Reliability analysis of multi-state systems based on intuitionistic fuzzy bayesian networks

In view of the difficulty in determining the precision probability of each root node in different states when Bayesian Networks were employed to analyze multi-state systems reliability, a method based on intuitionstic fuzzy Bayesian networks is proposed, in which the comprehensive probability is figured out with intuitionistic fuzzy weighted average operator on the basis of the intuitionistic probability and corresponding weight of each of different states of each root node given by each of several experts, the precision probability of each of several different states of each root node is obtained with the score function and the corresponding data processing,. The probability of leaf node in each of several different states is obtained with Bayesian networks, and the probability in each of several different states and the importance of each root node are calculated for the leaf node in each of several different states. The simulation results and their analysis show preliminarily that the method can accurately obtain the probability of the root node in each of several different states, and can effectively improve the ability of Bayesian networks to deal with uncertainty problems.