Optimize Transition Stages of the Integrated SPC/EPC Process Using Neural Network and Improved Ant Colony Algorithm

Abstract

Product quality plays an important role in facing competition and gaining competitiveness. Both Engineering Process Controllers (EPC) and Statistical Process Control (SPC) are effective methods of monitoring and adjusting the transition stages to improve process quality. At the same time, neural network was adopted to monitor the process and a flexible model is developed to determine optimal adjustable point for the integrated SPC/EPC. We adopt the improved ant colony algorithm to deal with the above model under the advanced machine choose rule: After all ants crawled, this algorithm could adjust pheromone aiming at whether it got into part convergence, this could help algorithm to get best solution faster. In the end, simulation experiments are done to verify the advantages. Results show that this algorithm can not only reduce the volatility of the process output and enhance system performance; and the integrated control method is more potential cost advantages.