An AI-ISM Methodology for Structural Modeling of Sustainable Supply Chain Management Drivers

Achieving Sustainable Supply Chain Management (SSCM) has become a strategic priority for organizations integrating environmental, social, and economic objectives. Interpretive Structural Modeling (ISM) is a powerful methodology for analyzing interdependencies among SSCM drivers by structuring them into a multilevel hierarchical framework. However, traditional ISM relies on subjective expert judgments, making it prone to inconsistencies and biases. To address these challenges, this study proposes a novel Hybrid AI-ISM methodology, integrating Genetic Algorithms (GA) to optimize the reachability matrix by minimizing transitivity violations and reciprocal inconsistencies. The research adopts a structured, expert-driven approach, engaging 25 domain specialists from executive, production, and supply chain management sectors within the steel industry. By leveraging AI-driven optimization, the proposed framework enhances accuracy, objectivity, and scalability, refining hierarchical structuring for effective SSCM decision-making. This study contributes to the SSCM literature by introducing a bias-resistant, computationally enhanced ISM framework, facilitating granular decision-making across organizational levels. Findings reveal notable differences in expert perspectives across professional roles. Executives emphasize strategic sustainability initiatives, particularly the Adoption of Renewable Energy, while production managers prioritize operational aspects, including Waste Minimization and Supply Chain Flexibility. Conversely, supply chain managers focus on stakeholder engagement and risk mitigation, highlighting Community Engagement and Supply Chain Disruption Management as critical drivers. The study provides actionable insights for policymakers, industry leaders, and supply chain professionals seeking to drive sustainable transformations in global supply chains.