Data envelopment analysis (DEA) is a data‐driven tool for performance evaluation, measuring decision‐making units (DMUs) and designating them with specific weightings. The standard DEA model typically sets up that decision‐makers (DMs) are wholly rational to select the most favourable weights to obtain the maximum performance score, but does not take into account their attitude toward risk during the assessment. The prospect theory generally matches humans' psychological behaviours. Thus, our study captures the non‐rational behaviours of DMs, performing under risk scenarios, in order to construct a novel common‐weights DEA model that maximizes the total prospect value, which can vary more steeply for losses than for gains, hence obtaining a more realistic common weight scheme. Our proposed model not only generates DMUs, with higher total prospect values, but also greater degrees of satisfaction. The current study shows that the prospect theory can be aptly extended to the DEA research area, supplying a proper guideline for future DEA research.

中文翻译：

---

前景理论下数据包络分析中的通用权重集

数据包络分析（DEA）是用于性能评估，测量决策单位（DMU）并使用特定权重进行指定的数据驱动工具。标准的DEA模型通常会设置决策者（DM）完全理性地选择最有利的权重以获得最大的绩效得分，但并未考虑评估过程中他们对风险的态度。预期理论通常与人类的心理行为相匹配。因此，我们的研究捕获了在风险情况下执行的DM的非理性行为，从而构建了一种新颖的通用权重DEA模型，该模型使总预期价值最大化，对于损失比对收益而言，其变化可能更大。更现实的普通体重方案。我们提出的模型不仅会生成DMU，总潜在客户价值更高，但满意度也更高。当前的研究表明，前景理论可以适当地扩展到DEA研究领域，为将来的DEA研究提供适当的指导。