On Genetic Crossover Operators for Relative Order Preservation

Genetic Algorithms is a population-based optimization strategy that has been successfully applied to many real world and accademic optimization problems. The crossover operator plays a very important role in genetic algorithms since it is the element that generates the exchange of information between individuals during the search. This paper shows how the careful design of these crossover operators should be essential to avoid loss of information. This is shown graphically using the Travelling Salesman Problem as an example of one application of genetic algorithms where the relative order preservation of the structure of the configurations is extremly important when the diversity of the individuals is small. 1 Introduction With the name Genetic Algorithms (GA), we understand a population-based optimization strategy that can also be regarded as part of evolutionary techniques to deal with optimization problems. Their remarkable property is the fact that they are robust methods, which..