Int. J. Integrated Supply Management, Vol. 1, No. 1, 2004
Assessing supply chain flexibility: a conceptual framework and case study I. Nyoman Pujawan Center for Supply Chain and e-Business Management, Department of Industrial Engineering, Sepuluh Nopember Institute of Technology, Kampus ITS Sukolilo, Surabaya 60111, Indonesia E-mail:
[email protected] Abstract: Flexibility has been considered as a major determinant of competitiveness in an increasingly intense competition in the marketplace. A large body of literature has been addressing various issues of flexibility in the last two decades. However, the discussions have mainly been from the viewpoint of a manufacturing company as a single entity in a supply chain. The flexibility related to machine, process, routing, part, worker and the like are all associated with a manufacturing or a production system. With the advent of the supply chain management concepts, business communities have been realising that being flexible in a production system only is insufficient. Thus, flexibility concepts should be broadened from the perspective of a production system into a supply chain system. However, the study addressing supply chain flexibility is still limited. This paper presents a framework for assessing flexibility of a supply chain. Four main parts of flexibility are identified including flexibility of the product delivery system, production system, product development, and supply system. In each of these parts, a number of pertinent elements are defined. A general guideline for conducting flexibility assessment is also presented. In an attempt to assess the model validity, a case study also forms a part of the paper. Keywords: supply chain; flexibility assessment; framework; case study. Reference to this paper should be made as follows: Pujawan, I.N. (2004) ‘Assessing supply chain flexibility: a conceptual framework and case study’, Int. J. Integrated Supply Management, Vol. 1, No. 1, pp.79–97. Biographical notes: Dr I. Nyoman Pujawan is a Senior Lecturer and Director of the Center for Supply Chain and e-Business Management. He received a first degree in Industrial Engineering from Sepuluh Nopember Institute of Technology (ITS), Surabaya, Indonesia, Master of Engineering in Industrial Engineering from Asian Institute of Technology (AIT), Bangkok, Thailand, and PhD in Management Science from Lancaster University, UK. His research interests are in the area of Supply Chain Management and Production/Operations Management. His papers have been published in the International Journal of Production Economics, European Journal of Operational Research, and International Journal of Operations and Quantitative Management. He is a member of Decision Sciences Institute.
Copyright © 2004 Inderscience Enterprises Ltd.
79
80
1
I.N. Pujawan
Introduction
Today’s customers are smart and clever. They no longer accept standardised products as they have in the past. To a supply chain, such diverging customer needs represent two things. First, it is a source of intense competition. Suarez et al. [1] stated that high competition means highly volatile markets, short product life cycle (PLC) and highly sophisticated buyers. Second, it is a cause of uncertainty. With customers requiring highly diverging products in the market, it becomes highly difficult for the supply chain to accommodate the customer needs into a product design and to predict the level of demand for a certain product. These all contribute to difficulties in managing the operations of a supply chain. In addition, high product variety (PV) increases the costs associated with production and delivery of products to the customers. This is logical since, when the variety of products demanded by the customers is high, the supply chain will lose the opportunity to exploit the economies of scale in many of its activities. The cost implication for this situation is clear: the unit price to be paid by the customer is higher, especially when the supply chain is not flexible enough to manage products with high variety. A study conducted by Berry and Cooper [2] has shown that productivity of a production system decreases when the PV is increased. This suggests that in order to be competitive in the marketplace, a supply chain is required to be able to produce various different products and deliver to the market in an acceptable speed and cost. This implies that flexibility is an important competitive advantage a supply chain should pursue to win the intense competition. Given that flexibility is important but pursuing high flexibility is costly, there should be an assessment on how much flexibility a supply chain should have. In many types of products, such as salt, sugar and mineral water, the need for flexibility is much less than for innovative products such as fashion products and computers. Fisher [3] provides a nice classification of products into two types: functional and innovative. Functional products are characterised by a relatively long life cycle, few product variations and easy to predict demand, thus error in forecasts at the time the production is committed is less than 10%. On the other extreme, the innovative products are characterised by a short PLC, wide variety of products and, consequently, the forecast errors are normally high. The focus of the supply chain in responding to these two types of products should certainly be different. A supply chain supplying innovative products should pursue responsiveness while for functional products costs should be the primary focus. Based on this classification, innovative products certainly require higher supply chain flexibility than the functional products do. It is important therefore that the assessment of flexibility for a manufacturing company as well as for a supply chain should relate the ability and the requirements to be flexible. Suarez et al. [1] argued that a company’s competitiveness is determined by its ability to answer the need from the market in terms of quality, efficiency and flexibility. It implicitly implies that a company does not need to be very flexible if the market does not require it. This notion is important because investment for flexibility is often costly and thus, high flexibility should be pursued only if the market indicates the need for it. The majority of the current literature on flexibility considers a manufacturing or a production system as a single entity. Various flexibility elements have been proposed in the manufacturing flexibility literature such as machine flexibility, material handling flexibility, process flexibility, routing flexibility, etc. Various research methodologies have been used including empirical investigation [1,4–6], development of mathematical
Assessing supply chain flexibility: a conceptual framework and case study
81
models [7–10], development of simulation models [11,12], development of theoretical framework [13] and literature survey [14,15]. Despite the rich development of the flexibility literature, there are at least two weaknesses exhibited by the current literature. First, the focus of the current literature has been mainly from the viewpoint of a manufacturing system or a production system as a single entity in a supply chain. With the advent of the supply chain management concepts, business communities have been realising that being competitive as a single company is no longer adequate; instead, competitiveness requires consideration of all channels in the supply chain [16,17]. Likewise, being flexible in the context of a manufacturing system is no longer adequate in the current competition. Flexibility should therefore be pursued by a supply chain, or at least by every function related to supply chain activities. In their review of empirical research on manufacturing flexibility, Vokurka and O’Leary-Kelly [15] presented 15 dimensions of manufacturing flexibility. Many of the dimensions are the same as those identified in Koste and Malhotra [13], which focused on elements of manufacturing flexibility. Although the latter identified other types of flexibility such as market and delivery flexibility, it seems that the focus is not specifically focused on the interface between channels in the supply chain. As stated by Golden and Powell [18], flexibility requires interorganisation data sharing in a supply chain. Despite its importance, the availability of the literature addressing supply chain flexibility is still limited to date. Second, there is a limited literature on supply chain flexibility which provides the readers with a relatively simple and ready to use framework to assess supply chain flexibility. Lummus et al. [19] presented a practical instrument to assess overall supply chain performance. Some of their ideas can be applied to assess supply chain flexibility, but what elements to be assessed, how to determine the relative importance of flexibility in each dimension and several other aspects relevant to supply chain flexibility would still require significant development. Taking into account the above two gaps, this paper presents a framework for supply chain flexibility assessment. The framework is designed in such as a way that a supply chain could compare between the desired level of flexibility and the current capability in various elements of supply chain flexibility.
