Abstract

Cross-efficiency evaluation in data envelopment analysis (DEA) is effective for evaluating the efficiency of decision-making units (DMUs). The use of cross-efficiency evaluation methods based on prospect theory has recently increased. However, the internal structure of DMUs is often ignored in efficiency evaluation; further, a fixed status quo is often selected as the reference point to calculate relative gains and losses, which is not ideal in the context of prospect theory. To address these issues, we investigate the basic two-stage cross-efficiency evaluation in DEA based on prospect theory. An optimism coefficient is introduced to formulate parameterised dynamic reference points, and a target identification model is developed to obtain target efficiency values that are attainable for all DMUs. Based on the prospect value and target efficiency values, we propose multiple novel aggressive, benevolent, and neutral two-stage cross-efficiency evaluation models. The models proposed herein can be applied to various decision environments and arbitrarily extended to other network system structures. A case study in sustainable supplier selection is performed to demonstrate the effectiveness of the proposed models for DMU ranking. The sensitivity analysis results show that the psychological characteristics of the decision maker under risk affect the evaluation results.

摘要

数据包络分析（DEA）中的交叉效率评估对于评估决策单元（DMU）的效率是有效的。最近，基于前景理论的交叉效率评估方法的使用有所增加。然而，DMU的内部结构在效率评估中往往被忽视；此外，通常选择固定的现状作为参考点来计算相对收益和损失，这在前景理论的背景下并不理想。为了解决这些问题，我们研究了基于前景理论的 DEA 中基本的两阶段交叉效率评估。引入乐观系数来制定参数化动态参考点，并开发目标识别模型以获得所有 DMU 均可达到的目标效率值。基于前景值和目标效率值，我们提出了多个新颖的激进、仁慈和中立的两阶段交叉效率评估模型。本文提出的模型可以应用于各种决策环境，并可任意扩展到其他网络系统结构。进行了可持续供应商选择的案例研究，以证明所提出的 DMU 排名模型的有效性。敏感性分析结果表明，风险下决策者的心理特征会影响评价结果。进行了可持续供应商选择的案例研究，以证明所提出的 DMU 排名模型的有效性。敏感性分析结果表明，风险下决策者的心理特征会影响评价结果。进行了可持续供应商选择的案例研究，以证明所提出的 DMU 排名模型的有效性。敏感性分析结果表明，风险下决策者的心理特征会影响评价结果。