Dynamic enterprise resilience assessment for port systems: A framework integrating Bayesian networks and Dempster-Shafer evidence theory

Ports act as vital nodes in the global transportation network, facilitating 80 % of international trade and supporting economic development. Despite their importance, port enterprises face growing vulnerabilities to global disruptions. Enterprise resilience (ER) is a critical capability that enables these dynamic and complex systems to address such challenges. This study develops a comprehensive framework for dynamically assessing ER, addressing the urgent need for enhanced resilience in port enterprises. The proposed framework integrates Dynamic Bayesian Networks (DBNs) with the Dempster-Shafer evidence interval theory, enabling the incorporation of both objective data and subjective expert judgments while managing uncertainty and conflict. Two time-evolution resilience models are introduced, encompassing multidimensional factors across economic, environmental, social, and technological domains. Case studies involving four major Chinese port enterprises—Shanghai, Ningbo Zhoushan, Tianjin, and Guangzhou Port—illustrate the framework's applicability. The analysis reveals varying temporal patterns in ER, identifies critical factors such as technological innovation and learning capabilities, and highlights the dynamic nature of resilience. This research contributes to ER theory by emphasizing the significance of learning capabilities in the dynamic adaptation of systems. It offers a novel approach to resilience research and management, providing a transferable framework for decision-makers in maritime transportation and other complex systems.