Research on Safety Level Evaluation of Construction Ships in the Channel of Yangtze River Estuary

During water construction operations, engineering ships often occupy part of the traffic channel, resulting in a surge in the density of ships in the channel which increase navigation and operational risks of ships. However, the existing research on risk evaluation in navigable waters is mostly oriented to the influence of the natural environment under navigation conditions and cannot systematically elaborate on the multi-source risks faced by construction ships in navigable waters. With a view to analyzing the complex working conditions during the construction of the deep-water channel in the Yangtze River Estuary, this paper introduces a multi-level Bayesian network to establish a safety level evaluation model. Firstly, ten factors affecting water construction safety in three categories are extracted and a multi-level index system is established; secondly, the optimal conditional probability is extracted from the information inferred by experts using the modified entropy weighting method; finally, the prior probability of the root node is calculated for four actual working conditions and based on Bayesian probabilistic inference theory, the Bayesian network is determined and the safety level of construction ships under four working conditions is calculated. The main influencing factors of safety level are analyzed, and the evaluation results are consistent with the actual engineering situation, which provides decision-making references for construction safety management in navigable waters.