Application Research Of Improved Ant-Genetic Algorithm In Shop Scheduling Model Base

As the fact development of the market economy, the scale of the manufacturing business is getting larger and larger. In the real manufacturing environment, different types of resource constraints, manufacturing capacity constraints and processing constraints coexist. Therefore there is a great need to arrange the manufacturing jobs reasonably. What is more, a good scheduling requires a good algorithm. In addition, for the time being, most of the researches are concentrating on either the processing or assembling scheduling. The fruit of these researches do not apply effectively enough in the business which deals with mixed manufacturing. Take a deeper look into this problem, it is because the process of production is very complicated and it is very rare that it only involves producing or assembling in the manufacturing process. Apart from this, it is also important to take real time information of warehouse, the information of those products being manufactured currently and the capacity of the workshop into consideration in order to make the scheduling dynamicly. its complexity is far beyond that of the current used international standard job-shop scheduling and some single-mode scheduling system. To solve to the above problems, it not only needs a good optimize algorithm, but also needs a good dynamic solution of shop attemper for the real shop.Combining advantages of Genetic Algorithm with Ant Algorithm, the paper proposed a new Ant Algorithm-Genetic Algorithm (DAAGA). The algorithm overcomes the shortcoming that is the slow convergence speed of traditional hybrid algorithm, and adopts a best hybrid point strategy to transfer two algorithms. At the same time, it imported an iterative adjusting threshold to control the genetic operation and the ant number at the evening of the algorithm, namely, when the population has already evolved to the neighborhood of the best value, the algorithm did mutation operation so as to reduce the genetic computational work and added the ant number to obtain the best solution earlier. The simulation results of Muth and Thompson problem indicated the effectiveness of the new algorithm.In the meantime, according to the practical situation of the mixed manufacturing business, this article designs a new coding method according to the mixed manufacturing situation which can apply to practical job shop problem. What is more, it also proposes an algorithm model with multiple objectives and cost-oriented. If compared with the conventional multi-objective optimizing algorithm which only concentrated on the manufacturing constraints, the new algorithm is much more feasible.This article developed a shop scheduling plat of the shop scheduling model database which is targeting at some factory's actual problem. Furthermore, this algorithm has been embedded into the multi-objective optimizing model. If we use this new method to solve the practical problem, the result shows the solution is both feasible and effective.