Development of a Model to Evaluate the Efficiency of Maintenance Units for Automated Guided Vehicles in Industry 4.0Using an Integrated Approach Based on Data Envelopment Analysis and Game Theory

Massahi, Maziyar; Tohidi, Hamid; Golrizgashti, Seyedehfatemeh

doi:10.22034/ijrrs.2025.549194.1217

Development of a Model to Evaluate the Efficiency of Maintenance Units for Automated Guided Vehicles in Industry 4.0Using an Integrated Approach Based on Data Envelopment Analysis and Game Theory

Articles in Press

Document Type : Original Research Article

Authors

Maziyar Massahi ¹

Hamid Tohidi ¹

Seyedehfatemeh Golrizgashti ²

¹ Department of Industrial Engineering, ST.C., Islamic Azad University, Tehran, Iran

² Department of Industrial Engineering, ST.C., Islamic Azad University, Tehran, Iran.

10.22034/ijrrs.2025.549194.1217

Abstract

With the advent of the Fourth Industrial Revolution, the use of specialized tools and technologies in industrial systems has significantly expanded. Automated Guided Vehicles (AGVs) have emerged as a vital component in factories aligned with Industry 4.0 standards. However, like any other equipment, AGVs require systematic and ongoing maintenance to ensure optimal performance.

This study aims to evaluate the efficiency of AGV maintenance units through an integrated model that combines Data Envelopment Analysis (DEA) and Game Theory. The model is applied under conditions that assess the impact of various maintenance strategies—corrective, preventive, and predictive—as well as the spatial configuration of maintenance and depot sites. Game Theory is incorporated to account for cooperative effects among decision-making units (DMUs), enhancing the realism and analytical depth of the evaluation.

The results of the integrated DEA–Game Theory analysis indicate that incorporating the proposed network structure provides valuable insights into the sources of inefficiency across DMUs. This enables the identification of optimal maintenance strategies and the most effective spatial arrangements for maintenance and depot facilities. A comparative analysis with the classical DEA model reveals that, out of three efficient DMUs in the traditional framework, only one remains efficient under the integrated model, while the other two exhibit inefficiencies across at least three sub-networks.

Ultimately, the findings suggest that adopting a predictive maintenance approach—alongside corrective and preventive strategies—yields superior performance. Moreover, the co-location of AGV maintenance and depot facilities with existing production machinery maintenance sites is identified as the most efficient configuration.

Keywords

Maintenance Efficiency Evaluation

Data Envelopment Analysis

Reliability of automated guided vehicles

Game Theory

Network DEA Model

Hybrid Decision-Making Approach

Subjects