This paper considers a multi‐repairmen problem comprising of M operating machines with W warm standbys (spares). Both operating and warm standby machines are subject to failures. With a coverage probability c, a failed unit is immediately detected and attended by one of R repairmen if available. If the failed unit is not detected with probability 1‐c, the system enters an unsafe state and must be cleared by a reboot action. The repairmen are also subject to failures which result in service (repair) interruptions. The failed repairman resumes service after a random period of time. In addition, the repair rate depends on number of failed machines. The entire system is modeled as a finite‐state Markov chain and its steady state distribution is obtained by a recursive matrix approach. The major performance measures are evaluated based on this distribution. Under a cost structure, we propose to use the Quasi‐Newton method and probabilistic global search Lausanne method to search for the global optimal system parameters. Numerical examples are presented to demonstrate the effectiveness of our approach in solving a highly complex manufacturing system subject to multiple uncertainties.
