Fuzzy Bayesian-based Approach for Collision Probability Prediction between Commercial and Fishing Vessels

A fuzzy Bayesian network-based prediction approach is proposed, which can be used to estimate the probability between commercial and fishing vessel collisions (CAFV). The kernel of this proposed method is first to collect the collision investigation reports between commercial and fishing vessels, to construct the fault tree between commercial and fishing vessel collisions based on factor analysis, and to use fuzzy sets to quantify the basic events and to calculate their occurrence probabilities. Afterwards, the fault tree is transformed into Bayesian network and the conditional probability table is constructed using Noisy-OR gate. Finally, the model is applied to Rizhao Port, the results are reasonable by comparing the results with the real data. The findings of this paper can provide a reference for reducing the collisions between commercial and fishing vessels.