|
Abstract: In the process of quality control, certain interrelations
exist between product faults and the inspection and selection of the
related parts based on statistics theory. The data mining model is
established in order to discover these interrelated rules, it includes
quality forecast algorithm and quality association-association rule
algorithm. The goal is to find rules inwhich believability is larger
than the presetting value and to forecast the situation of the product
quality in the future. The data mining tool is developed and the data
warehouse and data market for generating the product quality information
are constructed, and the discovering flows of predicting quality and
finding interrelated rules based on data mining and grey theory are also
designed. Finally, two examples of forecasting the situation of
selection and inspection of electronic parts and finding classification
of the fault modes are presented, and the algorithms’ availability and
efficiency are verified.
Key words: Computer aided quality control(CAQC) Data mining
Quality information database Grey theory Association rule
CLC No: TH166
国家自然科学基金(50375071)、江苏省自然科学基金(BK2003091)、江苏省教育厅科学基金(2004JD028J)和镇江市科学基金(2004JD020J)资助项目.
Received 20041020, received in revised form 20050415
|