An Importance Evaluation for Affiliated Stores Using Data Envelopment Analysis

Abstract

In recent years, there have been systems in which E-commerce website management companies provide a sales platform for a particular product, and individual stores or companies “affiliate” with these companies. The individual stores and companies can then sell their products online through this platform. In systems that use such sales models, the departments that are responsible for placing orders and managing contracts with affiliated stores need to evaluate the importance of each affiliated store as a part of the operations of the E-commerce website. As it stands, affiliated store evaluations are often conducted based on qualitative and ambiguous criteria. If these evaluations are influenced by ambiguous impressions of the affiliated stores, or ambiguous evaluation criteria, then this may result in incorrect measures being taken. Therefore, this study employs a Data Envelopment Analysis (DEA) model, referred to as a range adjusted measure (RAM) model, to evaluate the importance of each affiliated store, and conducted numerical experiments using actual data with this model. This model enables the provision of importance evaluation scores for each affiliated store. In addition, this model also enables the provision of specific numerical improvement targets for each evaluation criterion, and enables the provision of guidelines for which factors to improve to affiliated stores with low scores. Through this study, it was possible to quantitatively evaluate and analyze each affiliated store using the importance evaluation model.