To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Traveling Salesman Problem using Neural Network Techniques |
|
Author |
Sabry M. Abdel-Moetty, Mohamed Ali Gomaa, Gaber A. Elsharawy, Asmaa Heakil |
| Citation |
Vol. 9 No. 2 pp. 341-347 |
|
Abstract |
The traveling salesman problem (TSP) is well known as a non-deterministic polynomial (NP) optimization problem. is that there is a list of cities that are to be visited by a salesman. A salesman starts from a city and come back to the same city after visiting all the cities. The objective is to find the path, which follows following constrains: 1. Salesman has to visit each city. He should not leave any city unvisited. 2. Each city should be visited only one time. 3. The distance that he travels till he returns back to the City he has started should be minimum. Two methods for solving the TSP Has been discussed . The first method is ant system (Ant colony system (ACS)) and the second one is the Neural Network (Hopfield Neural Network). In ACS, a set of cooperating agents called ants cooperate to find good solutions to TSPs. Ants cooperate using indirect form of communication mediated by pheromone they deposit on the edges of TSP graph while building the solutions. A proposed algorithm and standard Algorithm that are applied to the TSP are derived. Experiments are accomplished using ACS to understand ants behaviors . Also we present a proposed algorithm using continuous Hopfield neural network to solve TSP. To enable the N neurons in the TSP network to compute a solution to the problem, the network must be described by an energy function in which the lowest energy state (the most stable state of the network) corresponds to the best path. Results of programs (standard, proposed) are presented and compared. |
|
Keywords |
Traveling Salesman Problem (TSP), Ant Colony System (ACS), Artificial Neural Network (ANN) |
|
URL |
