REAL-TIME ADAPTIVE RISK ASSESSMENT OF A VESSEL BASED ON INTELLIGENT MONITORING AND A DIGITAL TWIN

Abstract

This article examines the issue of ensuring a vessel’s operational safety in the context of an increasingly dynamic maritime environment, digitalization, and rising levels of uncertainty. It demonstrates that traditional risk assessment approaches, based on static scenarios, do not provide the necessary adaptability and timeliness for real-time decision-making. The feasibility of transitioning to integrated models that combine smart monitoring, digital twins, and methods of intelligent data analysis is substantiated. A mathematical model for adaptive ship risk assessment is proposed, based on the integration of streaming data from sensor systems, AIS, environmental parameters, and digital twin forecasts. The model implements a closed-loop “data–assessment–forecast–decision” cycle and ensures the formation of an integrated risk indicator, taking into account multi-criteria factors, their weights, and their dynamics over time. The paper employs the Analytic Hierarchy Process (AHP) to determine the weighting coefficients of risk factors, the FMEA approach to assess the criticality of failures, and Bayesian updating to adapt the probabilities of risk events based on new data. The proposed integrated risk indicator takes into account both the current state of the vessel and forecast assessments of the situation’s development, ensuring a proactive approach to safety management. The simulation results confirm that the proposed adaptive model allows for reducing peak risk values, shortening response times to hazardous situations, and increasing the system’s resilience compared to static approaches. The practical significance of the study lies in the possibility of integrating the model into decision support systems, integrated bridge systems, and digital ship management platforms, while the results obtained form the basis for the further development of intelligent systems for ensuring ship operational safety