Session:

GSW - Graduate Student Workshop

Title:

A Study of Mechanisms to Handle Constraints in Evolutionary Algorithms

Authors:

Efren M. Montes
Carlos A. Coello

Abstract:

In this work, we study different mechanisms to incorporate constraints into an evolutionary algorithm used for global optimization. The aim of the work is twofold. First, we propose a competitive constraint-handling approach which does not require a penalty function (nor penalty factors), and which is able to produce very competitive results while performing less fitness function evaluations than other algorithms representative of the state-of-the-art in the area. Second, we measure the rate at which our approach reaches either the feasible region of the search space or even the global optimum solution. Finally, we propose additional test functions and perform an empirical study that aims to find some of the features that make a constrained optimization problem difficult to solve by an evolutionary algorithm.