2
Dimensions of flexibility
In the last two decades, manufacturing flexibility has been an issue that attracts much attention of the academics. A large body of literature has addressed flexibility as an important competitive advantage. D’souza and Williams [20] classified manufacturing flexibility into externally driven and internally driven manufacturing flexibility where each has two elements. The externally driven manufacturing flexibility includes two dimensions, volume and variety flexibility, while the internally driven flexibility includes process and material handling flexibility. Each of the dimensions has two elements: range and mobility. The authors offered a quite general definition on the two elements. Range was defined as the range of output volumes at which the firm can run profitably. Mobility, on the other hand, was measured in terms of the cost implication and the time required to increase or decrease the volume of output.
82
I.N. Pujawan
Koste and Malhotra [13] presented a comprehensive review on manufacturing flexibility. Ten dimensions of flexibility were identified from previous literature addressing flexibility. The dimensions include flexibility in machine, labour, material handling, routing, operations, expansion, volume, mix, new product and modification. The ten dimensions were then mapped into four elements: range-number, range-heterogeneity, mobility and uniformity. While the dimensions seem to cover a wide definition of flexibility, they only address the elements of flexibility internal to a manufacturing system. Chang et al. [6] conducted an empirical study on the relationships between business strategy and manufacturing flexibility. Six dimensions of manufacturing flexibility were used: product, product mix, product modification, volume, delivery and service. They suggested that companies should select appropriate dimensions of manufacturing flexibility and related those dimensions with the strategy of the firm. In an attempt to develop a supply chain flexibility model, Duclos et al. [21] considered six elements of flexibility: production system, market, logistics, supply, organisational and information system. Swafford et al. [22] developed a similar model on supply chain flexibility and agility. They presented a measure of global supply chain flexibility that considers the supply chain’s ability to adapt in a timely and cost-effective manner to a rapidly changing global competitive environment in providing products and services. The agility of a supply chain, according to the authors, is impacted by flexibility in product development, procurement, manufacturing and logistics. Each dimension of flexibility is defined by range and adaptability. Similar to the elements proposed by Slack [23] and D’souza and Williams [20], range is defined as the number of different states, such as levels, options and positions, that can be achieved with existing resources, while adaptability is the ability to change from one state to another in a timely and cost-effective manner. In a recent paper, Prater et al. [24] stated that supply chain agility is determined by speed and flexibility of sourcing, manufacturing and delivery. Speed is a measure of the time it takes to ship or receive a good. Flexibility, on the other hand, is the degree to which the firm is able to adjust the time in which it can receive or ship the goods. According to the authors, the degree of agility of a firm’s supply chain is determined by the configuration of the three dimensions (sourcing, manufacturing and delivery) to incorporate speed and flexibility. Each author may use different dimensions of supply chain flexibility. However, the dimensions should be related to supply chain functions. This usually includes procuring the materials (sourcing), developing new products, manufacturing/production and delivering the finished products. Hence, as in Swafford et al. [22], four dimensions of supply chain flexibility are considered in this paper: sourcing, product design, manufacturing/production and delivery.
2.1 Sourcing flexibility One of the keys in achieving supply chain flexibility is flexibility of any activities related to procurement of materials. Often, it is the ability of the suppliers that limits the ability of a manufacturer to respond rapidly to customer requirements [25]. The sourcing function is said to be flexible if it has sufficient extra supply capacity to anticipate sudden increases in the volume of materials required, the suppliers are able to deliver materials in various different speed options and to mix different items into a delivery load so that
Assessing supply chain flexibility: a conceptual framework and case study
83
small requests can be satisfied easily. Along these lines, Duclos et al. [21] defined supply flexibility as the ability to reconfigure the supply chain, altering the supply of product in line with customer demand. In Table 1, we propose ten items that need to be assessed when evaluating supply flexibility. Note that the numbers in the table belong to the case company to be presented in a latter section. They are put in the same table to reduce the paper pages. Table 1 No.
