Advances and Applications in Statistics
Volume 52, Issue 3, Pages 171 - 201
(March 2018) http://dx.doi.org/10.17654/AS052030171 |
|
THE SELECTION OF PROJECTS WITHIN CONSTRAINED RESOURCES USING STOCHASTIC DATA ENVELOPMENT ANALYSIS
Maliheh Piri, Farhad Hosseinzadeh Lotfi, Mohsen Rostamy-alkhalifeh and Mohammad Hasan Behzadi
|
Abstract: The selection of an appropriate place to create new projects for the purpose of providing services (or products) for customers is an important issue that has been studied extensively. In certain performance evaluation situations when a resource constrained environment is encountered, it is necessary to select a subset of projects among a larger set of projects. It is expected that each project makes the best use of available resources. In the meantime, we encounter multiple inputs and outputs. As a result, we will need ways which provide a suitable benchmark for evaluating the performance of each unit by combining the inputs and outputs in the form of a single index. In addition, the presence of uncertain inputs and outputs or lack of accurate data on their values is problematic. Thus, the solution to these problems is the use of DEA (data envelopment analysis) with the help of probability theory. DEA is an approach to estimating the amount of inefficiencies involved in inputs and outputs generated by DMUs (decision making units). For each DMU based upon optimizing total output relative to total input, the DEA models provide the chance to make selections when multiple inputs and outputs are needed. The problem considered in this paper is the concept of chance constrained programming approaches used to expand the selection of projects within constrained resources in SDEA (stochastic data envelopment analysis), and its deterministic equivalent which is a mixed-binary nonlinear program. |
Keywords and phrases: stochastic data envelopment analysis, chance constraints, quadratic programming, project prioritization. |
|
Number of Downloads: 451 | Number of Views: 3366 |
|