Non-dominated Sorting Tournament Genetic Algorithm for Multi-Objective Travelling Salesman Problem

Abstract

A Travelling Salesman Problem (TSP) is an NPhard combinatorial problem that is very important for many real-world applications. In this paper, it is shown, that proposed approach solves multi-objective TSP (mTSP) more effectively than other investigated methods, i.e. Non-dominated Sorting Genetic Algorithm II (NSGA-II). The proposed methods use rank and crowding distance (well-known from NSGA-II), combining those mechanisms in a novel, unique way: competing and coevolving in the evolution process. The proposed modifications are investigated and verified by the benchmark mTSP instances, and results are compared to other methods.

The Best Efficient Solutions for Multi-Criteria Travelling Salesman Problem Using Local Search Methods

“…In order to solve these problems, the suggested algorithm has high convergence with minimum local value. Myszkowski et al [5] suggested a new method to solve multiobjective TSP (they notation it by mTSP) where this method is better than other methods which is called Sorting Genetic Algorithm II (NSGA-II). The new modification considers the improvment of the results which are verified by the benchmark of (mTSP).…”