Worksheet for assessing supply chain flexibility* Description
Desired score
Current score
Supply flexibility (Weight = ) 1.1
The company has more than one qualified suppliers for each item
5
2
1.2
The costs incurred to switch the purchase of item from one supplier to another is low for most cases
4
3
1.3
Most suppliers have the capability of producing/supplying various different types of items
5
3
1.4
There is a large extra total supply capacity for most items
4
3
1.5
Most suppliers are capable of producing a large quantity of items in a relatively short time
4
2
1.6
Most suppliers are capable of producing a small quantity due to relatively low setup costs
5
2
1.7
For most items, multiple modes of transportation are available to deliver the items from the suppliers
5
3
1.8
Both minimum order quantity as well as multiple order quantity are small or LTL/LCL delivery policy is available for most items
4
3
1.9
It is technically and economically possible to mix different items into a delivery load
5
4
1.10
Most suppliers are able to deliver urgent delivery requests with faster mode of transportation where such a policy incurs reasonably low extra costs
5
3
Product development flexibility (Weight = ) 2.1
The level of product development activity can be changed by either outsourcing or having the team employed at other functions when the product development activity decreased temporarily
5
3
2.2
Outsourcing product development activity, where applicable, incurs reasonably low costs
5
4
2.3
The product development team has the capability of developing various products of different types and specification
5
2
2.4
The team has a software and other resources that enable them to easily create, modify and simulate the designs
5
3
Weighted gap
84 Table 1 No.
I.N. Pujawan Worksheet for assessing supply chain flexibility* (continued) Description
Desired score
Current score
Product development flexibility (Weight = ) 2.5
When the product design activities is done with collaboration of remotely dispersed team, there is a mechanism to easily communicate ideas, design files, etc.
5
3
2.6
The team is able to produce a large number of different designs from many standard modules
5
3
2.7
When a new design requires new materials, it is easy to obtain suppliers confirmation on their ability to supply the new materials
4
2
Production flexibility (Weight = ) 3.1
There are multiple production facilities that are located at different sites
4
3
3.2
The total production capacity is large enough to accommodate reasonably large increases in demand
3
3
3.3
When the total demand cannot be satisfied by the in house capacity/capability, it is easy to do outsourcing
4
2
3.4
Overtime and/or temporary hiring/layoff is possible to cope with short term demand fluctuation
5
4
3.5
Most employees are multi-skilled, thus they can be easily shifted from one job/task to another
5
4
3.6
The machines are multipurpose so they are able of processing various different tasks/jobs
5
3
3.7
The final configuration of products can be postponed until the customer orders are specified
5
3
3.8
The setup times for most machines are low, thus enabling them to economically process small batch sizes
4
3
3.9
There are alternative routings to produce a product
5
4
3.10
The planning system enables the planning personnel to easily change the existing production schedule
5
4
3.11
Costs implication of changing the schedule is low, thus changes may be requested within a short interval of time
5
4
Weighted gap
Assessing supply chain flexibility: a conceptual framework and case study Table 1 No.
85
Worksheet for assessing supply chain flexibility* (continued) Description
Desired score
Current score
Weighted gap
Delivery flexibility (Weight = ) 4.1
Different modes of transportation are available in delivering products to the customers
5
3
4.2
It is both technically easy and economical to mix different products into a delivery load
5
3
4.3
The minimum delivery quantity is small, thus small delivery order quantity from the customer can be satisfied
4
3
4.4
There is no restriction that the delivery quantity should be in full truck, container or other size. In other words, the multiple delivery quantity is small for most products
5
4
4.5
In case of emergency needs, speeding up the delivery of products is possible either by choosing a faster mode of transportation or by other means
5
4
4.6
It is possible to satisfy one delivery order of a customer from more than one warehouses or factories as well as by transhipments
5
3
4.7
The company set up a short frozen delivery schedule, thus the customers may change the quantity, types, and/or due date of delivery in a quite short period
4
2
4.8
The costs implication of changing the quantity, types and/or the due date of delivery is low
4
3
*The scores are for the case company presented in Section 6.
2.2 Product development flexibility Product development flexibility is determined by the ability of the company to produce various new designs in a timely and cost-effective manner and to flexibly deploy resources related to product development. Tatikonda and Rosenthal [26] revealed that there is a positive and statistically significant impact of resource flexibility to the success of product development projects. Activities related to product design are not solely the responsibility of the manufacturing company alone. A close relationship with suppliers also plays a role in increasing new product flexibility [1,25]. On the other hand, when the responsibility of developing new products is on more than one organisations’ shoulder, there should be a mechanism to easily communicate the design ideas, including files, prototypes, etc. Product development flexibility assessment should take into account such linkages in the context of a supply chain. As can be seen in Table 1, we propose seven items related to assessment of product development flexibility.
86
I.N. Pujawan
2.3 Production flexibility Manufacturing or production flexibility is related to the ability of the manufacturing system to produce products of different types and different volume at an acceptable speed and cost. Manufacturing flexibility has been a topic of interest for many researchers. Synthesising many earlier papers on manufacturing flexibility, Koste and Malhotra [13] defined ten dimensions of manufacturing flexibility including machine, labour, material handling, routing, operations, expansion, volume, mix, new product and modification. As noted by D’Souza and Williams [20], while most researchers agree on the general workable definition of manufacturing flexibility, there is significant variation in perspectives when manufacturing flexibility is broken down into more specific dimensions, elements and measures. When bringing flexibility to the supply chain context where manufacturing is considered only one functional dimension, such dimensions need restructuring. For example, new product flexibility in our model stands apart as one flexibility dimension. We propose 11 elements that need evaluating to assess manufacturing flexibility of a supply chain as can be seen from Table 1.
