Reliability Evaluation of Linear or Circular Consecutive k-out-of-n: F System Using Dynamic Bayesian Network

Abstract

In the field of reliability theory, one of the most significant topics to discuss is the process of determining the reliability of a complex system based on the reliabilities of its individual components. The consecutive k-out-of-n:F system is used in telephone networks, photographing in nuclear accelerators, spacecraft relay stations, telecommunication system consisting of relay stations connecting transmitter and receiver, microwave relay stations, the design of integrated circuits, vacuum systems in accelerators, oil pipeline systems and computing networks. The reliability estimation of the consecutive k-out-of-n:F system is studied because it plays an important role in many physical systems. Dynamic Bayesian networks are graphical models for time-varying probabilistic inference and causal analysis under system uncertainty. The dynamic Bayesian network is built for the proposed system since time is continuously measured. The consecutive k-out-of-n:F system depends on the k components, because the system fails when the consecutive k components fail, otherwise the system works. The contributions are the dynamic Bayesian network construction of the proposed system and the reliability analysis of the linear and circular consecutive k-out-of-n:F system. Furthermore, Dynamic Bayesian network-based reliability is shown to be significantly higher than the reliability achieved by Malinowski, Preuss and Gao, Liu, Wang, Peng and Amirian, Khodadadi, Chatrabgoun. The Dynamic Bayesian network-based Reliability of linear and circular consecutive k-out-of-n:F system is also compared.