Benchmarking in Data Envelopment Analysis: Balanced efforts to achieve realistic targets

Santiago

Conferencia; The 23rd Conference of the International Federation of Operational Research Societies; 2023

International Federation of Operational Research Societies

Benchmarking consists of evaluating and analyzing the processes, products, services or other aspects of companies to draw a comparison and to take this information as a reference point to guide future strategic decision-making. The intention is to learn from the experience of other units in the sector to improve the performance of the evaluated unit.Various support tools for developing management and decision-making in organizations are used, including statistical and econometric approaches and Operational Research methods. Data Envelopment Analysis (DEA) has proven to be a useful tool for the benchmarking of Decision Making Units (DMUs) involved in a production process. In fact, benchmarking in DEA has been applied in multiple fields.The identification of best practices allows the establishment of targets and, with them, the design of improvement plans in susceptible areas. In the benchmarking process, it is important to bear in mind the following considerations: 1) the identification of best practices must be done taking into account the circumstances and characteristics of the organizations being evaluated, 2) these best practices must reflect efficient behaviors and 3) the established targets must be achievable and, as far as possible, require control over the effort necessary for their achievement.In the total absence of information, the most realistic and conservative solution would be the one where the effort necessary to achieve optimal operating levels is as balanced as possible in all dimensions, without neglecting the overall necessary effort. In this regard, we propose three different approaches (Minimum Range Model, Distance to the Minimum Squared Model and Distance to the Mean Squared Model), with the aim of reaching an impartial distribution of efforts to achieve optimal operating levels without neglecting the overall effort required. Therefore, we offer different alternatives for planning improvements directed toward DEA efficient targets, where the decision-maker can choose the one best suiting their circumstances. Moreover, and as something new in the benchmarking DEA context, we will study the properties satisfying the targets generated by the different models proposed.Finally, an empirical example used in the literature illustrates the new methodologies, comparing with closest targets, proposed by Aparicio, Ruiz & Sirvent (2007) and the MRAM measure (Aparicio, Monge & Ramón, 2021)