Lot sizing, preventive maintenance, and warranty decisions for imperfect production systems

Sets out to develop an integrated model that considers simultaneously inventory production decisions, PM schedule, and warranty policy for a deteriorating system that experiences shifts to an out of control state. The time to shift follows a general probability distribution with increasing hazard rate, so that time‐based PM is effective in improving the system reliability.

A profit function is used to model the production system. Optimization techniques are used to generate optimal solutions for the problem. Although global optimality cannot be guaranteed, empirical results show that global optimal solutions are obtained.

The integrated model provides decisions on inventory levels, production run length, and PM schedule simultaneously. It is illustrated through numerical examples that investment in PM can lead to savings in warranty claims for repairable products. As a result, the overall profit per unit, in certain cases, is higher with PM than without PM.

The production system is taken, numerical examples are presented and a sensitivity analysis is conducted to gain more insight into the developed model. In particular, the numerical analysis shows that a better PM program reduces warranty claims.

In addition to the joint optimization of production/inventory decisions and PM schedule, such models can be very useful in making resource allocation decisions between warranty and PM programs. It is clear from the numerical analysis that a better PM program reduces warranty claims.

The paper provides a joint optimization of production inventory decisions and the PM schedule for a system subject to a time to shift that follows a general probability distribution. Previous research considered only an exponential distribution and did not consider PM.