Sunday 24th of September 2017
 

Analysis of the impact of parameters values on the Genetic Algorithm for TSP


Avni Rexhepi, and

Genetic algorithms (GAs) are multi-dimensional and stochastic search methods, involving complex interactions among their parameters. Researchers have been trying to understand the mechanics of GA parameter interactions by using various techniques. It still remains an open question for practitioners as to what values of GA parameters (such as population size, choice of GA operators, operator probabilities, and others) to use in an arbitrary problem. Genetic algorithm (GA) parameters are explored to minimize the time needed to find a solution. The basic GA code stays the same throughout the entire system. Variable parameters include mutation rate, crossover rate, crossover operator, number of generations, population size, etc. When an optimization problem is encoded using genetic algorithms, one must address issues of population size, crossover and mutation operators and probabilities, stopping criteria, selection operator and pressure, and fitness function to be used in order to solve the problem. This paper tests a relationship between (1) size of initial population, (2) mutation probability, and (3) number of generations in runs of Genetic Algorithm for solving the TSP. TSP is an NP hard problem, so using Genetic Algorithm we can find a solution on reasonable amount of time. In this paper we describe the results of the solution for the Traveling Salesman Problem (TSP) for Kosovo municipalities, using the genetic algorithm, with different settings for the parameters of the Genetic Algorithm.

Keywords: Genetic Algorithms, TSP, Parameter Selection, Initial Population, Crossover, Mutation

Download Full-Text


ABOUT THE AUTHORS

Avni Rexhepi
Teaching and research Assistant in Faculty of Electrical and Computer Engineering, University of Prishtina, since 1996.


Teaching and research Assistant in Faculty of Electrical and Computer Engineering, University of Prishtina, since 1993.


PhD in computer sciences (1989) - University of Zagreb, Croatia. Ordinary Professor at the Faculty of Electrical and Computer Engineering, in the University of Prishtina.


IJCSI Published Papers Indexed By:

 

 

 

 
About IJCSI

IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us
FAQs

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482
Email: info@ijcsi.org

More contact details »