Resilience Appraisal for Iranol Oil Company: Application of Adaptive Neuro-Fuzzy Inference System

Subject Areas : Decision theory

Mohammad Ghasemi Hamedan ¹ , Mahdi Homayounfar ^{2 *} , Mohammad Taleghani ³

1 - Department of Industrial Management, Rasht Branch, Islamic Azad University, Rasht, Iran
2 - Department of Industrial Management, Rasht Branch, Islamic Azad University, Rasht, Iran
3 - Department of Industrial Management, Rasht Branch, Islamic Azad University, Rasht, Iran

Accepted : 2025-04-07

Keywords:

Abstract :

In recent years, resilience has gained significant attention for understanding how organizations can prepare for and respond to turbulent environments. Despite extensive empirical research, previous studies highlight the need for greater clarity in measuring organizational resilience. This study proposes an advanced framework for resilience assessment in Iranol Oil Company, a leading lubricant and petroleum product manufacturer in Iran, during 2024. Unlike conventional approaches, this research employs an Adaptive Neuro-Fuzzy Inference System (ANFIS), which effectively integrates expert knowledge and data-driven learning to enhance predictive accuracy. The ANFIS model, leveraging Gaussian membership functions, captures complex, nonlinear relationships between resilience factors, making it superior to traditional statistical and fuzzy logic models in handling uncertainty and imprecise data. A structured methodology was implemented, involving a two-round fuzzy Delphi method to refine resilience indicators across five dimensions: organizational adaptability, collaborative factors, change management, HR management, and production management. Data collection included 126 qualified employees, with a training-testing-validation split ensuring model accuracy and generalizability. The ANFIS model demonstrated exceptional predictive performance, with mean square errors of 0.00244, 0.00279, and 0.00113 for training, testing, and validation datasets, respectively. Sensitivity analysis confirmed the robustness of the model under extreme conditions. Practical implications of this study extend beyond Iranol, providing a scalable approach for resilience assessment in various industries. By adopting ANFIS-based resilience evaluation, organizations can gain data-driven insights into their adaptability, innovation potential, and workforce engagement. This methodology enables companies to proactively identify vulnerabilities, optimize crisis management strategies, and enhance long-term sustainability in uncertain environments.