|
Development of returns to scale in two-stage network DEA via parametric analysis
|
T. Mollaalizadeh Koloukhi   , A. Mostafaee , S. Saati |
| Department of Mathematics, N.T.C., Islamic Azad University, Tehran, Iran |
|
|
Abstract: (269 Views) |
| Network Data Envelopment Analysis (DEA) has attracted considerable attention in both methodological research and practical applications of performance evaluation. This paper investigates a fundamental class of network DEA models, namely the two-stage network framework. In conventional DEA, where the production process is viewed as a “black box,” returns to scale (RTS) plays a critical role in guiding managerial decisions on whether to expand or contract operations. This study extends the traditional concept of RTS to a two-stage network by examining input variations from three perspectives: stage 1, stage 2, and the overall system. The proposed approach employs parametric analysis to capture how these variations affect the relationships among inputs, intermediate measures, and final outputs. To ensure practical applicability, the method can be implemented through existing linear programming formulations and remains computationally feasible even for larger-scale problems. In addition, we develop a linear programming model that supports central managers in coordinating resource allocation across different stages, thereby achieving system-wide improvements. A numerical example illustrates that RTS classifications at the system and sub-process levels may diverge, offering distinct insights into pathways for enhancing productivity. |
|
| Keywords: Data Envelopment Analysis, Returns to scale, Network |
|
|
Full-Text [PDF 996 kb]
(93 Downloads)
|
Type of Study: Research |
Subject:
General
Received: 2025/04/2 | Accepted: 2025/06/20 | Published: 2025/07/10
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
