Local Fitness Landscape Exploration Based Genetic Algorithms

Genetic algorithms (GAs) have been used to evolve optimal/sub-optimal solutions of many problems. When using GAs for evolving solutions, often fitness evaluation is the most computationally expensive, and this discourages researchers from applying challenging This paper presents an approach generating offspring based on a local landscape exploration increase speed search better solutions. The proposed algorithm, “Fitness Landscape Exploration Algorithm” (FLEX-GA) can be applied single multi-objective optimization Experiments were conducted several benchmark problems with without constraints. performance FLEX-based algorithm single-objective compared canonical GA other algorithms. For problems, comparison made NSGA-II, Lastly, Pareto are evolved eight real-world comparative presented NSGA-II. Experimental results show that FLEX improves up 50% quality also improves. These provide sufficient evidence applicability approximation-based solving