- 1 -

Sixteenth Annual Conference of POMS, Chicago, IL, April 29 - May 2, 2005

No. 003-0236

SUPPLY CHAIN MANAGEMENT
AND PLANT PERFORMANCE

-An Empirical Analysis of the Fisher Model

Jörn-Henrik Thun

Industrieseminar, Mannheim University
Schloss S 205, 68131 Mannheim, Germany

phone: ++49 621 181 15 84,
fax: ++49 621 181 15 79

Keywords Supply Chain Management,High Performance Manufacturing, Fisher Model, Performance

Abstract

In this paper Supply Chain Management will be analyzed empirically based on the data of the “High Performance Manufacturing”-Project, a research cooperation of universities from different countries. The aim of the paper is to investigate the relation between Supply Chain Management and plant performance. The empirical analysis investigates the FisherModel whichdistinguishestwo different kinds of Supply Chains: physically efficient vs. market-responsive Supply Chains. Differences concerning plant performance in terms of efficiency and responsiveness can be shown.

Introduction

For manyindustriesSupply Chain Management has become one of the most important concepts for improving the flow of material and information. The potential of this concept has often been mentioned in the academic literature (see e.g. Chopra and Meindl, 2001). The basics of Supply Chain Management areintroduced by Forrester when he describes the phenomenon of an amplifying demand between different supply tiers (Forrester, 1958). This phenomenon commonly referred to as the “bullwhip”-effect has become popular from the observation of logistic executives at Procter Gamble. While the customers demand diapers at a steady state,the logistic executives observed an increased order variabilityof disposable diapers moving up the supply chain. Supply Chain Managementcan mitigatethe negative influences of thiseffect bycoordinating information and material between companies (Lee et al., 1997).

In the academic literature there is nostandard definition of the term Supply Chain Management. Jayram and Bechtel distinguish betweenschools with different approaches of Supply Chain Management, i.e. the ‘FunctionalAwarenessSchool’, the ‘Linkage/Logistics School‘, the ‘InformationSchool’, the ‘IntegrationSchool’, and the ‘Future School’ (Bechtel and Jayaram, 1997). The ‘Functional Awareness School’ stresses the flow of material beginning at the supplier and ending at the customer, whereas the ’Linkage/Logistics School’ goes one step further aiming to investigate how linkages among the functional areas of a supply chain, i.e. purchasing, manufacturing, and distribution,can be exploited for competitiveness. The ‘InformationSchool’ focuses on the meaning of a bidirectional information flow within the chain, whereas the ‘Integration/Process School’ overcomes the view of a unidirectional flow of material and stresses the integration of business processes. Following the ‘FutureSchool’ the term Supply Chain Management should be replaced by the expression “seamless demand pipeline” with the customer as the activator of the chain.

Despite the different approaches of the supply chain schools some definitions will show the nature of Supply Chain Management. A definition related to the ‘Linkage/Logistic School’ is provided by Chopra and Meindl (2001): “Supply Chain Management involves the management of flows between and among stages in a Supply Chain to maximize total profitability.” Handfield and Nichols (1999) define Supply Chain Management as “…the integration of [..] activities through improved supply chain relationships, to achieve a sustainable competitive advantage.”Both definitions mention implicitly the meaning of management of relationships between different institutions within a chain, i.e. suppliers and customers.

Christopher notes explicitly suppliers and customers and defines the term Supply Chain Management as the “… management of upstream and downstream relationships with suppliers and customers to deliver superior customer value at less cost to the supply chain as a whole.” (Christopher, 2004) This means that the focus of Supply Chain Management is upon the management of relationships in order to achieve a more profitable outcome for all parties in the chain. Christopher argues that the term supply chain management, although it is widely used, should actually better be renamed.

First, the aspect that the chain has to be driven by the market is to be stressed. Christopher suggests the term ‘demand chain management’. Second, he argues that the word ‘chain’ is irritating since there will normally be multiple suppliers and customers. So the term ‘network’ seems to be more appropriate. Following Christopher asupply chain can simply be definedas a “...network of connected and interdependent organizations mutually and co-operatively working together to control, manage, and improve the flow of materials and information from supplier to end user.” (Christopher, 2004) This network will include several stages within the chain, i.e. following the Supply-Chain Council “… from your supplier’s supplier to your customer’s customer….” Additionally, it can be argued that also the term ‘network’ can better be replaced by the term ‘acyclic graph’ since there will be a flow of material without cyclic processes. For this paper is important to note that supply chain management involves different institutional entities which have to be coordinated in terms of information and material in order to generate a mutual benefit.

Despite the fact that Supply Chain Management has become on of the most discussed issues of production and operations management there is still a need for empirical work. Accordingly, it is the aim of this paper to contribute an empirical analysis concerning the relationship between Supply Chain Management and plant performance based on the data base of the international research project “High Performance Manufacturing”.

