AntColony Optimization (ACO) is a meta heuristic that is mainly used for tackling combinatorial optimization (CO) problems. The traveling salesman problem(TSP) is one of the most important combinational problems.It is known to be NPhard.

ACO is taken as one of the high performance computing methods for TSP. In ACO algorithm, theheuristic information is very important in generating high qualitytours.Because the value of the pheromone trails do not get enough informationin the early stage of learning and cannot guide the ants in constructing goodtours, the heuristic parameter may be set to a large value. On the other hand,in the later stage, the heuristic parameter may need a small value because thepheromone trails may have collected enough information to behavior as requiredand the heuristic information may mislead the search to local optimal solution.

The heuristic parameter is set as a constant in traditional ACO algorithms. Itstill has some drawbacks such as stagnation behavior,uneven distribution ofants, long computational time, and premature convergence problem of the basicACO algorithm on TSP. Those problems will be more obvious when the consideredproblems size increases. The proposed system based on basic ACO algorithm withwell distribution strategy and information entropy which is conducted on theconfiguration strategy for updating the heuristic parameter in ACO to improvethe performance in solving TSP.

ACOis a relatively novel meta-heuristic technique and has been successfully usedin many applications especially problems in combinatorial optimization. ACOalgorithm models the behavior of real ant colonies in establishing the shortestpath between food sources and nests. Ants can communicate with one anotherthrough chemicals called pheromones in their immediate environment. The antsrelease pheromone on the ground while walking from their nest to food and thengo back to the nest. The ants move according to the amount of pheromones, thericher the pheromone trail on a path is, the more likely it would be followedby other ants. So a shorter path has a higher amount of pheromone inprobability, ants will tend to choose a shorter path.

Through this mechanism,ants will eventually find the shortest path. Artificial ants imitate thebehavior of real ants, but can solve much more complicated problem than realants can.ACOhas been widely applied to solving various combinatorial optimization problemssuch as Traveling Salesman Problem (TSP), Job-shop Scheduling Problem (JSP),Vehicle Routing Problem (VRP), Quadratic Assignment Problem (QAP), etc.Although ACO has a powerful capacity to find out solutions to combinationaloptimization problems, it has the problems of stagnation and prematureconvergence and the convergence speed of ACO is very slow. Those problems willbe more obvious when the problem size increases.

Therefore, several extensionsand improvements versions of the original ACO algorithm were introduced overthe years. Various adaptations: dynamic control of solution construction 4,mergence of local search 3, 13, a strategy is to partition artificial antsinto two groups: scout ants and common ants 11 and new pheromone updatingstrategies 1, 3, 14, using candidate lists strategies 2, 16, 17 are studiedto improve the quality of the final solution and lead to speedup of thealgorithm. All these studies have contributed to the improvement of the ACO tosome extents, but they have little obvious effect on increasing the convergencespeed and obtaining the global optimal solution. In the proposed system, themain modifications introduced by ACO are the following. First, to avoid searchstagnation and ACO is more effective if ants are initially placed on differentcities. Second, information entropy is introduced which is adjust thealgorithm’s parameters. Additionally, the best performing ACO algorithms forthe TSP improve the solutions generated by the ants using local searchalgorithms.

Algorithmsand Artificial Intelligence: An Algorithm is a set of rules that we follow toachieve some goals.Goals might be some predictions or a solution of anyproblem. Articial intelligence is deception of human intelligence process thatemphasizes creation of intelligent machines that works like humanbrain.

Phenomena of articial intelligence can be achieved by following somealgorithms.Our focus would be to explore the unidentied areas and study itsapplication on traveling salesman problem(TSP) that means we will predictsolution of traveling salesman problem by following ant colony algorithm(ACO).ACOis a heuristic based algorithm.This algorithm is used for finding shortest pathby an ant to locate its food.When one ant nds a source of food it returns backto it’s colony leaving behind some marks called pheromones. Next ant when comeacross marks it likely to follow same path to an extent.If second ant does thesame it leaves its own marks and it gets stronger until bunch of ants travelingto various food sources near the colony.