Quality Improvement through SPC and EPC Information Sharing
Transcription
Quality Improvement through SPC and EPC Information Sharing
Quality Improvement through SPC and EPC Information Sharing Dr. Fugee Tsung Department of Industrial Engineering and Engineering Management Hong Kong University of Science and Technology Clear Water Bay, Kowloon, Hong Kong Email: [email protected] Introduction With recent advances in Information Technologies (IT), the research and practice of supply-chain information sharing has a direct impact on the production scheduling and inventory control. However, few investigations focused on the impact of supply-chain information sharing on product and process quality. Furthermore, it is still not clear how and what information should be shared/used, and how to quantify the benets of information sharing in terms of quality improvement. In this research, a \matching problem" is used to demonstrate the quality impact of information sharing. We quantify and compare the impacts of dierent information-sharing strategies on process and product quality, and indicate that real-time information sharing may lead to dramatic quality improvement for an assembly process, as an example of two-stage supply chain. SPC and EPC Information Sharing The real-time process data/information sharing becomes common due to recent advances in IT, which has great potential for quality improvement. Besides Statistical Process Control (SPC), Engineering Process Control (EPC) or Automatic Process Control (APC) is also a quality technique utilizing real-time process data. EPC is employed within process to compensate for drifts from target values by feedback adjustment. However, it is not always applicable as the within-run process adjustment may not be possible, such as in a stamping process. Especially, for the fuel injector assembly problem, the needle machine setting cannot be easily changed within run, but between runs or between days. Instead of abandoning the usage of real-time data in this case, we may gain by sharing the information with other processes, such as the body manufacturing. Here the Box-Jenkins Bounded Adjustment approach, one of the EPC methods, is modied to adjust the body machine setting for better matching quality by employing real-time needle process data (see Hunter, 1998, Box and Luceno, 1997, and references therein). As the information of needle dimention cannot be used for feedback adjustment, it is used for feed-forward adjustment of body machine setting by modifying the Box-Jenkins Bounded Adjustment approach. The proposed approach is used for forward adjustment instead of feedback, and an adjustment variable with correlated errors is considered, which is dierent from the original Box-Jenkins approach. The approach is rst to calculate the exponentially weighted moving average (EWMA) statistic from the real-time needle dimention data. Note that in industrial practice, a three-term adjustment scheme, i.e., Proportional-Integral-Derivative (PID) scheme, is also a commonly used method (Tsung and Shi, 1999, Tsung, Shi, and Wu, 1999). In many situations, only one or two of these three terms are used (Tsung, Wu, and Nair, 1998). In particular, if the Proportional and Derivative terms are set to zero, we have the Integral (I) scheme, which is equivalent to the EWMA. A more detailed study of the proposed SPC/EPC information sharing approach can be found in Tsung (1999). Conclusion In this research, we quantify and compare the impacts of dierent information-sharing strategies on process and product quality. We indicate that real-time information sharing may lead to dramatic quality improvement for an assembly process, an example being the two-stage supply chain. The proposed approach to evaluate the impact of information sharing on quality improvement may be directly extended to a more complex supply chain. However, what information to use/share and how to use/share it for quality improvement in a complex supply chain still warrant further research. REFERENCES Box, G. E. P. and Luceno, A. (1997) Statistical Control by Monitoring and Feedback Adjustment, New York: John Wiley. Hunter, J. S. (1998) The Box-Jenkins Bounded Manual Adjustment Chart. Quality Progress, 31, 129-137. Tsung, F. (1999) Impact of Information Sharing on Statistical Quality Control in a Supply Chain. working paper. Tsung, F. and Shi, J. (1999) Integrated Design of Run-to-Run PID Controller and SPC Monitoring for Process Disturbance Rejection. IIE Transactions, in press. Tsung, F., Shi, J. and Wu, C. F. J. (1999) Joint Monitoring of PID Controlled Processes. Journal of Quality Technology, in press. Tsung, F., Wu, H. and Nair, V. N. (1998) On the Eciency and Robustness of Discrete Proportional-Integral Control Schemes. Technometrics, 40, 214-222. FRENCH RESUME Recemment le developpement dans la technologie de l'information, la recherche et la pratique du partage de l'information dans la gestion de la cha^ne logistique ont eu un impact direct sur l'ordonnance de la production et la gestion des stocks. Par contre l'impact du partage de l'information dans la gestion de la cha^ne sur le produit et notamment le processus de la qualite n'ont pas ete sujets de nombreuses recherches. D'autant plus qu'il reste a developper quelles sont les informations et de quelle maniere les partager ainsi que la quantication des avantages du partage des informations en terme d'amelioration de la qualite. Notre recherche quantie et compare les eets des dierentes strategies de partage de l'information sur la qualite du produit, et indique que le partage de l'information en temps reel permet de realiser une amelioration notoire dans un processus d'assemblage, un exemple de cha^ne logistique bi-niveau.