The use of the DWV3 classification system in manufacturing companies for evaluating a market-specific supply chain strategy - A case study at Atlas Copco Industrial Technique

Abstract: The research topic of this study is market-specific supply chain strategy, and the research problem is defined as, how manufacturing companies can use the DWV3 classification system to evaluate the opportunity for a market-specific supply chain strategy. What has been written about the DWV3 classification system is somewhat general in its nature and the practitioner is left without detailed instructions on how to proceed with the analytical analysis. Key elements of the DWV3 classification system that is not explicitly described in the literature is (1) how to measure each of the classification variables, (2) how to define a suitable limit for each measure in order to classify the products and (3) how to reason when sequencing the classification variables in the clustering analysis. Hence, the purpose of this thesis is to make the DWV3 classification system more available to practitioners, and thus the aim is to illustrate how to tackle the key elements of the framework by applying it on the Atlas Copco Industrial Technique Business Area product portfolio. A single-case study design was chosen as a suitable research approach for this thesis. The application of the DWV3 system to the ITBA product portfolio was considered as the phenomenon under investigation, the case, of this study. Two sets of quantitative data were collected, demand data and product master data. The qualitative data collected was related to the ITBA supply chain set-up and the products as well as the customers’ responsiveness requirements for each assortment included in the study. All qualitative data was collected through interviews. The findings of this study are summarized in a number of conclusions that can serve as guidelines for practitioners that are about to apply the DWV3 system. These are (1) as far as possible use measures at the single product level, (2) use measures that express each classification variable in a way that is relevant to the matching of demand characteristics and supply chain strategy, (3) be prepared to redefine initial measures in order to describe the studied products’ characteristics in the best possible way, (4) develop measures that are based on available data or data that is feasible to attain, (5) adjust the number of codification levels to find the best trade-off between the level of detail in the cluster analysis and the number of populated segments, (6) alter the sequencing and repeat the cluster analysis to gain insight into the demand characteristics of the product portfolio, (7) the final sequencing of the classification variables must produce clusters that are relevant for the chosen production philosophy concepts.