考虑决策单元非同质性的DEA效率评价与效率改进理论、方法与应用

Data envelopment analysis (DEA) is one of important tool for organizations’ efficiency evaluation and improvement. In traditional DEA, decision-making units (DMUs) are assumed to be homogeneous, which is always criticized, while DEA efficiency evaluation and improvement considering the non-homogeneity among DMUs get more and more attentions. In this project, the non-homogeneities among DMUs are studied. Firstly, the factors leads to non-homogeneity will be abstracted from the real problems and will be divided into the internal factors and the external factors, and also be quantitative characterized, and the impact mechanism that each type of non-homogeneity on the efficiency evaluation and efficiency improvement will be considered. Secondly, we will study the methods of efficiency evaluation and efficiency improvement considering different types of non-homogeneity respectively, which can construct DEA theories and methods eliminating the assumption of homogeneity effectively, and its feasibility and effectiveness will be demonstrated. Finally, the proposed theories and methods will be applied to the cases of bank, university and hospital to verify its effectiveness. This project will correct and improve the traditional DEA methods, which is also a useful supplement to the theory and method of efficiency evaluation and improvement and has important theoretical significance and academic value. Besides, it will also provide important theoretical and methodological support for the efficiency evaluation, efficiency improvement and management decision making of non-homogeneous organizations.

数据包络分析（DEA）是组织单元效率评价和效率改进的重要方法之一。传统DEA方法针对决策单元的同质性假设受到质疑，而考虑决策单元间非同质性的DEA效率评价和效率改进方法在理论界和现实界得到越来越多的重视。本项目以决策单元间所存在的非同质性为研究对象。首先，从现实问题中抽象出导致非同质性的各种因素，对这些因素进行内外部分类和定量刻画，并研究每类非同质性对效率评价和效率改进的影响机理。其次，分别研究考虑不同非同质性的效率评价与效率改进方法，构建出能够有效解除同质性假设的DEA理论和方法体系，并论证其可行性和有效性。最后，将理论和方法分别应用于银行、高校以及医院等不同领域，以验证其有效性。本项目将对传统DEA方法进行修正和完善，并对效率评价和效率改进理论和方法进行有益补充，具有重要的理论意义与学术价值，也为实际运作中非同质决策单元的效率评价、效率改进和管理决策提供重要的理论和方法支持。

目前有关非同质数据包络分析方法（DEA）的研究尚未建立起一套完善的科学体系。本项目旨在从非同质DEA的理论完善、方法确立和评价改进三方面进行研究。首先，从非同质性的定义与分类、非同质程度的定量分析、以及非同质性对效率评价与改进的影响机理分析等方面完善了非同质DEA的研究体系。在新的理论体系指导下，主要研究了具有外部非同质性、内部非同质性、混合因素非同质性的效率评价和改进方法，取到了一定的成果。此外，该项目将所得的理论和评价方法应用于酒店管理、环境效率评价和资源分配等领域，具有一定的应用广泛性。围绕非同质DEA理论、方法和应用开展研究工作，项目组在国际学术期刊（接受）发表论文43篇，其中SCI检索期刊论文33篇，SSCI检索期刊论文10篇。研究工作得到了2015年度中组部（第二批）青年拔尖人才计划以及2018年度中国科学技术大学学术领军人才项目（A类）的资助。并且，项目主持人相继获得2017年度安徽省自然科学奖一等奖、2018年度中国科学院优秀研究生指导教师奖、2019年度中国科学技术大学唐立新优秀学者奖以及Highly Commended Award winner of 2019 Emerald Literati Awards等国内外奖项。