Generalized Data Envelopment Analysis Method by Using Sample Units
Zhanxin Ma, MA Sheng-yun
Abstract
Zhanxin Ma, MA Sheng-yun
Abstract
Traditional DEA method is evaluation method based on a self-assessment system, decision-makers can not choose a frame of reference freely.In order to solve the DEA method can also based on self-assessment system and other frame of reference for evaluation,the concept of the generalized DEA efficiency is presented at first.Then a class of generalized data envelopment analysis models based on the sample units are given,including generalized DEA model for input-oriented and output-oriented situation,and generalized additive DEA model.Finally,the relationship between these models and the traditional DEA models is analyzed,the relationship between the generalized DEA efficiency and corresponding Pareto efficiency of multi-objective programming is explored,and the property of projection and the method of ranking decision making units are given.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Traditional DEA method is evaluation method based on a self-assessment system, decision-makers can not choose a frame of reference freely.In order to solve the DEA method can also based on self-assessment system and other frame of reference for evaluation,the concept of the generalized DEA efficiency is presented at first.Then a class of generalized data envelopment analysis models based on the sample units are given,including generalized DEA model for input-oriented and output-oriented situation,and generalized additive DEA model.Finally,the relationship between these models and the traditional DEA models is analyzed,the relationship between the generalized DEA efficiency and corresponding Pareto efficiency of multi-objective programming is explored,and the property of projection and the method of ranking decision making units are given.
Key concepts: Data envelopment analysis, Ranking (information retrieval), Sample (material), Computer science, Projection (relational algebra), Frame (networking), Mathematical optimization, Mathematics