for lead time flexibility. order-point, order-up-to-level) system (Silver, Pyke,
and Peterson 1998). The (R, s, S ) policy describes
Total quantity of committed that every R units of time the buyer checks the
order ¼ CQ
Rules Reorder point ¼ s instantaneous inventory level. If it is at or below the
reorder point s, the buyer orders enough items to raise
LTF-A If (CQ 5 0.2  s) it to S. If the inventory level is above s, nothing is done
Delivery lead time ¼ 7 days; until at least the next review. For example,
Else if (0.2  s  CQ 5 0.5  s) a replenishment policy (20, 25, 100) is that the inven-
Delivery lead time ¼ 5 days;
Else if (0.5  s  CQ 5 1.0  s) tory level is reviewed every 20 days. If the inventory
Delivery lead time ¼ 3 days; level is determined at or below 25 units, the buyer will
Else if (CQ  1.0  s) order enough quantity to raise it to 100 units.
Delivery lead time ¼ 1 day. Otherwise, no order is placed by the buyer. It has
LTF-B If (CQ 5 0.5  s) been pointed out that, under general assumptions
Delivery lead time ¼ 7 days; concerning the demand pattern and the cost factors
Else if (CQ  0.5  s) involved, the best (R, s, S ) system is able to achieve
Delivery lead time ¼ 3 days. a lower total cost than any other system (Silver et al.,
LTF-C Delivery lead time ¼ 7 days (Fixed lead time) 1998). When the buyer does not have enough stock for
a customer’s demand, a shortage occurs and the
unfilled demand is backordered and will be delivered
levels of lead time flexibility for determining the due with the next shipment. In this case a partial shipment
dates of orders are presented in Table 4. is not allowed by the buyer. A partial shipment concept
In this study, when the buyer takes orders from his is that the supplier would ship on-hand stock even
downstream customers, these orders are committed to though it can only partially fill a buyer’s demand,
be delivered, but not necessarily to be shipped yet. rather than having the buyer having to wait for the
The shipping date is determined based on the due date entire order ready to be delivered (Banerjee, Banerjee,
set by the customers. Every time the buyer places an Burton, and Bistline 2001).
order for replenishing his stock, he will take into The purpose of this study is to demonstrate the
account the quantity of the committed orders on-hand. influence of the flexibility levels of order quantity and
Three rules representing different flexibility levels of delivery lead time. Testing involves using discrete-event
delivery lead time are presented in Table 4. Among simulation to model a three-echelon supply chain
the three rules, LTF-C does not offer any flexibility. consisting of a manufacturer, a buyer and several
(The delivery lead time is exactly 7 days). LTL-B offers customers. The simulation model is developed in Cþþ
two options for the buyer (3 days and 7 days). LTL-A programming language and implemented on a personal
allows the buyer to decide the delivery lead time when computer with an Intel Pentium 4 3.2 GHz CPU.
placing an order. It is noted that here we used the The elements in the supply chain such as a supplier/
reorder point (s) as a measure to determine the level of manufacturer, a buyer/retailer and a customer, are all
the ‘hot’ order. For instance, if the total quantity of built as modules in the system. All the activities
committed orders is greater than the reorder point, the regarding to each supply chain element are modulised.
buyer should realise that the committed orders may Each module is programmed as an object in Visual
result in a shortage if the replenishment takes too long. Cþþ. With such modulised elements, this simulation
In order to avoid a shortage, the buyer must ask the model can easily be extended as many stages and many
supplier to deliver as soon as possible, and decides on elements in the stage of the supply chain as user’s
option LTF-A, in which case the delivery lead time is needs. Compared to Das and Abdel-Mdlek’s mathe-
reduced to one day. matical model, which was only used to obtain a simple,
closed-form solution for the supply chain flexibility,
this discrete-event simulation model logically mimic
the interactions within the elements of the supply chain
4. Simulation modelling and can be used to see how the output measures of
In this study, we consider a scenario consisting of performance are affected by exercising numerically the
a manufacturer producing and delivering a product on model for the inputs. This simulation model and its
receiving an order from a buyer (a store or a retailer) results have also been verified and validated by the
who sells it under stochastic customer demand. techniques recommended by Law and Kelton (2000).
The manufacturer and the buyer both check their In this simulation model, the buyer takes orders
inventory status periodically. The replenishment policy directly from the customers. The time between
has been denoted as a (R, s, S ) policy–(periodic-review, individual customer demand arrivals follows an
1198 Y.-C. Wang

