Conference Agenda

Many organizations employ unique segmentation and customer evaluation methods to allocate appropriate managerial attention to each segment and individual customer. This presentation introduces an enhanced objective method for segmenting and comprehensively ranking customers. The proposed approach utilizes customer criteria with quantitative values derived from the organization's information system. Customer scores are determined objectively using Super Efficiency (SE), a ranking technique within Data Envelopment Analysis (DEA). This method allows tracking of each customer's relative position within their segment (e.g., Platinum, Gold, Silver, and Bronze) over time. It provides a precise, complete ranking based on company-defined criteria. The effectiveness of the proposed method was demonstrated through a real-world case study.

Data Envelopment Analysis (DEA) is widely used to evaluate the relative efficiency of decision making units (DMUs), including banks, hospitals, schools, and more, and provide benchmarking information (efficient targets) for them in operations research. In DEA, two common frameworks are employed for efficiency evaluation and target setting (benchmarking): the greatest-distance and least-distance frameworks. The greatest-distance framework identifies the farthest efficient target for the DMU under evaluation, essentially seeking an efficient target that suggests maximum improvements to achieve efficiency. Conversely, the least-distance framework aims to find the closest efficient target for the DMU under evaluation, seeking an efficient target that is most similar to the DMU under evaluation. This approach is often preferred as it provides a more attainable benchmark for improving the performance of inefficient DMUs.

In this research, we will introduce the operation of both conventional DEA models (referring to those based on the greatest-distance framework) and the least-distance DEA model to demonstrate the practicality of the closest efficient target. Additionally, we will present our newly proposed effective Mixed Integer Programming (MIP) approach for computing the closest efficient target. Furthermore, we will introduce a new DEA model called "Least-distance Range Adjusted Measure (LRAM)" for efficiency evaluation and benchmarking.