Impact of the Number of Generations on the Fitness Value and the Time Requiredto Find the Optimal Solution in Standard GA Applications

摘要

Genetic Algorithms (GA) are search and optimization algorithms and as such are used for minimizing or maximizing a given function and if possible finding its most suitable solution. They can be used for finding a solution to problems that are difficult to solve with traditional optimization techniques, including problems that are not well defined or difficult to be mathematically modeled, such as the traveling salesman problem and the 2D packing problem. The data we obtain from the applications that implement the GA are visually presented as a graph from which we can see the progress of the GA's fitness value minimization over the generations, the smallest fitness value, the generation of its occurrence and the solution itself. Within this visualization we also made an analysis on the impact of the number of generations on the fitness value and the time required for finding the optimal solution.