徐雅甄 博士候選人
(國立交通大學工業工程系)

Statistical Theory and Methods Applied to
Product Quality Assurance
(統計理論與方法在品質保證上的應用) 

96年04月11日(星期三)下午4:10

科技大樓數學系ST527

    Most commonly used control methods for monitoring quality characteristics of
the processes were developed under the assumption that the observations are 
independently and identically distributed (i.i.d.) with a normal distribution. 
The independence assumption may be unrealistic for some processes such as the 
milling processes and drilling processes that inherently produce correlated data.
The ordinary measures of process capability are inaccurate because the process
data is correlated and contaminated by the assignable cause variation. In order
to determine the accurate process capability, conventional capability calculation
must be modified. In this presentation, some statistical properties of the 
estimator are examined, and a technique for managing a process exhibiting a 
systematic assignable cause is discussed. Closed form of the exact sampling 
distribution of each process capability is derived. The practitioners can use the
proposed method to determine whether their process meets the quality requirement,
and make reliable decisions regarding the tool replacement time. For illustration
purpose, application examples involving tool wear are presented. 

聯絡人：東海大學數學系  潘青岳老師
電  話：04-23590121轉 3251 或 3668  
E-Mail：math@.thu.edu.tw