Modeling equipment availability for a system that follows a Kijima Type II imperfect repair process

This thesis explores the impact of a Kijima Type II imperfect repair model on the availability of a repairable system. A simulation model for equipment availability is constructed in Visual C. Initial results show that system availability converges quickly to a steady state value when a Kijima Type II repair process is used. Based on the simulation results, an approximate analytic equation for equipment availability was developed and analyzed. Then, we construct a meta-model for equipment availability using the equipments reliability and maintainability parameters. A set of 50 randomly generated experiments within the circumscribed central composite design space are analyzed to asses the robustness of the meta-model. The meta-model was found to be accurate and robust to changes in parameters. Finally, a brief investigation into the impact of preventive maintenance is conducted. Both perfect and imperfect preventive maintenance are analyzed.