sensitive analysis of the objective function in a new network structure; dea-based approach

Data envelopment analysis (dea) is an improved technique for optimization and performance evaluation of peer decision making units (dmu). in two-stage network structure undesirable factors are jointly produced with desirable ones either as final outputs or as intermediate products. defining an appropriate objective function affects any decision for managing a project. sometimes improving the desirable products is more important than reducing undesirable factors. in some cases decision-makers focus on less usage of inputs because of the shortage of the resources. likewise there can be define many different cases for defining different functions. this paper tries to investigate an appropriate function in a two-stage model by considering various demands of managers. properly modeling of the objective is very important in the optimization problems in line with the valid treatment of the constraints. in the presence of additional inputs to the second stage in two–stage structure as an example, this approach searches for presenting some possible objective functions by considering the desirable and undesirable outputs. this paper scenario is conducting the weak disposability concept at both stages.