Solution diversity in multi-objective optimization: A study in virtual reality

Solution diversity in evolutionary multi-objective optimization is considered. Although the Pareto front is ubiquitously used for the multi-objective optimization, the method of formation of the Pareto front in the evolutionary process is important to ensure the diversity of the solutions so that they are desirably evenly distributed along the front. Conventionally this is an issue and in some cases this is compromised with sub-optimality or layered Pareto fronts. This issue is dealt with in this research and a novel method termed as relaxed dominance for design applications is presented. The method is implemented for a design process as a case study and the effectiveness of the method is demonstrated.