Replacement-Repair Policy Based on a Simulation Model for Multi-State Deteriorating Products Under Warranty

In this paper, a replacement-repair model is developed to study a warranty servicing policy for a class of multi-state deteriorating and repairable products, based on a computer simulation analysis. In each working state there is a determined probability for transition to each of the subsequent states, given that it has made a transition out of that state. There are two parameters that determine the manufacturer's decision to repair or replace a failed item, assuming that the buyer's claim is valid; the deterioration degree of the item and the length of the residual warranty period. Beside these two parameters, other inputs to the model are: the number of working and failure states, the di erent rates of transition from each working state, given that the life distribution is exponential, the length of the entire warranty period, the probabilities of expected costs, generated by a mathematical approach, for a few special situations including transition between states and the cost of repair and replacement in each failure state. The output of the model is the mean and standard deviation of the total simulated costs generated for a given set of inputs, such that the best values of the two mentioned parameters can be obtained, based on the statistical test. The model is validated by comparing its output with the optimal expected costs generated by a few special situations where this comparison is possible.