Modeling ESG Performance Drivers in the Shipping Industry: An LLM‐Assisted q‐ ROFRS‐Delphi‐WINGS‐FCM Framework for Causal Analysis and Simulation

ABSTRACT The shipping industry plays a vital role in global trade and also exerts significant environmental and social pressures due to high fuel consumption, greenhouse gas (GHG) emissions, and intensive port activities. Implementing environmental, social, and governance (ESG) practices offers a strategic pathway to enhance sustainability performance in this sector; however, it remains difficult to determine which managerial and operational actions can most effectively drive improvements in ESG performance. To address this challenge, this study proposes a novel framework that integrates a large language model (LLM), an extended q‐ rung orthopair fuzzy rough set ( q‐ ROFRS)‐based Delphi‐weighted influence nonlinear gauge system (WINGS) model, and a fuzzy cognitive mapping (FCM) simulation. This approach systematically identifies critical success factors (CSFs), analyzes their cause–effect relationships, and simulates their dynamic influence on ESG performance in the shipping industry. The findings reveal that six CSFs, including structured board oversight, global policy engagement, comprehensive legal compliance framework, employee‐centric culture, effective communication strategy, and integrated water resource management, serve as key causal drivers within the ESG system. FCM‐based scenario simulations suggest that full activation of these modeled drivers could substantially improve projected ESG performance under the assumed scenario conditions. This research offers a dual contribution: Theoretically, it extends fuzzy decision‐making and AI‐driven causal modeling to complex sustainability systems; practically, it provides actionable guidance for shipping companies and policymakers to strengthen ESG integration, meet the International Maritime Organization's net‐zero targets, and advance the United Nations Sustainable Development Goals (SDGs).