Research On Job Shop Scheduling Method Based On The Improved Smoothing Adaptive Ant Colony Algorithm

With the development of science and technology, the production scale of themanufacturing industry is growing, production environment is more complex andchangeable. Job shop scheduling problem by reasonable dispatching decision, can improvethe machine utilization rate, reduce the production cost, minimize total completion time,thereby improving the production efficiency and economic benefit. So reasonablescheduling decision is manufacturing industry wide attention and become the core part ofthe production process.All kinds of intelligent optimization algorithms are proposed and used in solving theworkshop scheduling problem continuously. They make the workshop scheduling problemimproved greatly. Among them, according to the mechanism of ants searching for foodsource in nature, the ant colony algorithm (ACO) finds the optimal solution throughconstant iteration of artificial ants. This algorithm through the positive feedback principleas well as the ants pheromone updating and transfer with each other, which make themcoordinate with each other in the search process to find the optimal solution. So the antcolony algorithm (ACO) is especially suitable for solving JSSP problem.An improved smoothing auto-adaptive ant colony algorithm is proposed, which isbased on the Job shop scheduling characteristics and ant colony algorithm self-weakness oflocal optimum and slow convergence speed. In the process of pheromone updating therewards and punishment measures are used to speed up the convergence speed of thealgorithm. When algorithm falls into local optimum, the smoothing mechanism is to adjustthe volatile coefficient of pheromone and expand the ant search scope, which makes thealgorithm "jump out of the trap"; improves the global search ability of the algorithm. Andthe self adapt adjustment is used according to the number of each path pheromone, whichhelp to improve the algorithm convergence speed. In the process of adaptive parameterupdates, in accordance with the algorithm performance parameter values of the appropriate adjustments. According to the experience of other researcher, when the algorithm run to aquarter of the total number of iterations change the parameters. The simulation resultsverify the feasibility and effectiveness of the proposed algorithm.The optimization results and efficiency of ant colony algorithm is affected byα,β,ρ,Q,m parameters largely. The value of the parameters play a decisive role on theperformance of the ant colony algorithm. In this paper, the main parameters of theimproved smoothing adaptive ant colony algorithm are analyzed and simulated. It showsthat the influence and rule of the parameters for the algorithm.With the problems of expanding of the scale of production, and single workshopproduction model can’t meet the demand. The improved smoothing adaptive ant colonyalgorithm is applied to distributed more workshop scheduling in this paper. Compared withother algorithms, the results show that the proposed algorithm are improved on theoperation time and search the optimal solution aspects.