Generalized matrix-based Bayesian network for multi-state systems

To achieve a resilient society, the reliability of core engineering systems should be evaluated accurately. However, this remains challenging due to complexity and large scale real-world systems. Such can efficiently modelled by Bayesian network (BN), which formulates probability distribution through graph-based representation. On other hand, issue addressed matrix-based (MBN), allows for efficient quantification flexible inference discrete BN. MBN applications have been limited binary-state systems, despite essential role multi-state Therefore, paper generalizes introducing concept composite state. The definitions operations developed are modified accommodate state, while formulations parameter sensitivity also MBN. facilitate generalized MBN, three commonly used techniques decomposing an event space employed quantify i.e. utilizing definition, branch bound (BnB), decision diagram (DD), each being accompanied example system. numerical examples demonstrate efficiency applicability supporting source code data download at https://github.com/jieunbyun/Generalized-MBN-multi-state.