الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 90 |  |  |
| | | from | 90 | | |
Abstract
This thesis discusses combinatorial optimization methods and gives a big focus on one type of them. This type is called Ant Colony Optimization. Three versions of an Ant Colony Optimization algorithm are discussed: Ant System algorithm, Maximum Minimum Ant System algorithm and Ant Colony System algorithm.
The thesis presents three designed local search techniques for improving the performance of the Ant Colony Optimization algorithm. The first technique is called Inner Local Search. The second one is called Outer Local Search and the third one is called Simulated Annealing based Local Search. Each technique was used in the three versions.
We used the TraveJing Salesman Problem as a testbed. It is one of the most widely known combinatorial optimization problem. It is used in many applications such as computer network, printed circuits boards and drilling applications.
We present in this thesis some new results obtained for the Traveling Salesman Problem by using the improved versions of the Ant Colony Optimization algorithm. Some experiments were presented to analyze the performance of Ant Colony Optimization algorithm.