Modelling and Optimization of Warranty Policy for Heterogeneous Second-Hand Products Based on Historical Failure Information

Considering the heterogeneity in past and existing end users, a novel preventive maintenance model for second-hand products is proposed in this paper. The historical failure information before the acquisition and these between the resale and the inspection time are utilized to derive probabilities that the past end user and the exiting end user belong to “the general sub-population” (the sub-population with smaller failure intensity function). Preventive maintenance or replacement actions are carried out according to the relationships between the aforementioned probabilities and the corresponding thresholds. Taking the profit rate of second-hand products over the warranty period as the objective function, the level of the preventive maintenance and the probability thresholds are optimized jointly. A numerical example is given to illustrate the application of the policy.