SEQUENTIAL NONPARAMETRIC-ESTIMATION OF AN OPTIMAL AGE REPLACEMENT POLICY - A SIMULATION STUDY

Maintenance policies can reduce the number of system failures and minimize costs by scheduling times for replacement. Even in the most basic scenario, where the underlying system lifetimes are assumed to be independent and identically distributed (iid), the problem of updating maintenance policies using past maintenance history has not been adequately solved. In this paper, Monte-Carlo methods are used to study a particular nonparametric procedure of Aras and Whitaker (1991) which updates an optimal age replacement policy after each replacement. In particular, we will focus on finding conditions under which this sequential procedure does well.