Supply Chain Management and the FisherModel

Performance Measures of Supply Chain Management

Although there exists a great variety of possible performance measures, in this paper two factors will be identified subsuming the most important aspects for measuring the performance of supply chain management. The performance measures are derived from the Fisher model (Fisher, 1997). Fisher distinguishes between two different kinds of products: Functional products and innovative products. Functional products are characterized by long product life cycles, low product variety, high stability of demand, etc., whereas innovative products have relatively short product life cycles, show a great variety of variants, and their demand is unpredictable. Following Fisher, supply chains should be designed according to the type of product. The crux is matching supply chains with products, i.e. “functional products require an efficient process; innovative products, a responsive process” (Fisher, 1997: 109).

In his approach Fisher distinguishes between physically efficient and market-responsive supply chains. Physically efficient supply chains serve standard products for a market with stable, predictable demand. Market-responsive supply chains offer a great variety of innovative products and can change quickly without high switching costs due to flexible processes. They are close to the market and respond quickly to unpredictable demand.Based on these two kinds of supply chains performance measures can be identified.

The performance of a physically efficient supply chain can be measured by indicators like high average utilization rate, low inventory, or in general low cost. Contrary to that a market-responsive supply chain tries to match best with customers expectations in terms of speed, flexibility, and quality. Relevant indicators are e.g. on-time delivery ratio or short delivery times. Fisher claims that supply chains matching the particular type of product will have superior performance, whereas supply chains with a mismatch tend to be the ones with problems. The following figure depicts the Fisher Model.

Type of Product
Functional
Products / Innovative
Products
Type of Supply Chain / Efficient
Supply
Chain / match / mismatch
Responsive
Supply
Chain / mismatch / match

Figure 1: The FisherModell

Despite the great popularity of the SCOR model, supply chain management practices will be discussed in the following in order to create a generic framework for supply chain management from a institutional viewpoint. This framework will serve as a basis for the empirical analysis.

Practices of Supply Chain Management

In the literature there is a vast variety of approaches and practices for supply chain management. Therefore, different aspects will be discussed in order to identify the most important practices, some of which can be derived from the schools of supply chain management discussed in the last section.

The basic idea of supply chain management is the planning and control of the material flow.Therefore, cross-company-analysis of supply chains is reasonable, becausematerial management need not stop at the factory doors. The activities within a supply chain should be coordinated between the participating partners thus the supply chain can be managed as a whole. A local optimization of every company will generally not lead to the maximum of the whole supply chain. Here the ‘FunctionalAwarenessSchool’ comes into play stressing the flow of material beginning with the supplier and ending at the customer.

Another crucial aspect in terms of supply chain management is the relationship with suppliers. In order to establish a sustainable cooperation suppliers must be regarded as partners rather than as opponents. Accordingly, in terms of supplier relationships often the expression “durable arm’s length relationships”(Dyer et al., 1998) is used. Supplier Partnership will be seen as one column of supply chain management.

By involvementsuppliers are included in supply chain related aspects. Thereby it is important to establish efficient communication processes. Following supplier involvement, problems must be shared with suppliers. Additionally the involvement of suppliers is important to reconfigure processes within a supply chain. Accordingly, the ‘information school’ and the ‘integration school’ build the foundation for supplier involvement.

Beside the suppliers customers play an essential role for supply chain management as well. It is important to stay in close contact with customers, respond quickly to their needs, and give feedback about quality and delivery performance. Customer integration is a focal point of the ‘integration school’, whereby the integration of business processeswith customers and suppliers must be stressed. Also in the ‘InformationSchool’ customer involvement is important in terms of the bidirectional information flow within the chain. Customer involvement will be seen as the forth column of supply chain management. Although there are other aspects like Information Technology, from an institutional viewpoint, i.e. putting the stakeholders supplier, manufacturer, and customer into consideration, supply chain management will be characterized by the columns Supply Chain Planning, Supplier Involvement, Supplier Partnership, and Customer Involvement. Figure2 gives an overview of the institutional view of supply chain management.

Figure 2: Columns of Supply Chain Management

An Empirical Analysis of the Fisher Model

The “High Performance Manufacturing”-Project

The empirical analysis of the impact of supply chain management on the performance of manufacturing companies is based on data collected within the international empirical research project “High Performance Manufacturing”. The “High Performance Manufacturing”-project is an international cooperation with the purpose to evaluate critical success factors in operations management. High Performance Manufacturing declares the ability of a production unit to reach continuous improvements in the manufacturing area through integration and utilization of different management concepts. Former analysis of previous rounds is done by Schroeder, Flynn et al. (Schroeder and Flynn, 2001). The basic aim of the project, in which research groups from the U.S., Japan, Germany, Sweden, Finland, and Koreaparticipate, is to identify the management practices pursued by plants which are commonly seen as being at the edge in their industry with respect to performance.

The data base comprises qualitative and quantitative information of over 189 manufacturing plants collected in the automotive supply industry, the electronics industry, and the machinery industry in the six countries. In each plant, 23 persons from various levels within the hierarchy and the important functional areas had to fill in the questionnaires ensuring a transversal image of the plant. The persons being asked in each plant are the Plant Manager, the Inventory Manager,the Plant Superintendent, the Product Development Manager, the Process Engineer, the Manager of Accounting, the Human Resource Manager, the Production Control Manager, the Quality Manager, the Information Systems Manager, 3 Supervisors, and 10 Direct Labors. Figure 3 gives an overview of the participating countries and the plants from the different industries.