exponential distribution. The quantity of each custo- Table 5. The replenishment policies for seven test problems.
mer’s demand is uniformly distributed between
15 units and 25 units. The due date for each customer Mean demand Demand inter-arrival Policy
Case per period time (days) (R, s, S )
demand is set with a uniform distribution between
1 day and 10 days after the demand is received. 1 10 40 (20, 6, 40)
The buyer is asked to fill the customer demands 2 20 20 (20, 14, 62)
exactly at their due dates. The interval for checking the 3 30 13.3 (20, 23, 66)
4 40 10 (20, 33, 87)
inventory level is 20 days. The unfilled demands are 5 50 8 (20, 42, 108)
backordered and will be delivered right after the next 6 60 6.7 (20, 52, 129)
replenishment. After taking an order from the buyer, 7 70 5.7 (20, 62, 81)
the manufacturer/supplier immediately ships the order
to the buyer if there is enough stock for the order.
The delivery lead time for a regular order is 7 days,
but the delivery time for a backorder is 5 days.
each system parameter setting. Since the system
The manufacturer employs the replenishment policy
involves a buyer and a manufacturer, the follow-
(R, s, S ) – (100, 150, 800) to manage the inventory
ing system performance measures are selected for
stock and to set the production schedule. If the
evaluating order quantity flexibility and lead time
inventory position is below 150 when reviewing the
flexibility:
stock, the manufacturer makes products up to 800.
The manufacturing lead time is 20 days. Some of the . Total ordering cost (OC): the total ordering
above data are from one of the largest machine tool cost is calculated by multiplying the total
providers in Taiwan who offers comprehensive range number of orders placed by the buyer by the
of metal working machineries and their accessories. ordering cost per order.
The machine tool industry is chosen in this study . Total holding cost (HC): the total inventory
because it is one of the largest manufacturing activities cost can be obtained by multiplying the
in the world and it is economically significant to many average inventory level by the simulation
countries. This machine tool provider has complained duration by holding cost rate.
about his upstream suppliers and has been searching . Total shortage cost (SC): the total inventory
for an upstream supplier who is able to offer better cost can be obtained by multiplying the total
options on some flexibility dimensions. number of stock-out occurrences by the
penalty cost per backorder.
. The service level (SL): the fraction of customer
5. Performance analysis demands to be met without backorders.
In order to avoid the impact of various replenishment For order quantity flexibility, a larger quantity than
policies adopted by the buyer, we have chosen seven desired is ordered by using QF-B as well as QF-C
cases from the test problems used in the study of Zheng (especially QF-C) due to its lesser flexibility. The buyer
and Federgruen (1991) since the parameters for the is forced to order more quantity than is actually
optimal replenishment policies were found in these needed. Under the same rate of customer demands,
cases. The following parameters are common to all ordering more quantity means holding more items in
seven cases. The ordering cost is $24 per order. stock and placing fewer orders (incurring less ordering
The holding cost rate is $1 per unit per day. cost). These conditions can be observed in Figures 1
The penalty cost rate is $9 per backorder. The and 2. With more quantity in stock, the number of
replenishment policies based on the optimal algorithm shortage occurrences decreases. QF-C shows the
of Zheng and Federgruen for each of these seven cases advantages on lower shortage costs and higher service
are presented in Table 5. The seven cases are chosen for level in Figures 3 and 4. Therefore, there is a trade-off
examining the system performances under various for such order quantity flexibility. It depends on the
system loading levels. cost per shortage, cost per order and holding cost per
A horizon of 10,000 customer-demand completions unit per unit time. When the unit holding cost is high
was determined as an appropriate run length to compared to the other two costs, the flexibility of order
guarantee that the system had reached steady-state. quantity provides significant saving benefits on hold-
After completing these 10,000 demands as a system ing cost since the buyer does not need to order extra
warm-up, 20,000 additional demands are processed by quantity. Although QF-C outperforms the other two
the system, and statistics are collected and calculated. rules on the measure of shortage cost, it may not be so
The system is conducted with 10 replications under attractive to an inventory manager because improving
 International Journal of Systems Science 1199

