Determining Optimal Replacement Policy with an Availability Constraint via Genetic Algorithms

Abstract EN

We develop a model and a genetic algorithm for determining an optimal replacement policy for power equipment subject to Poisson shocks.

If the time interval of two consecutive shocks is less than a threshold value, the failed equipment can be repaired.

We assume that the operating time after repair is stochastically nonincreasing and the repair time is exponentially distributed with a geometric increasing mean.

Our objective is to minimize the expected average cost under an availability requirement.

Based on this average cost function, we propose the genetic algorithm to locate the optimal replacement policy N to minimize the average cost rate.

The results show that the GA is effective and efficient in finding the optimal solutions.

The availability of equipment has significance effect on the optimal replacement policy.

Many practical systems fit the model developed in this paper.