Abstract
Multiobjective evolutionary algorithms (MOEAs) are useful tools capable of searching problems that contain several conflicting criteria. Although MOEAs have been shown to be capable of finding a wide spread of Pareto-optimal solutions for a given problem, they are still hindered by the requirement for significant computation. This paper investigates a new MOEA that incorporates spatial structure into the population. The introduction of space into the algorithm alters the behaviour of the algorithm so that computational complexity increases linearly with population size. In addition, the paper suggests paths that could be taken to improve the algorithm’s ability to successfully converge upon the global Pareto-optimal front of a given problem.