2.4 Delivery flexibility Beamon [27] defined delivery flexibility as the ability to change delivery dates. However, this definition seems to be insufficient. It should be broadened to include the ability of the supply chain to deliver different types of products to the customers with a wide range of volume at an acceptable cost and time. When assessing distribution system flexibility, one needs to take into account a number of considerations. These may include the possibility to schedule different routes in each day of the delivery, the ability of the company to obtain trucks from different sources and the possibility to work on a transhipment system that enables the products to be delivered from another parallel channel in addition to shipment from an upstream channel of the supply chain. In addition, the customers may frequently require a small quantity of products to be delivered immediately. To satisfy such requirements, a delivery system should have the capability to either mix different products into a truckload and/or use different modes of transportation. More detailed aspects to be evaluated in assessing delivery flexibility can be seen in Table 1.
3
Assessing the drivers of flexibility
The need for flexibility is largely determined by the operating and environment characteristics of a supply chain. Suarez et al. [1] pointed out that more volatile markets, shorter PLC and more sophisticated buyers have all contributed to flexibility’s emergence as a new strategic imperative. Other aspects such as uncertainty [13] and global competition [20,21] are also considered as factors behind the need for flexibility. Vokurka and O’Leary-Kelly [15] classified external factors on manufacturing flexibility into environmental factors, organisational attributes, strategy and technology. D’Souza and Williams [20] noted that there are external and internal drivers of flexibility. While the market situation and supply uncertainty (SU) are examples of external drivers, operating characteristics such as process similarity (PS) are internal drivers. To assess the degree of flexibility requirements in each functional dimension of a supply chain, one
Assessing supply chain flexibility: a conceptual framework and case study
87
should be able to assess each of the supply chain flexibility drivers. To enable such an assessment, we propose in this paper seven flexibility drivers that include both operating (internal) and environment (external) factors.
3.1 The length of product life cycle PLC is a measure of how long a product stays in the market before it disappears due to the introduction of new products or due to inability of the product to penetrate the market. A short PLC would mean that the product development team is required to be able to produce various different designs in a shorter time, the production system is required to be able to cope with the evolution of product designs, and the suppliers are also required to be able to cope with the changing requirements of the materials. Hence, the length of PLC affects the need of these three flexibility dimensions.
3.2 Product variety PV pertains to the number of different product types that are produced by the company. When there is a wide variety of products, not only the production function, but also the delivery and product development functions need to be flexible although with varying degrees of requirement. For example, a large PV usually means that the demand for each product is relatively small and delivering only one type of product in one truck or container would often be uneconomical. This finally leads to a requirement that the delivery function should be able to mix various types of product into one delivery load.
3.3 Customer requirements disparity (CRD) CRD is defined as the differences in the speed and service levels required by different customers or customer segments. The disparity in customer requirements has a high implication on the need for delivery flexibility. For example, a customer may require 2 days of response time, so delivery should be by air while the other may accept one week or more, so a slower mode of transportation may be employed. In responding to such diverging customer requirements, a delivery system should be able to economically employ different modes of transportation, different delivery policies as well as different service level targets for each customer or customer segment.
3.4 Order stability (OS) We define OS as the stability of orders placed by the customers in terms of due date, order quantity and the types of items required. Very often the customers request changes in the due date and the quantity of orders as well as the types of items to be delivered. Such changes, especially if not communicated early to the company, would result in many disturbances, often referred to as nervousness, to the supply chain unless there is a high flexibility in the supply chain functions [28]. The degree of flexibility requirements in the delivery, production and sourcing functions are all affected by the degree of OS.
88
I.N. Pujawan
3.5 Component commonality (CC) A company may consider itself as having low, medium or high CC dependent on how many different materials and/or components are used in the majority of the finished products to be produced. CC has impacts on various performance of production system, including load variability and system disruption [29]. Low CC implies that the production system, the product development team as well the sourcing system deal with a large variety of materials and components, thus flexibility in those three functions need to be high.
3.6 Process similarity Similarity in the process required to produce various different products in the production floor affects the needs for production flexibility. Low PS would require a production system that is highly flexible in handling different machine tool requirements, different process sequences, different process times, etc. PS is considered to be high if different products pass through many similar processes with similar machine tools and process times, and low if different products pass through only small number of similar processes, requires significantly different machine tools and process times.
3.7 Supply uncertainty The uncertainty of supply is related to the competition in obtaining the materials, the availability of alternative sourcing, the nature of raw materials availability (e.g. seasonal for agricultural items), etc. The supply characteristics directly affect the need for supply flexibility. Indirectly, SU also corresponds to the need of production and delivery flexibility. For example, when there is an unexpected short term supply shortage of a raw material, an inflexible production system would incur high costs due to its inability to reschedule production or to decrease the production capacity temporarily. As mentioned above, each of the above drivers may affect one or more flexibility dimensions with varying degrees of importance. Figure 1 shows the relationships between the drivers and the dimensions of flexibility where different thickness of the lines represents the degree of relationship, i.e. the thick line represents a strong relationship, the medium line represents a moderate relationship and a thin line represents a weak relationship. For example, there is a strong relationship between CC with sourcing and production flexibility and a weak relationship with product development flexibility. Each line can be translated into a score, where the thick, medium and thin lines correspond, respectively, to 5, 3 and 1. These will be referred to as relationship scores.
