A Bayesian network-based TOPSIS framework to dynamically control the risk of maritime piracy

Piracy has long plagued the maritime industry, and has led to significant losses of life and goods. The risk of maritime piracy is largely predicted at present based on static analysis. This does not suitably address practical needs because the behavior and activities of maritime pirates are dynamic. Selecting an appropriate strategy for reducing the risk of piracy under a dynamic environment featuring uncertainty thus remains a key challenge. In this study, we propose a two-stage technique for order of preference by similarity to an ideal solution (TOPSIS) model based on the Bayesian network (BN). A data-driven BN is constructed in the first stage of the proposed method to identify the causal relationships influencing the behaviors of pirates. The second stage involves calculating a decision matrix of the strategies by using TOPSIS, where this enhances the strength of risk prediction and dynamic diagnosis by the BN. The main novelty of the proposed model is that it can provide quantitative measurements of strategies for reducing the probability of piracy in a dynamic environment. It provides a decision-making tool for researchers, shipping companies, and national navies to assess the risk of piracy and select effective measures to reduce its likelihood.