On defect propagation in multi-machine stochastically deteriorating systems with incomplete information

In many manufacturing environments, costly job inspection provides information about the random deterioration of the machines. The resulting maintenance and inspection problem is extensively studied for a single machine system by using the framework of Partially Observable Markov Decision Processes (POMDPs). In this work, this concept is extended to multiple operations and multiple job types by considering two process flow topologies: (i) re-entrant flow, (ii) hybrid flow. The resulting (significantly large sized) POMDPs are solved using a point based method called PERSEUS, and the results are compared with those obtained by conventionally used periodic policies.

► We study job scheduling/inspection problems for systems deteriorating randomly.
► The previous studies focused on a single machine with a single operation.
► We extend the approach to multiple operations and multiple job types.
► The resulting POMDPs are solved using a point based method called PERSEUS.
► The results show significant improvement over the conventionally used periodic policies.