A novel hybrid AHP-entropy weighted dynamic network DEA framework for comprehensive efficiency assessment
•A novel hybrid AHP-Entropy dynamic network DEA model is proposed.•The proposed Methodology blends expert insights with data-driven weighting to fix traditional shortcomings.•Incorporated carryovers and intermediate products for deeper efficiency insights.•Investigate the efficiency of the school ed...
Saved in:
| Published in: | Expert systems with applications Vol. 296; p. 129137 |
|---|---|
| Main Authors: | , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Ltd
15.01.2026
|
| Subjects: | |
| ISSN: | 0957-4174 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | •A novel hybrid AHP-Entropy dynamic network DEA model is proposed.•The proposed Methodology blends expert insights with data-driven weighting to fix traditional shortcomings.•Incorporated carryovers and intermediate products for deeper efficiency insights.•Investigate the efficiency of the school education system across 28 Indian states, utilizing data from 2018 to 2021.
This study presents a novel hybrid AHP-Entropy Weighted Dynamic Network Data Envelopment Analysis (DEA) framework for comprehensive efficiency assessment. Traditional DEA models often fail to capture the intricate interdependencies and multidimensional, dynamic nature of real-world systems, typically treating performance measurement as a static process. Even advanced network and dynamic DEA models face challenges in weight determination, as flexible schemes can distort efficiency assessments and yield unrealistic results. Existing weighting methods either rely solely on subjective judgments, potentially introducing bias, or purely on objective data, which may ignore contextual knowledge and stakeholder preferences. To address these fundamental limitations, this research synthesizes the Analytic Hierarchy Process (AHP) with entropy methods, integrating subjective expert judgments and objective data-driven weights. This hybrid approach mitigates potential biases and enables a more nuanced understanding of performance by capturing interdependencies across different stages and periods. These integrated weights are incorporated into dynamic network DEA frameworks employing directional distance vectors for flexible efficiency measurement. Applying this framework to evaluate the efficiency of the Indian state-level school education system from 2018 to 2022, utilizing a dynamic network DEA model, reveals significant disparities in performance across states. Key findings indicate that only a few states, such as Himachal Pradesh, Nagaland, Goa, Tripura, and Mizoram, achieved complete efficiency across all educational divisions, while others, like Gujarat, Karnataka, and Bihar, were found to be inefficient, particularly in Division 4 (Higher Secondary). The novel integration of weighting methodologies within a dynamic network DEA framework contributes significantly to the literature on performance measurement by advancing both methodological understanding and practical application in complex organizational performance assessments. |
|---|---|
| ISSN: | 0957-4174 |
| DOI: | 10.1016/j.eswa.2025.129137 |