ENHANCED IMPROVEMENT OF INDIVIDUALS IN GENETIC ALGORITHMS

Abstract

In this paper, a new modification of the genetic algorithms (GAs) based on an enhanced improvement of individuals is discussed. The basic philosophy of the proposed approach is to accelerate the convergence speed of the genetic search by maintaining compact populations of the outstanding quality individuals – “super-individuals”. The super-individuals are obtained by means of powerful iterated local search techniques. The increase in time for the improvement of individuals is compensated by decreasing the size of populations. We tested our approach on a well-known combinatorial optimization problem, the quadratic assignment problem (QAP). The results of the experiments show that using the enhanced improvement in GAs makes it possible to achieve very encouraging performance.