Assessing supply chain flexibility: a conceptual framework and case study Figure 1
89
Relationships between drivers and dimensions of flexibility
In addition, there should be an assessment on the above seven drivers when a company would like to determine how much flexibility each dimension it should have. In this paper, each dimension is classified into three levels. For example, the length of PLC can be classified as short, medium or long. Table 2 presents the worksheet than can be used to assess the seven flexibility drivers. The score of 1, 3 and 5 are then given to the corresponding answers in each driver, where 1, 3 and 5 are representing, respectively, a low, medium and high need for flexibility. The score, referred to as driver score, will be used in quantifying the relative importance of flexibility in each supply chain functional dimension to be described in a later sub section.
90
I.N. Pujawan
Table 2
Worksheet for assessing flexibility drivers Score
Flexibility drivers
1
3
5
Product life cycle for majority of products
Long
Medium
Short
Product variety
Small
Medium
Large
Customers’ requirements disparity
Low
Medium
High
Order stability
High
Medium
Low
Component commonality
High
Medium
Low
Process similarity
Large
Medium
Small
Supply uncertainty
Low
Medium
High
Filling the above worksheet would not be easy. Almost all of the answers should be given subjectively, although one may develop limits for each answer for several drivers. For example, PLC of less than 1 year maybe considered as short, from 1 to 3 years as medium and more than 3 years as long. Similarly, PV may be classified as small if there are less than 25 varieties at a time, medium if there are 25–100 varieties and large if there are more than 100 varieties. However, such limits cannot be done easily for other drivers. Hence, it would be better if the assessment of the drivers is done by a consensus involving a number of key people in the company such as the relevant functional managers. Collecting objective data by relevant managers before making a consensus is desirable. The team of different managers may then use such data to facilitate the process of making the consensus.
4
Determining weight
Weight will be given to each flexibility dimension. It represents the relative importance of the corresponding flexibility dimension in relation to the operating and environmental characteristics facing the company. To obtain the weight, a number of alternatives are available including the analytical hierarchy process, which is based on pair wise comparisons between the dimensions. While such alternatives are possible to use, we believe that the idea of generating weights based on the assessment results of the drivers would reflect more appropriately the real situations requiring flexibility. We propose in this paper a mechanism of relating the flexibility drivers to the relative importance of supply chain flexibility. The weights are assessed by using the results of the driver scores obtained from Table 2 and the relationship scores shown in Figure 1. To convert the drivers’ scores and the relationship scores into the weights let us define: i: flexibility driver (i = 1, …, 7) j: flexibility dimension (j = 1, …, 4) vi: score for driver i sij: relationship score between driver i and flexibility dimension j. We first calculate the importance ratio, i.e. the ratio between the total score and the maximum possible score for each flexibility dimension, represented as βj, as follows:
Assessing supply chain flexibility: a conceptual framework and case study β j = ∑ i =1 vi sij 7
∑
7 i =1
v s max ij
91 (1)
We define vmax as the possible maximum score for each driver, which is five in our scale. The relative importance of flexibility dimension j is then calculated as follows: Wj = β j
5
∑
j
βj
(2)
Assessing the desired and current scores
The desired and current scores will be given to each statement in the assessment sheet shown in Table 1. The desired score is given to a desired situation while the current score represents the current state in each statement. Lummus et al. [19] stated that one proven method that can be used to evaluate the current capability of a supply chain is through the evaluation of a series of relevant questions. In our model, there are a total of 36 statements that need to be evaluated where ten statements correspond to supply flexibility, seven to product development flexibility, 11 to production flexibility and eight to delivery flexibility. In each statement, a Likert scale of 1–5 may be used to rate both the desired and the current scores where the scores are defined as follows: 1
very low flexibility in the corresponding flexibility element
2
low flexibility in the corresponding flexibility element
3
moderate flexibility in the corresponding flexibility element
4
high flexibility in the corresponding flexibility element
5
very high flexibility in the corresponding flexibility element.
It is recommended that the assessment is conducted by a panel of relevant functional managers and preferably led by a supply chain manager or a general manager. When getting those managers together is difficult, each of them may give the score separately with the aid of information from the relevant function. The scores are given based on subjective judgement. However it is advisable to obtain pertinent objective data before the assessment is conducted. For example, to give a score to statement 1.1, the procurement manager should obtain the data on the number of raw materials and components purchased from the suppliers and the number of qualified suppliers available for each type of item. If most of the items can be obtained from only one qualified supplier, the score of the current state should be given very low. In rating statement 3.5, as another example, the production manager needs to provide information on how to easily do overtime and/or to hire/layoff direct labour. There may be a situation that both options are unavailable, so the score would be very low. In some countries, overtime is possible but temporary hiring/layoff is prohibited by government regulation. In such a case, the score may be given low. If both options are available but limited, for example, the direct labour can only be laid off after 3 months of hiring, the answer may be moderate. It is scored very high only if both options can be done very easily.
