Research on Ship Accident Scenario Deduction Based on Similarity Algorithm

Abstract Emergency response and rescue in ship accidents are complex, urgent, and high-risk, making it difficult to comprehensively assess the situation on site based solely on personal experience. Existing research on decision support primarily focuses on case similarity, leading to significant deviations in predicting accident development trends. Therefore, this paper proposes a study on accident scenario deduction based on similarity calculations. Specifically, it first adopts a method for decomposing accident scenarios into multiple discrete elements. Then, it designs a scenario deduction algorithm based on similarity calculations to deduce accident development trends at a finer granularity level. The feasibility of this algorithm is verified through practical applications. Ultimately, it is proven that the scenario deduction algorithm proposed in this paper can efficiently deduce accident scenarios, providing powerful support for emergency response and rescue in ship accidents.