Dynamic hesitant fuzzy Bayesian network and its application in the optimal investment port decision making problem of “twenty-first century maritime silk road”

Most traditional decision making methods cannot deal with intensive uncertain data that varies with time effectively, and they only derive the static risk analysis by different aggregation operators. There is little research of the dynamic decision making problems with massive uncertain information. To manage with different kinds of uncertain knowledge and hesitancy in the dynamic decision-making process, this paper combines the advantages of hesitant fuzzy sets (HFSs) in depicting information and Bayesian Network (BN) in uncertain reasoning. Considering the uncertain information of risk factors varies with time, the concept of Dynamic Hesitant Fuzzy Bayesian Network (DHFBN) is proposed to deal with dynamic decision making problems under the hesitant fuzzy environment. Then, an improved Particle Swarm Optimization (PSO) algorithm and the Expectation-Maximization (EM) algorithm are adopted for the structure learning and parameters learning of DHFBN respectively. Based on the learned optimal DHFBN, a dynamic reasoning and prediction method is developed. Furthermore, a case about the optimal port investment decision making problem of “21st Century Maritime Silk Road” is presented to illustrate the application of the proposed method. Finally, we also conduct a comparative experiment to testify the validity and advantages of the method in detail.