Rokhsaneh, Yousef Zehi
(2023)
Robust Optimization Approach In Data Envelopment Analysis Models: Extension To The Cases With Uncertain Production Trade-offs, Integer Data And Negative Data.
PhD thesis, Universiti Sains Malaysia.
Abstract
Data envelopment analysis (DEA) is a popular performance measurement technique and since it was first introduced, DEA models have been extensively applied in real-world managerial problems. One of the challenges in applying DEA models in real-world problems is uncertainty and inaccuracy in data which can be due to error in measurement, calculation, prediction etc. As uncertainty is an inevitable factor in many optimization problems, therefore the uncertainty in data should be taken into consideration to ensure reliable optimal solutions and benchmarking. Robust optimization is one of the most recent approaches for handling uncertainty in DEA models which immunize the uncertain parameters over a pre-specified uncertainty set to determine an optimal solution which is guaranteed to be the best for all or most of the possible realizations of the uncertain parameters. Applying robust optimization approach in DEA models resulted to Robust DEA field which is a relatively young yet growing field in DEA, introduced in 2008. The goal of this thesis is to fulfil some of the theoretical and practical gaps in robust DEA field. The previous works on robust DEA models only considered inputs and outputs data to be uncertain, thus one of the objectives of this thesis is to assess the effect of uncertainty in the other involved parameters in the optimization such as weights assigned to inputs and outputs and production trade-offs. Moreover, a comparative analysis between the proposed robust DEA model and other approaches of handling uncertainty in data such as interval DEA will be provided.
Actions (login required)
 |
View Item |
