A new collaborator selection method of cooperative co-evolutionary genetic algorithm and its application

Cooperative Co-evolutionary Genetic Algorithm (CCGA) is an effective way to solve complex problems like high-dimensional and multi-Objective problem, but there are also performance issue of high time complexity in the application of the algorithm. For the issue of collaborator selection is a key element of the success of applying the algorithm, the paper proposes a new method to select collaborators called Distance-based Collaborators Selection Algorithm (DBCCGA), which draws on the idea of classification in machine learning, two individuals are selected as reference individuals in each population, then evaluate individuals according to the distance of candidate individual and reference individuals by which evaluate operation is needed only once, a rule of getting individuals around the best individual is set to make the search more directional. The availability and validity of this algorithm are verified by experiments on the typical function optimization problem as well as on the Job Shop scheduling problem.