Hidden markov models in reliability and maintenance

Although the hidden Markov models (HMM) are very popular in many applied areas their use reliability engineering is limited. Problems such as selection of HMM model by choosing appropriate number states, or problems prediction failures have not been widely covered literature. This paper concerned with HMMs where state system directly observable and instead certain indicators true situation provided via a control system. A can provide key information about dependability failed component system, related measures. maximum-likelihood estimator obtained its asymptotic properties studied. Finally, maintenance considered this context new preventive strategies defined efficiency measured terms expected cost. To prove finite sample performance methodology, an extensive simulation study developed.