350000 0.8

300000 0.7

250000 0.6
Total ordering cost

0.5

Service level
200000
0.4
150000
0.3
100000
QF-C 0.2 QF-C
50000 QF-B QF-B
QF-A
0.1
QF-A
0
1 2 3 4 5 6 7 0
1 2 3 4 5 6 7
Case
Case
Figure 1. Total ordering cost (quantity flexibility).
Figure 4. The service level (quantity flexibility).

25000000 350000

300000
20000000
250000
Total ordering cost
Total holding cost

15000000 200000

150000
10000000
100000
QF-C LTF-B
5000000
QF-B 50000 LTF-A
QF-A LTF-C
0 0
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Case Case

Figure 2. Total holding cost (quantity flexibility). Figure 5. Total ordering cost (lead time flexibility).

140000 18000000
16000000
120000
14000000
Total holding cost

100000 12000000
Total shortage cost

80000 10000000
8000000
60000
6000000
40000 4000000 LTF-B
QF-C LTF-A
2000000
20000 QF-B LTF-C
QF-A
0
1 2 3 4 5 6 7
0 Case
1 2 3 4 5 6 7
Case Figure 6. Total holding cost (lead time flexibility).
Figure 3. Total shortage cost (quantity flexibility).

it is noted that LTF-C achieves less holding cost than

service level can be achieved by raising the values of the other two because the buyer has no option of early
order point (s) and order-up-to level (S ). delivery and therefore holds less inventory on average.
For lead time flexibility, three rules do not seem to Figures 7 and 8 show some differences on the measures
be significantly different in Figures 5 and 6. However, of shortage cost and service level. The main idea of
1200 Y.-C. Wang

LTF-A and LTF-B is that the buyer takes into shortage occurrences. Therefore, if the cost per stock-
account the on-hand committed orders. Those com- out occasion is high, the lead time flexibility holds the
mitted orders must be shipped to the customer in the edge in reducing the possibility of a shortage occurring
near future. A longer lead time may likely result in and so improving the service level.
Table 6 presents the comparison of LQ-C/LTF-C
and LQ-A/LTF-A, the least and the most flexible
140000 combination rules, respectively, on various perfor-
120000
mance measures for the buyer. All the values in
this table are calculated by comparing the performance
100000 of the system with LQ-C/LTF-C and with LQ-A/
Total shortage cost

LTF-A, which is used as the basis. For instance,

80000
the value on row two and column two in Table 6 is
60000 23.37%, which represents that the holding cost will
C
drop 23.37% when the least flexible combination rule
40000
LTF-B (LQ-C/LTF-C) employed by the supplier is replaced
20000 LTF-A by the most flexible combination rule. By finding
LTF-C the average value for each performance measure
0
1 2 3 4 5 6 7 from Case 1 through Case 6, except the extreme case
Case (Case 7), if the most flexible combination rule is
Figure 7. Total shortage cost (lead time flexibility). employed instead the least one, the holding cost drops
27.37%, the ordering cost increases 38.96%, the
shortage cost increases 22.26%, and the service level
drops 9.01% on average. The values again demonstrate
0.8 the fact that flexibility provides the benefits for
0.7 lowering the holding cost but deteriorates the other
three measures. However, the total cost summed the
0.6
holding cost, ordering cost and shortage costs together
Service level

0.5 actually drops 26.65%. That is, in this case, placing

0.4 precise order quantity leads to lower holding cost
which dominates the other two costs. For Case 7, the
0.3
extreme case, customer demands arrive according to an
0.2 LTF-B exponential distribution with a mean of 5.7 days.
0.1 LTF-A The service level is only around 53.68%. With order
LTF-C
0
quantity flexibility and lead time flexibility, the service
1 2 3 4 5 6 7 level drops to 30% which may not be acceptable for
Case
any members of a general supply chain system. These
Figure 8. The service level (lead time flexibility). numbers reveal that attention to must be paid on the

Table 6. Comparison of QF-C/LTF-C and QF-A/LTF-A on various measures for the buyer.

