963. NONCONVEX CENTRALIZED ALLOCATIONS WITH MULTIPLE TECHNOLOGIES
Invited abstract in session TB-60: DEA under uncertainty, stream Data Envelopment Analysis and its applications.
Tuesday, 10:30-12:00Room: Western LT
Authors (first author is the speaker)
| 1. | Giovanni Cesaroni
|
Abstract
Until recently, the literature on centralized resource allocations (CRA) within non-parametric methods was mainly characterized by the assumption of a common convex technology shared by a set of decision-making units (DMUs) under control of a given planner/manager, and determined by means of the Data Envelopment Analysis (DEA) technique. On one hand, the seminal contribution of Fang (2013), and that of Lozano (2014), extended this convex approach by considering the existence of multiple technologies in the set of DMUs under analysis. On the other hand, Cesaroni (2024) introduced an explicit solution for the CRA of a set of DMUs sharing a common nonconvex technology, such as the free disposal hull (FDH). The purpose of the present work is extending the nonconvex CRA to the case of multiple technologies. Besides determining the aggregate technology in the presence of multiple FDH technologies, we develop the optimization program as a MILP problem, which is simpler and allows to substantially reduce the number of binary decision variables when compared with the approach of Fang (2013) and Lozano (2014). We perform computations on a data set describing a multi-input and multi-output technology relating to Italian local public transit. In particular, following Lozano (2014), we compare the results and the computational features of nonconvex CRAs obtained under either a common or multiple technologies.
Keywords
- Data Envelopment Analysis
- Efficiency Analysis
- Economic Modeling
Status: accepted
Back to the list of papers