92
I.N. Pujawan
The desired score need not always be 5. For example, if many of the suppliers are close geographically to the company then using a multimodal transportation system is not an urgent matter. Thus, the desired score of statement 1.7 may be 3. The score should not be 1 or 2 unless there is no supplier at all located very far geographically where such a multimodal transportation system is necessary. If the current score for a certain element is significantly lower than the desired score, an improvement is necessary. It is also possible to have a lower desired score than the current score. For example, when the available capacity is too large and never fully utilised even when the demand is peak then the desired score for statement 3.2 should be lower than the current score. The desired score may also be given in relation with the competitor capability. But this can only be done if information on what the competitors can do in the corresponding flexibility element is available. The difference between the desired and the current scores is the gap. Ideally, both the current and the desired capability should be close to each other. When either one is significantly higher than another then there is a significant gap which shows that either the supply chain is over- or under-designed in the corresponding flexibility element. For simplicity, we will define that the gap is the score for the current capability minus the desired capability. Thus, a positive gap represents a need for improvement of flexibility in the corresponding element. The results of the analysis will provide the management with an insight on what aspect of the supply chain needs improvement and/or investment in order to increase the overall flexibility. Logically one should prioritise the elements of the supply chain flexibility, which has a large gap and a high degree of importance (weight). Thus, it is necessary to obtain the weighted gap for each flexibility element in all four dimensions. The weighted gap can be obtained with the following formula: WG jk = W j ( E jk − C jk )
(3)
where WGjk = weighted gap for the kth flexibility element of dimension j Wj = weight or degree of importance of dimension j Ejk = desired score for the kth flexibility element of dimension j Cjk = current score for the kth flexibility element of dimension j Ideally, each element in each dimension has its own weight. However, when the number of elements is large, obtaining individual weight would be difficult and tedious. Hence, we use a single weight for each dimension, Wj, calculated from the process described in the previous section. Averaging the weighted gaps over all elements in each dimension will then provide an insight on which dimension the company should prioritise to improve. Visualising the results using a radar diagram as used by Lummus et al. [19] or mapping the weighted scores in a two-dimensional graph would be helpful in interpreting the results.
6
A case study
In order to provide insights on the applicability of the above framework, it was decided to conduct a case study. The company involved in the case study was a manufacturer of containers for cosmetics products such as hair gel and body lotion. The company
Assessing supply chain flexibility: a conceptual framework and case study
93
currently employed about 80 people with a production capacity of approximately 1.8 million units of containers per month. However, due to a rapid increase in demand, the company is in the process of expanding its capacity to about 200% of its current capacity in the near future. The company is producing a large variety of products. At the time the study was conducted, the company produced 860 product types. The PV increases rapidly. On average, about 100 new products are developed annually, but the load of product development activities are quite fluctuating. In the peak periods, the product development team is required to be able to develop as many as 20 new products in a month. Rapid new product introduction also shorten the PLC. The general manager of the company stated during an interview that the PLC for most products is currently approaching 1 year. The company is located in East Java, Indonesia. There are currently 72 customers located in Java and Bali Islands. About 10% of the products are supplied to customers in East Java, about 20% to West Java, 62% to Jakarta, 3% to Central Java and 5% to Bali. In addition to differences in geographic locations, the customers are also of varying sizes and management sophistications. While several customers are large multinational companies such as Unilever and Avon, others are local small- to medium-size companies. They normally have different requirements in terms of quality level, response time, typical order quantity and firmness of the orders. Small customers tend to be more volatile in terms of their requirements on the due date and order quantity compared to large companies. Before the study was conducted, a formal letter was sent to the general manager of the company explaining the purpose of the study and asking his participation in the study. The first meeting was then conducted after the company agreed to participate. The purpose of the meeting was to explain the essence of the study and the methodology to be used to a team of five people including the general manager and one representative from each of the following functions: production, purchasing, product development and delivery. The team was also given two sets of questionnaires, one was in English and the other was in Indonesian language. A handout explaining the general concept of flexibility as well as the guidelines of answering the questionnaire was also distributed and explained to the team. The Indonesian versions of the questionnaire and the handout were provided to help the team understand the questionnaire easily and to prevent misinterpretation of the questionnaire. The team was also given an alternative to either fill in the questionnaire with a consensus among them or fill it separately by each member of the team. The first method was chosen but the team asked a period of 2 weeks to complete the questionnaire. During this period several communications by phone took place mainly to make clarifications on several statements in the questionnaire. The completed questionnaire was then discussed in the second meeting. Not all elements were discussed, but only about 20% of them, chosen randomly. The discussion was mostly to obtain more detailed information and confirmation on why the team has given those ratings to their company. The team initially gave a score of 3 for customer requirement disparity. Our discussion revealed that the team appeared to have underestimated the high disparity in the ordering pattern, typical order quantity, and the service levels required by different customers. In the middle of the discussion, they recognised how divergent their customers actually were and changed the score to 5. Finally, as shown in Table 3, the team gave a score of 5 for PV and CRD, a score of 1 for CC and a score of 3 for the other four drivers. Using the relationship score defined in Figure 1, the importance ratio and the weight for each flexibility dimension was then
94
I.N. Pujawan
calculated. The results are presented respectively in the last two columns of Table 3. To provide an example, calculation of the importance ratio for dimension 1 (sourcing) is as follows:
β1 = Table 3
3(1) + 5(0) + 5(1) + 3(1) + 1(5) + 3(0) + 3(5) = 0.48 5(1 + 0 + 1 + 1 + 5 + 0 + 5) Calculation of weights Relationship score*
Flexibility dimension
PLC
PV
CRD
OS
CC
PS
SU
βj
Weight
Sourcing
1
0
1
1
5
0
5
0.48
0.185
Product development
5
3
0
0
3
0
0
0.60
0.233
Production
3
5
3
5
3
5
3
0.67
0.261
Delivery
0
3
5
5
0
0
1
0.83
0.321
Driver score
3
5
5
3
1
3
3
2.58
*Relationship scores are defined in this paper, but driver scores should be obtained from assessment using Table 2.
