Evaluation of sequential, multi-objective, and parallel interactive genetic algorithms for multi-objective optimization problems†

Evolutionary Computation (EC) is the field of computational systems that use ideas and get inspiration from natural evolution [1]. Genetic Algorithms (GA) fall into the category of EC. GA are a type of search and optimization algorithm based on the mechanisms of genetics and natural selection. The canonical form of GA encodes each candidate solution to a given problem as a binary, integer, or realvalued string, referred to as the chromosome. GA simulate the genetic evolution of a population of individuals using recombination operators such as crossover and mutation. Crossover exchanges genetic material between two parents during mating while mutation flips a bit in the chromosome, typically of the offspring. Mutation is carried out to prevent premature convergence of the design variables and promote diversity, which means preventing all the bit structures of strings in the mating pool from becoming identical in an early stage of the evolution. Each individual is evaluated once per generation according to some fitness criterion enabling a numerical fitness value to be assigned to the individual. New individuals are created by the recombination operators for the next generation. It is important to realize that GA are stochastic, meaning that there is randomness involved; mainly in the initial generation of a random population, random choice of parents, random choice of which genes to inherit from Evaluation of sequential, multi-objective, and parallel interactive genetic algorithms for multi-objective optimization problems† Alexandra Melike Brintrup,1 Hideyuki Takagi,2 Ashutosh Tiwari1 and Jeremy J. Ramsden1,* School of Applied Sciences, Cranfield University, Bedfordshire, MK43 0AL, UK Faculty of Design, Kyushu University, 4-9-1 Shiobaru, Minami-ku, Fukuoka 815-8540, Japan