Overall Efficiency and its Decomposition in Two-Stage Network DEA Model

Guo-liang Yang, Song Yaoxuan, Dong Xu, Jian-Bo Yang

Research output: Contribution to journalArticlepeer-review


This paper aims to propose a new approach to decompose an overall DEA model into equivalent two-stage models. In this approach we use a minimax reference point method to set the weights and reliabilities of the two stage models so that the combined efficiency of the two stages is equal to the overall efficiency. The equivalent multi-stage models are useful to support planning for performance improvement. An illustrative example is first explored to compare the results from the new approach with those of 4 other existing approaches. The main finding from the comparisons is that the new decomposition approach of this paper satisfies the proposed several assumptions. A case study is then conducted on a two-stage process of steel manufacturing to illustrate the validity and applicability of the proposed approach.
Original languageEnglish
JournalJournal of Management Science and Engineering
Issue number3
Early online date8 Dec 2017
Publication statusPublished - 2017


  • Two-stage network DEA
  • overall efficiency, decomposition
  • evidential reasoning
  • performance improvement


Dive into the research topics of 'Overall Efficiency and its Decomposition in Two-Stage Network DEA Model'. Together they form a unique fingerprint.

Cite this