The total of the four βj is 2.58. Thus, for example, the weight of the sourcing flexibility is 0.48/2.58 which is 0.185. Following the same procedure, the weights for product development, production and delivery are 0.233, 0.261 and 0.321, respectively. The results indicate that achieving high flexibility for the delivery function is more important than for the other three functions. These results seem to be sensible. The company deals with a relatively large number of customers but has only a few suppliers. The variety of products is much higher than the variety of materials. Thus, the above weights seem to generally confirm the situation facing the company. The result of the assessment of the current and the desired scores for each elements of flexibility is presented in Table 1. Using the weights calculated above and the results of the assessment in Table 1, the weighted gaps were then obtained. Table 4 presents the weighted gaps for each element over four dimensions where number 1.1, etc., refers to the corresponding elements in the assessment sheet presented in Table 1. Table 4 shows that the company needs significant improvement in almost all of the elements as the gap is positive in almost all elements. The table also shows that the delivery and product development functions exhibit significantly larger average weighted gap than the other two functions. This indicates that flexibility in the delivery and product development functions need more improvement than the production and sourcing functions. The individual weighted gap provides more detailed information on which element of flexibility improvements need to be prioritised. In Table 4, ten elements with highest weighted gaps are indicated by the shaded cells. Four out of the ten shaded cells are associated with delivery functions that support the finding that improvement of flexibility in the delivery function should be given the highest priority. The results of the assessment were communicated to the team. In general, they endorsed the results and commented that the framework was useful in forcing them to think about the detailed elements of flexibility. Many of those elements had never been
Assessing supply chain flexibility: a conceptual framework and case study
95
thought to be important. The team did encounter several difficulties when deciding the desired scores. For example, they did not have enough knowledge on the competitors’ competence in many of the elements to be assessed. Indeed, for many elements in the assessment questionnaire, it would be easier to give the desired scores when the competitors’ capability is known. But this is more to do with the completeness of information owned by the company rather than the framework itself. Table 4
The weighted gap for each element of the flexibility assessment
Sourcing
7
Product development
Production
Delivery
No.
Weighted gap
No.
Weighted gap
No.
Weighted gap
No.
Weighted gap
1.1
0.555
2.1
0.466
3.1
0.261
4.1
0.642
1.2
0.185
2.2
0.233
3.2
0.000
4.2
0.642
1.3
0.370
2.3
0.699
3.3
0.522
4.3
0.321
1.4
0.185
2.4
0.466
3.4
0.261
4.4
0.321
1.5
0.370
2.5
0.466
3.5
0.261
4.5
0.321
1.6
0.555
2.6
0.466
3.6
0.522
4.6
0.642
1.7
0.370
2.7
0.466
3.7
0.522
4.7
0.642
1.8
0.185
Average
0.466
3.8
0.261
4.8
0.321
1.9
0.185
3.9
0.261
Average
0.482
Average
0.329
3.10
0.261
3.11
0.261
Average
0.308
Concluding remarks
A general framework for supply chain flexibility assessment has been presented in this paper. The paper extends the current literature on flexibility which focuses on manufacturing flexibility. This extension is important with the advent of supply chain management concepts in the last two decades. The framework proposed in this paper allows a supply chain manager to compare between the desired and the current capability of a supply chain in various flexibility dimensions. A case study presented in this paper suggests that the framework is useful in helping the managers to identify which elements of flexibility should be given emphasis in an attempt to increase supply chain flexibility. The framework in this paper has included assessments on the relationships between a company and its downstream and upstream channels. While the framework can be used internally by a manufacturing company without directly involving suppliers and customers in the assessment process, it would be more valuable if such an assessment can be done together with other channel members of the supply chain. For example, when assessing sourcing flexibility, a joint assessment with the suppliers would add significant value to the improvement efforts to be taken for a better supply chain flexibility. In addition, revising or customising the framework may be necessary before it is applied to any specific situation.
96
I.N. Pujawan
Acknowledgements The earlier version of the paper was presented in the 7th Asia Pacific Decision Sciences Conference held in Bangkok, 24–27 July 2002. The author gratefully acknowledges the constructive comments received from two anonymous reviewers on the earlier draft of the paper.
