A Novel Hybrid Solution for Traveling Salesman Problem

The Traveling Salesman Problem (TSP) is one of the most notorious challenges in the fields of optimization and operation research. The TSP asks an apparently easy question, given a list of locations and the distances between them, what is the shortest feasible path that visits each city precisely once before returning to the initial city. Despite its obvious idea the TSP is known for being challenging to solve since the number of alternative routes grows exponentially with the addition of more cities. This makes it an NP-hard problem in computational complexity theory. In this paper, a novel hybrid solution that incorporates the Genetic Algorithm (GA) with the Hopfield Neural Network (HNN) is proposed to solve the TSP. This solution exploits the evolutionary exploration capabilities of the GA and the pattern recognition abilities of the HNN. The proposed technique provides promising results in terms of optimal solution and time needed to achieve this solution. Results prove the effectiveness of the proposed algorithm in solving complex combinatorial optimization problems.