A unified framework for statistical process control of a dynamic-stochastic system in manufacturing transitions.

摘要

Driven by global competition and evolving customer needs and expectations, manufacturing systems today have witnessed a significant increase in process transitions—an attempt to shift the process from one operating level to another. The transition behavior of a manufacturing process may be represented as a dynamic system with input variables, output variables, and a noise disturbance. An important outcome of the process transition is the induced transition period, a temporal trend that is inherent to the process transition. Because of poor understanding and control of the process transitions, large product and dollar losses often result. While much research effort has been dedicated to the advancement of monitoring and adjustment methodologies at process steady-states, little attention has been given to the process transitions.; The goal of this research is to develop the process monitoring and adjustment methodologies for addressing the process transition problem so that system performance improvement may be attained. The methodologies will provide a scientific means to acquire critical knowledge in process transitions as well as improved control and quality, leading to the enhancement of economic position. The three major developments in this research are: (1) The characterization of the process transition with the appropriate monitoring procedure. (2) The development of an adaptive monitoring procedure for the transition period, a characteristic part of the process transition. (3) The integration of the transition monitoring and adjustment methodologies and the robustness study of the adaptive monitoring procedure when an adjustment is applied to the process transition.; A plastic extrusion process is an integral part of this research as it provides a real environment for the development of the transition monitoring and adjustment procedures. Mathematical modeling of the process transition phenomenon is also an important element for performance evaluation and comparison of the methodologies. This research lays a solid foundation for future research on monitoring and adjustment methodologies for process transitions.