Optimal replacement interval for multiple repairable units with incomplete data

This paper studies the optimal replacement interval for multiple units. The units are identical and the failure occurrences follow a non-homogeneous Poisson process (NHPP). The failure time data has a mixture of left truncation and right censoring. Maximum likelihood (ML) method is used to estimate the NHPP parameters. Two hazard rate functions (power law and log-linear) are considered. After the parameters are estimated, the expected life cycle cost is estimated. Periodic replacement with minimal repairs is considered and the optimal replacement interval is obtained that provides the minimal life cycle cost. The cost of minimal repairs and the cost of replacement are considered in the estimation of the expected life cycle cost. Failures of rolling stock brake control module cards are used as real case study.