Development of Integrated Evaluation Model for Ship Berthing Risk

Abstract When a ship docks at a port terminal, berthing velocity refers to the speed at which the ship first contacts the fender. This velocity requires careful management as it directly affects the likelihood of berthing accidents. Although numerous studies have focused on berthing velocity, there remains a significant lack of comprehensive methodologies for assessing berthing risk. While various factors influence the berthing process, existing research predominantly emphasizes berthing velocity and energy, often overlooking the potential risks posed by other contributing factors. This study aims to introduce a novel method for berthing risk evaluation by thoroughly analyzing these influencing factors. Machine learning techniques will be employed to estimate the relative weights of each factor. Over a five-year period, berthing processes at a tanker terminal were closely observed, and relevant data were systematically collected. This proposed method not only identifies risks associated with berthing velocity and energy but also accounts for additional hazards, such as adverse marine weather conditions, individual pilot characteristics, and the presence of very large vessels. This approach is expected to significantly contribute to the prevention of berthing accidents in actual port terminal operations. Furthermore, the main findings of this study can serve as foundational data for advancing autonomous ships technologies.