Figure 3: Data Set of the “High Performance Manufacturing”-Project

Within the fields Just-in-Time, Quality Management, Human Resources, Manufacturing Strategy, Information Management, and Technology Management, also information about Supply Chain Management are contained in the data base as well as indicators for the competitive performance of the plants.

Operationalizing the Factors of Supply Chain Management

The asked items are answered based on a 7-point-Likert scale. Altogether 21 items have been taken from the questionnaire of the “High Performance Manufacturing”-project in order to create four factors reflecting the four institutional supply chain management columns discussed in the previews section (see appendix for the items). The items have been aggregated by a factor analysis. The following table shows the criteria for validity and reliability.

Factors / Items / Eigenvalue / Var. expl. / Alpha
Supply Chain Planning / 5 / 2.893 / 57.86 / 0.816
Customer Involvement / 6 / 3.129 / 52.15 / 0.805
Supplier Partnership / 5 / 3.016 / 60.32 / 0.826
Supplier Involvement / 5 / 2.659 / 53.19 / 0.765

Table1: Validity and Reliability of the SCM-scales

The validity of the factors is tested by theirEigenvalues and the explained variance. Firstly, the Eigenvalue of each factor must be greater than 1. Secondly, all loadings of a factor must be greater than 0.5, which holds for every factor. Furthermore the explained variance should be greater than 50%. The reliability of the factors is tested by Chronbach’s Alpha.Following Nunnally (1978) a value of 0.7 for Chronbach’s Alpha is regarded as acceptable. All four factors fulfill the criteria for validity and reliability and can be used for further analysis.

To show differences concerning the degree of implementation of supply chain manufacturing a cluster analysis was performed. For the clustering Ward’s method with the squared Euclidean distance was used. Based on the cluster analysis the plants can be divided into cluster of high and low degree of implementation. The following figure shows the average values of the factors for the created cluster.

Figure4: Mean values of the SCM-cluster

As it can be seen from the figure the cluster with a high implementation degree has higher average values than the other cluster. These results are confirmed by t-tests. Accordingly, the cluster can be used to identify those plants with a high implementation of the generated factors. In the following only plants from the cluster with a high implementation degree will be considered to guarantee that only those plants are investigated that actually do supply chain management. Accordingly, almost 50% of the plants are excluded from the sample leaving 96 plants for further analysis. The following figure depicts the average values of the performance measures of the two clusters.

Figure5: Mean values of the performance measures

As it can be seen from figure 5 the cluster “SCM high” shows a better performance in all criteria. But the differences are not significantly high. A reason for this might be that there are plants in the cluster “SCM high” pursuing a different supply chain strategy, i.e. some might aim for efficiency and some for responsiveness.

A Comparison of Responsive and Efficient Supply-Chain-Plants

The plants from the cluster with a high implementation degree are separated concerning the type of their products. This is done by creating a measure reflecting the customization or standardization of the product. For each plant this measure called “Degree of Customization” will be calculated in order to distinguish between plants with functional or innovative products. In terms of the type of supply chain another factor is calculated in order to distinguish between plants which focus on responsiveness instead of efficiency called “Degree of Responsiveness”. The relevant items are the manufacturing goals inventory turnover and volume flexibility which are mentioned by Fisher as well (Fisher, 1997: 109). Both created factors have been standardized by means of comparable results leading to a mean of 0 and a standard deviation of 1. The resulting positioning of the plants in the Fisher Model is depicted in figure6.

Type of Product
Functional
Products / Innovative
Products
Type of Supply Chain / Efficient
Supply
Chain /
Responsive
Supply
Chain

Figure 6: Positioning of SCM-plants in the FisherModel

The figure shows that the plants are positioned on or close to a diagonal. This leads to the presumption that there exists indeed a relationship between the type of product and the corresponding type of the supply chain like it has been claimed by Fisher. This presumption can be tested by a linear regression analysis. For the linear regression analysis the following equation with the factor “Degree of Customization” (DoC) as independent variable and the factors “Degree of Responsiveness” (DoR) as dependent variable results on a high level of significance with p< 0.01:

DoR=– 0.47*DoC

The linear regression equation indicates that the presumption can not be rejected. In the following the impact on performance measures will be investigated.

Linking the Fisher Model with plant performance

For the empirical analysis of the relationship of the Fisher Model and plant performance the plants are grouped according to their position in the matrix described in the previous section. In a first step the plants are separated into two groups according to the type of product. The group with functional products and an efficient supply chain contains 15 plants. In the group with innovative products and a responsive supply chain there are 19 plants. In a second step a comparison of means is done. For the comparison of the performance measures for efficiency and responsiveness will be chosen, because the plants in the two groups aim at different goals. Figure6 gives an overview of the mean values of plants with an efficient supply chain.