| As the electrical manufacture technology developing continually, the launch time for electron product is becoming shorter and shorter. In order to keep competitive in fierce market competition, and to get high reward as well as meet clients' request for higher product grade and higher consumption satisfaction, electronic product manufacturers exert their efforts to reduce production cost by improving productivity. In the process of printed circuit board (PCB) serial production, insertion assembly is the bottleneck of the whole line, which determines the whole line's efficiency. So how to optimize insertion operation has become a problem to which corporations and researchers pay much attention. The former insertion process is done through workers experience. Though some methods have been provided by equipment suppliers in the recent years, these methods are simple and not efficient.Considering that the ant colony algorithm (ACA) has the functions of distributed parallel computing, positive-feedback and self-organization, this paper puts its optimization idea into the insertion operations and brings forward a new optimization model and algorithm. This paper's innovation points lies in: 1) the insertion assembly problem was described as the traveling salesman problem (TSP) based on machines' hardware structure and mechanical characteristic on the premise that the distribution of charging tank and the insertion order of the elements are the same, and then a mathematic model with the target of minimizing total production time. 2) considering ACA's tendency of getting into local optimum and long time for computing , heuristic algorithm—2-opt was brought in this paper and improved ACA was developed to seek the best solution for insertion operation order.This paper firstly introduced the PCB insertion operation skills including its fixing styles and technological process; then it described the basic principles and parameter settings of ACA, and mathematical model and structure flowchart were also built; furthermore, the paper analyzed the PCB assembly system in detail and implemented the scheme by programming with improved ACA; finally, with Matlab language, a series of simulation experimentations were carried out based on production data. The results indicated that the solution could enhance the production efficiency. |