Measure Case 1 (%) Case 2 (%) Case 3 (%) Case 4 (%) Case 5 (%) Case 6 (%) Case 7 (%)

HC 23.37 39.09 36.11 26.11 20.75 18.8 61

QC þ57.89 þ55.21 þ50.73 þ33.74 þ18.5 þ17.66 þ7.91
SC þ26.65 þ38.06 þ28.12 þ16.09 þ13.8 þ10.83 þ49.24
TC 24.47 38.32 35.08 25.42 20.23 18.35 57
SL 18.24 12.12 9.87 4.92 4.74 4.69 42.49

Table 7. Comparison of QF-C/LTF-C and QF-A/LTF-A for the supplier/manufacturer on service level (%).

Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7

QF-A/LTF-A 99.99 99.33 98.03 95.45 92.39 90.05 91.83

QF-C/LTF-C 99.98 99.79 98.78 96.47 93.48 91.08 92.89
 International Journal of Systems Science 1201

original service level of an organisation when the Machine Tools and Manufacture, and Computers and
Industrial Engineering.
managers of the organisation are thinking about
positioning flexibility for benefit.
Table 7 shows the service levels for the manufac-
turer/supplier under these cases. The service levels
References
show very close for those seven cases. In fact, the one
with the most flexible combination rule, LQ-A/LTF-A, Anand, G., and Ward, P.T. (2004), ‘‘Fit, Flexibility and
is slightly lower than the one with the least flexible Performance in Manufacturing: Coping with Dynamic
combination rule, LQ-C/LTF-C. In other words, Environments,’’ Production and Operations Management,
offering the downstream buyer the most flexible 13, 369–385.
options only slightly deteriorates the supplier’s service Aprile, D., Garavelli, A.C., and Giannoccaro, I. (2005),
level. ‘‘Operations Planning and Flexibility in a Supply Chain,’’
Production Planning and Control, 16, 21–31.
Banerjee, S., Banerjee, A., Burton, J., and Bistline, W. (2001),
‘‘Controlled Partial Shipments in Two-echelon
6. Conclusions
Supply Chain Networks: A Simulation Study,’’
Flexibility in manufacturing systems has been a very International Journal of Production Economics, 71, 91–100.
active issue for more than 20 years. Proper flexibility Barad, M., and Sapir, D.E. (2003), ‘‘Flexibility in Logistic
has offered a promising solution for improving the Systems – Modeling and Performance Evaluation,’’
performance of manufacturing systems. However, International Journal of Production Economics, 85,
a manufacturing system is just one of several important 155–170.
links in a supply chain. The issues of flexibility in Beach, R., Muhlemann, A.P., Price, D.H.R., Paterson, A.,
a supply chain system have become very interesting for and Sharp, J.A. (2000), ‘‘A Review of Manufacturing
Flexibility,’’ European Journal of Operational Research, 122,
researchers in recent years. This study investigated the
41–57.
advantages of a supply chain with different degrees of
Chan, F.T.S., and Chan, H.K. (2006), ‘‘A Simulation Study
order quantity and delivery lead time flexibility. When with Quantity Flexibility in a Supply Chain Subjected
unit holding cost dominates the total costs, flexibility to Uncertainties,’’ International Journal of Computer
of order quantity may provide significant cost savings. Integrated Manufacturing, 19, 148–160.
However, service level of an enterprise must be an Duclos, L.K., Vokurka, R.J., and Lummus, R.R. (2003),
important consideration when positioning order quan- ‘‘A Conceptual Model of Supply Chain Flexibility,’’
tity flexibility for the greatest benefit. It must be noted Industrial Management and Data Systems, 103, 446–456.
that service level should be maintained at least at Das, S.K. (1996), ‘‘The Measurement of Flexibility in
a certain level to keep customers’ loyalty. On the other Manufacturing Systems,’’ International Journal of Flexible
hand, when the cost per shortage is relatively high, lead Manufacturing Systems, 8, 67–93.
Das, D.K., and Abdel-Malek, L. (2003), ‘‘Modeling the
time flexibility allows the buyer to improve its service
Flexibility of Order Quantities and Lead-times in Supply
level and reduce the shortage cost. These advantages
Chains,’’ International Journal of Production Economics,
were demonstrated by the results of a supply chain 85, 171–181.
simulation. De Toni, A., and Tonchia, S. (1998), ‘‘Manufacturing
Flexibility: A Literature Review,’’ International Journal of
Production Research, 36, 1587–1617.
Garavelli, A.C. (2003), ‘‘Flexibility Configurations for the
Notes on contributors Supply Chain Management,’’ International Journal of
Yi-Chi Wang is an Assistant Professor Production Economics, 85, 141–153.
in the Department of Industrial Eng- Gerwin, D. (1993), ‘‘Manufacturing Flexibility: A Strategic
ineering and Systems Management at Perspective,’’ Management Science, 39, 395–410.
Feng Chia University. He received his Groover, M.P. (2001), Automation, Production Systems,
PhD in Industrial Engineering from and Computer-Integrated Manufacturing, New Jersey:
Mississippi State University, his MS Prentice Hall.
in Manufacturing Engineering from
Gupta, Y.P., and Goyal, S. (1989), ‘‘Flexibility of
Syracuse University, and his BS in
Mechanical Engineering from Tatung Manufacturing Systems: Concepts and Measurements,’’
Institute of Technology. His major research interests are in European Journal of Operational Research, 43, 119–135.
the area of metal cutting optimisation, joint replenishment Law, A.M., and Kelton, W.D. (2000), Simulation Modeling
problems and supply chain management. His work has app- and Analysis, Singapore: McGraw-Hill.
eared in journals including Robotics and Computer- Koste, L.L., and Malhotra, M.K. (1999), ‘‘A Theoretical
Integrated Manufacturing, Engineering Applications and Framework for Analyzing the Dimensions of
Artificial Intelligence, International Journal of Advanced Manufacturing Flexibility,’’ Journal of Operations
Manufacturing Technology, International Journal of Management, 18, 75–93.
1202 Y.-C. Wang

