UNCERTAINTY ANALYSIS FOR A PERIODIC REPLACEMENT PROBLEM WITH MINIMAL REPAIR: PARAMETRIC BOOTSTRAPPING APPROACH

In this paper we consider a statistical estimation problem for a periodic replacement problem with minimal repair which is one of the mostfundamental maintenance models in practice, and proposetwo parametric bootstrapping methods which are categorizedinto the simulation-based approach and re-sampling-basedapproach, respectively. Especially, we concern two data analysis techniques: direct data analysis of the minimalrepair data which obeys a non-homogeneous Poisson process and indirect data analysis after data transformation to a homogeneous Poisson process. Through simulationexperiments, we investigate statistical features of the proposedparametric bootstrapping methods. Also, we analyze the realminimal repair data to demonstrate the proposedmethods in practice