基于指标差异化的双集团DEA效率研究—以某高校人文类-理工类学科效率评价为例

Two-group’s DEA efficiency analysis with different input and output consideration —a case study of the efficiency evaluation of the discipline between the humanities and engineering of the college

摘要:

传统DEA(数据包络分析)效率分析假定指标体系固定不变,然而现实生活中由于决策单元所处的情境不同,存在着主张不同指标体系的多个集团,如高校中理工类与人文类学科。本文研究了如何公平有效实现具有不同投入产出指标集的两个集团间决策单元DEA效率比较问题,通过设计“求同存异”的指标融合运算,针对形成的多种效率评价结果,运用OWA算子,进行集成计算,实现对双集团DEA效率的综合分析。所提出的方法拓展了传统DEA固定指标体系的限制,拓展了DEA的应用范围。最后通过案例演算了方法的可行性。

Abstract:

Traditional DEA (data envelopment analysis) efficiency analysis assumes that the index system (inputs and outputs) is fixed, but in real life different groups, which contains several decision-making units (DMUs) may adopt different index system to DEA analysis. For example, in a university, engineering and science related disciplines and social and humanities related discipline are typical examples. In this paper, a method is proposed to solve two-groups’ DEA efficiency comparison in a fair and effective way. Firstly, a index infusion algorithm is designed to obtain different DEA analysis scenarios by the concept of “seeking common ground while putting aside differences”. Then, OWA operator is applied to integrate different DEA results to reach a final result for each DMU. The proposed method expands the traditional DEA model by considering different index system. Finally, the feasibility of the method is demonstrated through a case study.