A Bayesian network for reliability assessment of man-machine phased-mission system considering the phase dependencies of human cognitive error

Existing researches on the reliability assessment of phased-mission systems (PMSs) focus mainly on the phase dependencies of the machine state. However, with regard to the man-machine PMS (MMPMS), it also has non-negligible phase dependencies of human cognitive error. For example, the error of omission in previous phase may result in an identical error if the operator experiences a similar working scenario in a subsequent phase. To address the phase dependencies of human cognitive error, a novel method for the reliability assessment of MMPMS is proposed. First, the phase dependencies of human cognitive error are analyzed and categorized into four types based on the framework of situation awareness. A decision tree is then developed to quantify the dependence level. Second, the Bayesian network (BN) is used to construct the system reliability model for each phase from the perspective of mental model. Subsequently, the phase dependencies of machine state and of human cognitive error are mapped to BN to integrate constructed single-phase models as a multi-phase system reliability model. Third, the reliability of MMPMS can be assessed based on the conditional probabilities of all the nodes. Finally, the proposed method is exemplified with a multi-phase reconnaissance mission of an unmanned aerial vehicle.

Automated summary

This study proposes a novel method for assessing the reliability of man-machine phased-mission systems, focusing on addressing the phase dependencies of human cognitive errors. By analyzing and categorizing the types of cognitive error dependencies and constructing system reliability models using Bayesian networks, a comprehensive assessment approach is successfully developed.