GAT: A Genetic Algorithms Toolkit

Abstract

During the last thirty years there has been a rapidly growing interest in a field called Genetic Algorithms (GAs). The field is at a stage of tremendous growth as evidenced by the increasing number of conferences, workshops, and papers concerning it, as well as the emergence of a central journal for the field. With their great robustness, GAs have proven to be a promising technique for many optimization, design, control, and machine learning applications. A genetic algorithms toolkit (GAT) has been developed to help researchers facilitate GAs. With the readily available tool users can reduce the mechanical programming aspect of simula- tion and concentrate on principles alone. The toolkit was established to help users operate and control not only the structural identification but also the paramet- ric identification of GAs. It outlines how to design ge- netic algorithms and how to set parameters of different kinds of problems. The purpose of making this sys- tem available is to encourage the experimental use of genetic algorithms on realistic optimization problems, and thereby to identify the strengths and weaknesses of genetic algorithms. This paper describes the toolkit, shows how it can be used to solve various problems, and provides details on its implementation.