Application of Genetic Algorithm and DMADV Technique to Improve the Process and Out-of-Plant Distribution in the Factory - Taking the Glass Factory as an Example

• Main Authors: Zhao-Fu Wen, 溫兆福
• Other Authors: Dr. Shui-Chuan Chen
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/juaptb

碩士 === 國立勤益科技大學 === 工業工程與管理系 === 105 === With the rapid development of science and technology and the rapid changes in market demand, the process is increasingly complex, how to continuously improve the quality, control the timing, taking into account the cost, risk, performance and many other factors, through DMADV approach to improve the process is a good way The This study will use the six standard deviation design to improve the process of the glass factory, import QR CODE to replace the one-dimensional bar code, increase the amount of data storage, improve order errors, work orders and manual specifications caused by copying, misjudgment of bad products, improve the process Of the yield to reduce costs, and the establishment of engineering monitoring system in the production and distribution process to immediately grasp the status of production and construction process, to immediately deal with unusual circumstances. The factory in the finished product inventory because the use of mobile shelves, so the plant is too large and no fixed storage of finished products, so the picking stage is easy to spend too much time to find the finished product position, so this study will use RFID on the mobile shelves , In the picking through the RFID positioning can quickly find out into. In addition, the transportation of finished products is also a part of the glass factory. This study will simulate the transportation problem. In this paper, we use integer linear programming to establish a mathematical model in the transportation environment of multi-vehicle multi-vehicle, and use genetic algorithm to solve the problem. In the case of small scale, the optimal solution can be obtained by using the integer programming, and when the problem is large, the genetic algorithm is used to obtain the approximate solution in a short time.