Correlation-Based Dynamic Allocation Scheme of Fitness Evaluations for Constrained Evolutionary Optimization

Constrained optimization is an active research topic in evolutionary computation. It challenges the evolutionary algorithms (EAs) in allocating fitness evaluations to the minimization of constraint violations and the optimization of objectives. Most existing EAs implement fixed allocation schemes by using nonpriority, priority, or priority-complete comparison criteria. This article argues that different constrained optimization problems (COPs) should be solved by algorithms with a dynamic allocation scheme (DAS), and the key to adjusting the allocation scheme is a judgment on whether the objective optimization or the constraint violation minimization is beneficial to finding the optimum. In this article, correlations between objectives and constraints are measured for judgment. Based on the correlations, the DAS for fitness evaluations is proposed for constrained evolutionary optimization. The objective priority criterion or the objective priority-complete criterion is dynamically selected to allocate more fitness evaluations to the objective optimization if the optimization of objectives is beneficial. Otherwise, the constraint priority criterion or the constraint priority-complete criterion is selected to allocate fitness evaluations. The experimental results show that algorithms with the selected criterion significantly outperform state-of-the-art algorithms in terms of attaining feasible solutions with better objective values. Furthermore, algorithms with the DAS have the advantage of consistently finding feasible solutions with better objective values than algorithms based on nonbeneficial, fixed, and randomly selected schemes. The results demonstrate the positive impact of the correlation-based DAS on EAs for solving COPs.