Application of Taguchi Methods to the Design of an Automated Equipment for Serial Numbers Programming

• Main Authors: Tsai, Chia-Wei, 蔡稼衛
• Other Authors: Jeng, Mu-Der
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/83633511452903956642

碩士 === 國立臺灣海洋大學 === 電機工程學系 === 104 === The purpose of this thesis is to focus on the current IC problem, currently, every single IC contains different serial numbers. As a result of that, there is no suitable mass production equipment that can be used at this stage, and the programs must be completed manually. Due to the IC size became smaller and smaller, manually operating the IC generates a certain amount of cost and quality loss and resulting in low production and quality. The Taguchi methods with the concept of starting from design concept, based on the existing equipment cost and compatibility considerations and the use of XY-Table gantry suspension system re-architecture design to achieve the purpose of “automation programming objective record SN”. To make sure the stability of the programming process SN is not affected by the change of external environmental factors, and to lower production loss can effectively reduce costs while improving process yield. First of all, the thesis discusses how equipment structure and process parameters are selected by the applications of “System Design” and “Parameters Design” on Taguchi robust design stage. The main design of algorithms are the process of “programming SN” and the troubleshooting process of “programming up SN”, and to compile abnormal distribution of production yield by the experimental results and to find the factors of affecting the quality loss/variation. To analyze the causes for abnormalities by the fish-bone diagram, then used them as factors to define the standard reference. To consider the production yield as a quality characteristic of Taguchi experiment and to analyze the data by designing three level orthogonal experiments. Then to obtain a robustness combination of process optimized parameters by S/N ratio, factor response, classification induction factor and equipment with the ideal quality characteristics of “expecting better” verified by experiment. In conclusion, to simplify into one production process by applying two different equipment processes of two programming stations (same content) and serial number (different content) in accordance with rule no. N+1 discharge. Taguchi methods are efficient and streamlined the number of experiments to improve the process, and we analyzed and summarized a set of optimal parameter combination which is able to effectively enhance the effectiveness of the device as the basis for future volume production improvements. For follow-up researching, the data control of programming serials, the stability of power system, the cohesion of uninterruptible power supply (UPS) and abnormal exclusion of mechanical could be discussed further afterwards. Keywords : Taguchi Methods, Robustness Design, Automated Equipment, IC, Programming, Serial Number