Optimal opportunistic maintenance policy using genetic algorithms, 1: formulation

Describes the development of two genetic algorithm (GA) programs for cost optimization of opportunity‐based maintenance policies. The combinatorial optimization problem is formulated and it is shown that genetic algorithms are particularly suited to this type of problem. The theoretical basis and operations of a standard genetic algorithm (SGA) are presented with an iterative procedure necessary for implementation of the SGA to least‐cost part replacement. However, an SGA used in an iterative manner may limit the global search capability of the evolutionary computing technique and may lead to suboptimal solutions. To avoid this problem, an improved GA which considers more than two groups simultaneously is devised. This model is based on the permutation representation and genetic sequencing operators originally developed for the travelling salesman problem. The same example used with the SGA confirmed that the improved GA can bring additional savings.