References 1 2
3 4 5 6
7 8
9 10
11
12
13 14 15 16
Suarez, F.F., Cusumano, M.A. and Fine, C.H. (1995) ‘An empirical study of flexibility in manufacturing’, Sloan Management Review, Fall, pp.25–32. Berry, W.L. and Cooper, M.C. (1999) ‘Manufacturing flexibility: methods for measuring the impact of product variety on performance in process industries’, Journal of Operations Management, Vol. 17, pp.163–178. Fisher, M.L. (1997) ‘What is the right supply chain for your products?’, Harvard Business Review, March–April, pp.105–116. Ariss, S.S. and Zhang, Q. (2000) ‘The impact of flexible process capability on the product-process matrix: an empirical examination’, DSI 2000 Proceedings, pp.990–992. Chang, S., Lin, N. and Sheu, C. (2000) ‘Aligning manufacturing flexibility with environmental uncertainty in high-tech. industry’, DSI 2000 Proceedings, pp.984–986. Chang, S., Yang, C., Lin, N. and Sheu, C. (2001) ‘Mapping manufacturing flexibility with business strategy: an empirical study’, CD ROM Proceedings of the 6th Asia Pacific Decision Science Conference, Singapore, July. Benjaafar, S. (1994) ‘Models for performance evaluation of flexibility in manufacturing systems’, International Journal of Production Research, Vol. 32, No. 6, pp.1383–1402. Laengle, K., Griffin, P.M. and Griffin, S.O. (1994) ‘A quantification of economic value of flexible capacity’, International Journal of Production Research, Vol. 32, No. 6, pp.1421–1430. Takahashi, K., Hiraki, S. and Soshiroda, M. (1994) ‘Flexibility of production ordering systems’, International Journal of Production Research, Vol. 32, No. 7, pp.1739–1752. Wainwright, C.E.R. and Bateman, N. (1998) ‘Auditing system flexibility in the context of manufacturing strategy information’, International Journal of Physical Distribution and Logistics Management, Vol. 28, Nos. 9–10, pp.724–740. Gertosio, C., Mebarki, N. and Dussauchoy, A. (2000) ‘Modeling and simulation of the control framework on a flexible manufacturing systems’, International Journal of Production Economics, Vol. 64, pp.285–293. Grag, S., Vrat, P. and Kanda, A. (2001) ‘Equipment flexibility vs. inventory: a simulation study of manufacturing system’, International Journal of Production Economics, Vol. 70, pp.125–143. Koste, L.L. and Malhotra, M.K. (1999) ‘A theoretical framework for analyzing the dimensions of manufacturing flexibility’, Journal of Operations Management, Vol. 18, pp.75–93. Sethi, A.K. and Sethi, S.P. (1990) ‘Flexibility in manufacturing: a survey’, The International Journal of Flexible Manufacturing Systems, Vol. 2, pp.289–328. Vokurka, R.J. and O’Leary-Kelly, S.W. (2000) ‘A review of empirical research on manufacturing flexibility’, Journal of Operations Management, Vol. 18, pp.485–501. Lambert, D.M., Cooper, M.C. and Pagh, J.D. (1998) ‘Supply chain management: implementation issues and research opportunities’, The International Journal of Logistics Management, Vol. 9, No. 2, pp.1–19.
Assessing supply chain flexibility: a conceptual framework and case study 17
18
19 20 21 22
23 24
25 26
27 28
29
97
Lummus, R.R. and Vokurka, R.J. (1999) ‘Defining supply chain management: a historical perspective and practical guidelines’, Industrial Management & Data Systems, Vol. 99, No. 1, pp.11–17. Golden, W. and Powell, P. (1999) ‘Exploring inter-organisational systems and flexibility in Ireland: a case of two value chains’, International Journal of Agile Management Systems, Vol. 1–3, pp.169–176. Lummus, R.R., Alber, K. and Vokurka, R.J. (2000) ‘Self-assessment: a foundation for supply chain success’, Supply Chain Management Review, July–August, pp.81–87. D’Souza, D.E. and Williams, F.P. (2000) ‘Toward a taxonomy of manufacturing flexibility dimensions’, Journal of Operations Management, Vol. 18, pp.577–593. Duclos, L.K., Lummus, R.R. and Vokurka, R.J. (2001) ‘A conceptual model of supply chain flexibility’, DSI 2001 Proceedings. Swafford, P., Ghosh, S. and Murthy, N. (2000) ‘A model of global supply chain agility and its impact on competitive performance’, Proceedings of the 31st National DSI Meeting, Orlando, Florida, November 2000, pp.1037–1039. Slack, N. (1983) ‘Flexibility as a manufacturing objective’, International Journal of Operations and Production Management, Vol. 3, pp.4–13. Prater, E., Biehl, M. and Smith, M.A. (2001) ‘International supply chain agility: tradeoffs between flexibility and uncertainty’, International Journal of Operations and Production Management, Vol. 21, Nos. 5–6, pp.823–839. Christopher, M. (2000) ‘The agile supply chain: competing in volatile markets’, Industrial Marketing Management, Vol. 29, pp.37–44. Tatikonda, M.V. and Rosenthal, S.R. (2000) ‘Successful execution of product development project: balancing firmness and flexibility in the innovation process’, Journal of Operations Management, Vol. 18, pp.401–425. Beamon, B. (1999) ‘Measuring supply chain performance’, International Journal of Operations and Production Management, Vol. 19, No. 3, pp.275–292. Pujawan, I.N. and Kingsman, B.G. (2000) ‘System nervousness and inventory locations: issues for buyer and supplier partnerships’, Proceedings of the 31st National DSI Meeting, Orlando, Florida, November 2000, pp.1174–1176. Vakharia, A.J., Parmenter, D.A. and Sanchez, S.M. (1996) ‘The operating impact of part commonality’, Journal of Operations Management, Vol. 14, pp.3–18.
