An improved genetic algorithm with a local optimization strategy and an extra mutation level for solving traveling salesman problem

TL;DR

This paper introduces an enhanced genetic algorithm with local optimization and an extra mutation step to more effectively solve the NP-complete Traveling Salesman Problem, outperforming conventional GAs in path quality.

Contribution

The paper presents a hybrid genetic algorithm combining local optimization strategies and an additional mutation level, improving solution quality for TSP.

Findings

Proposed method finds better paths than conventional GA.

Enhanced algorithm maintains acceptable computation time.

Local optimization strategies improve solution quality.

Abstract

The Traveling salesman problem (TSP) is proved to be NP-complete in most cases. The genetic algorithm (GA) is one of the most useful algorithms for solving this problem. In this paper a conventional GA is compared with an improved hybrid GA in solving TSP. The improved or hybrid GA consist of conventional GA and two local optimization strategies. The first strategy is extracting all sequential groups including four cities of samples and changing the two central cities with each other. The second local optimization strategy is similar to an extra mutation process. In this step with a low probability a sample is selected. In this sample two random cities are defined and the path between these cities is reversed. The computation results show that the proposed method also finds better paths than the conventional GA within an acceptable computation time.