Incorporating undesirable outputs into the best and worst practice DEA models for customers scoring

Abstract

Determining the profitability of customers and ranking them is an issue that has been frequently studied in the past decades. To differentiate marketing activities towards the profitable segments of customers from the non-profitable ones, it is necessary to separate customers into different categories (Jonker et al., 2004). To this end, in this paper, data envelopment analysis (DEA), as a multiple criteria decision making tool is used to evaluate customers and classify them as good and bad customers. Particularly, the objective of this paper is to use the best practice frontier (BPF) DEA in the presence of undesirable outputs to recognise good customers. As well, the worst practice frontier (WPF) DEA in the presence of undesirable outputs is developed to find bad customers and save the marketing expenses for the best practice ones. A case study demonstrates the use of the proposed models.