Efficiency decomposition in three-stage network with fuzzy desirable and undesirable output and fuzzy input in data envelopment analysis

PurposeThe purpose of this paper is to improve the recent models for the evaluation of the efficiency of decision making units (DMUs) comprising a network structure with undesirable intermediate measures and fuzzy data.Design/methodology/approachIn this paper a three-stage network structure model with desirable and undesirable data is presented and is solved as linear triangular fuzzy planning problems.FindingsA new three stage network data envelopment analysis (DEA) model is established to evaluate the efficiency of industries with undesirable and desirable indicators in fuzzy environment.Practical implicationsThe implication of this study is to evaluate the furniture services and the chipboard industries of wood lumber as a three-stage process.Originality/valueIn some cases, DMUs include two or multi-stage process (series or parallel) operating with a structure called a network DEA. Also, in the real world problems, the data are often presented imprecisely. Additionally, the intermediate measures under the real-world conditions include desirable and undesirable data. These mentioned indexes show the value of the proposed model.