Chance Constrained Stochastic DEA Model With Multiple Objective Programming And Its Application

In this dissertation, the development of DEA theories, DEA methods and DEA application are discussed. The basic DEA models and DEA efficiency theories are analyzed.Basing on the Multiple Objective DEA mode and thinking of the undesirable outputs in the practical production process, this paper constructs a Multiple Objective DEA model with undesirable outputs. In the meanwhile, the data of inputs and the outputs in practical situations are ever-changeful and unexpectedly, so we consider stochastic factors into the model, and construct chance strained stochastic DEA model with multiple Objective programming,Then, we apply reformulation of stochastic DEA model with chance-constrained techniques to the model, and have a effective analysis to the DMUS.Finally, we apply the new type of stochastic DEA model into practice. The paper analyze the experimental data in Study 193A of Aids Clinical Trail Group (ACTG), Because of the lack of the data, a deleting method is used to handle the large amount of experimental data. Firstly, DMUs are classified into 4groups by patients'age, then, classified by treatments and treatment phases through patients of the same age; lastly, we apply stochastic DEA model with chance-constrained techniques to the DMUS. Through this, the optimal medicine amounts are then provided to the AIDS patient for him to choose a treatment that is most effective for his age.