Modeling multiple levels of targets in operational efficiency evaluation with DEA
Abstract
Purpose
Researchers have noticed that in efficiency assessment, some attributes exhibit specializations including non-discretionary, non-controllable or undesirable. This paper aims to focus on other special factors which have target levels to achieve, i.e. the inputs (outputs) are no longer the-less-the-better (the-more-the-better).
Design/methodology/approach
In this paper, the authors further study the target variables when some attributes have multiple levels of targets. Such a situation can be found in many operational efficiency evaluations with various targets or bounded scale in inputs or outputs. They suppose that decision-making units (DMUs), reaching any target level, are identical efficient. To some extent, it is mitigation between common targets and individual targets. Using the closest target rule, the authors propose a target-level-oriented method to evaluate DMUs locally.
Findings
First, the authors found that some factors have multiple levels of targets to improve its efficiency in real world practice. Second, the proposed technique is able to deal with the multiple-levels-targets problems in data envelopment analysis (DEA) framework. Third, the decision-maker can select the improvement directions more freely than that in the traditional setting.
Originality/value
First, this is the first paper to discuss the multiple-levels-targets problems in DEA framework. Second, the proposed technique can help the decision-maker to select the best improvement strategies. Third, the technique developed in this paper can be used in many areas. For example, it can support the environmental efficiency evaluation with different standards of pollution emission.
Keywords
Acknowledgements
Financial support from National Natural Science Foundation of China (Grant No. 71301039) is gratefully acknowledged.
Citation
Long, X. and Xia, Q. (2018), "Modeling multiple levels of targets in operational efficiency evaluation with DEA", Journal of Modelling in Management, Vol. 13 No. 2, pp. 418-433. https://doi.org/10.1108/JM2-03-2016-0024
Publisher
:Emerald Publishing Limited
Copyright © 2018, Emerald Publishing Limited