Koste, L.L., Malhotra, M.K., and Sharma, S. (2004), Silver, E.A., Pyke, D.F., and Peterson, R. (1998), Inventory
‘‘Measuring Dimensions of Manufacturing Flexibility,’’ Management and Production Planning and Scheduling,
Journal of Operations Management, 22, 171–196. New York: John Wiley & Sons.
Sánchez, A.M., and Pérez, M.P. (2005), ‘‘Supply Chain Suarez, F.F., Cusumano, M.A., and Fine, C.H. (1996), ‘‘An
Flexibility and Firm Performance: A Conceptual Model Empirical Study of Manufacturing Flexibility in Printed
and Empirical Study in the Automotive Industry,’’ Circuit Board Assembly,’’ Operations Research, 44,
International Journal of Operations & Production 223–240.
Management, 25, 681–700. Upton, D.M. (1994), ‘‘The Management of Manufacturing
Scala, J., Purdy, L., and Safayeni, F. (2006), ‘‘Application of Flexibility,’’ California Management Review, 36, 72–89.
Cybernetics to Manufacturing Flexibility: A Systems Vickery, S., Calantone, R., and Droge, C. (1999), ‘‘Supply
Perspective,’’ Journal of Manufacturing Technology Chain Flexibility: An Empirical Study,’’ The Journal of
Management, 17, 22–41. Supply Chain Management, 35, 16–24.
Sethi, A.K., and Sethi, S.P. (1990), ‘‘Flexibility in Zheng, Y.-S., and Federgruen, A. (1991), ‘‘Finding Optimal
Manufacturing: A Survey,’’ International Journal of (s, S) Policies is about as Simple as Evaluating a Single
Flexible Manufacturing Systems, 2, 289–328. Policy,’’ Operations Research, 39, 654–664.