Two-step benchmarking: Setting more realistically achievable targets in DEA

TL;DR

This paper introduces a two-step benchmarking method in DEA that sets more realistic short-term targets for poorly performing units, offering practical improvement plans aligned with long-term goals.

Contribution

It proposes a novel two-step DEA benchmarking approach that provides attainable short-term targets and alternative improvement strategies for decision-making units.

Findings

Enables setting more achievable short-term targets for poor performers.

Offers multiple improvement pathways towards DEA efficient targets.

Illustrated with research performance of Spanish universities.

Abstract

The models that set the closest targets have made an important contribution to DEA as tool for the best-practice benchmarking of decision making units (DMUs). These models may help defining plans for improvement that require less effort from the DMUs. However, in practice we often find cases of poor performance, for which closest targets are still unattainable. For those DMUs, we propose a two-step benchmarking approach within the spirit of context-dependent DEA and that of the models that minimize the distance to the DEA efficient frontier. This approach allows to setting more realistically achievable targets in the short term. In addition, it may offer different alternatives for planning improvements directed towards DEA efficient targets, which can be seen as representing improvements in a long term perspective. To illustrate, we examine an example which is concerned with the research performance of public Spanish universities.