Abstract:

The traveler is aiming to find the least costly tour in the traveling salesman problem, which is only one time out of each of the known cities. Although it is easy to identify the traveling salesman problem, obtaining the optimal solution is very difficult and NP-hard problem. The basic difficulty of this problem is that the number of possible tours increases in large numbers as the number of cities increases, which makes the problem impossible to solve with definite methods, so different methods have been proposed to solve the problem. One of these methods is genetic algorithms. Genetic algorithms are particularly suited to solve difficult optimization problems where traditional optimization methods are less effective. It is determine how to create and the size of the initial population that significantly affect the performance of the Genetic Algorithm solution. The initial population is often randomly selected but is used in different heuristics to improve the performance of genetic algorithms. In the study, the initial population was created with the nearest neighbour intentionally and randomly, and the different population sizes were considered and compared

Keywords: