Condition based maintenance: implementation and optimization of a two-unit serial system model with multi-threshold policy

This work starts from an analytic model found in literature of a condition based maintenance two unit serial system which stochastically, independently and gradually deteriorates, the system is considered to by maintained in "as good as new" conditions by preventive and corrective maintenance interventions. For every inspection or substitution we will be considering the operative unit cost and also the setup cost. The numeric solution to our problem without some restricted hypothesis on the model results very complicated, if ever existent. The solution that we purpose here is to represent the model with an event driven simulator. With the use of a software it is possible to free the user from the mathematical aspects of simulation, and in the same time accomplish some improvements that analytically were not possible, do to the complexity of some mathematical aspects, such as: the use of gamma function that describes the wear of the systems industrial components; the possibility of considering a time not zero that statistically represents the time necessary for accomplishing a maintenance intervention, optimization of the global cost of maintenance with a series of condition based thresholds. All these improvements allow having more accurate and significant values of costs and time for a more valid prevision on the maintained system.