Power Heuristics for Handling Constraint in Evolutionary Algorithms

The focus of our study is to propose a new constraint handling technique for real value encoding problem, specifically on the animal diet problem. This new proposed technique is a modification of mutation operator known as Power Mutation. This technique named as Power Heuristics is able to choose a better combination of feed ingredients based on certain performance. As the intention is to cater for strictness of nutrient and ingredients restriction, the Power Heuristics is introduced at the initial stage of the main methodology. The result shows that this technique is useful to remove inappropriate ingredients when an important constraint is violated.