Optimal Models of Preventive Maintenance and Replacement Policies

In this paper, we solve a family of problems related to machine replacement and preventive maintenance. We consider the optimal way of skipping the available technology and be ready for the next technological breakthrough. Skipping the existing technology may have a crucial impact on the firm’s performance in a highly competitive world where the up-to-date technology affects both revenues and prestige. We investigate two cases of technological replacement problem for a discrete time finite horizon. The first deals with two available machines: the one-in-use and a more improved existing one that can be purchased to replace the first. The second case is a consideration of a third machine type that will be available in future. The maintenance level of each machine is chosen for each period in order to control performance deterioration. We show that the optimal maintenance policy for a single machine life span has a bang-bang structure, and employ this property in a dynamic programming setting to formulate and solve the problem efficiently. Numerical example demonstrates that management may improve profitability by delaying (but not necessarily foregoing) replacement with a more improved existing machine that is currently available.