Different deviation value prediction method to control the effect of the different data distribution patterns

• Main Authors: Sih-Ying Chen, 陳思穎
• Other Authors: Hong-Tau Lee
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/82235857259954364018

碩士 === 國立勤益科技大學 === 工業工程與管理系 === 100 === Engineering process control (EPC) and statistical process control (SPC) are two techniques that are commonly used in on line process control. The SPC technique using the control chart for monitoring the process based on the assumption that the data for each sample is satisfied a normal distribution and independent to each other. This may not true especially in a continuous manufacturing process. For handling this situation the EPC technique was developed. It forecasts the possible manufacturing disturbance for the next period based on the correlation property of the past data. An adjustment of the controllable (independent) variable is conducted for avoiding the happening of the forecasted possible disturbance of the respondent (dependent) variable. From this view point, the quality of the disturbance forecasting is essential for the effectiveness of EPC. In this research, a moving average approach and grey theory are applied to improve the precision level of the forecasting. Different distribution types of data are involved to illustrate the suitability of the EPC technique.