The objective of this research project is to develop new regression-based methods for quality improvement in data-rich complex systems by considering various types of data uncertainties. This is motivated by the fact that many existing statistical and data mining methods for process monitoring, abnormality detection, and fault diagnosis have limited capabilities in handling data uncertainties, making them less effective when applied to real-world applications. On the other hand, research on data uncertainties has mainly been conducted from a pure statistics point of view, without linking the modeling and analysis of data uncertainties to quality improvement objectives. To overcome these limitations, four inter-related research tasks will be performed in this project, including (1) process modeling, monitoring, and fault detection based on data with uncertainties; (2) separation of process fault and sensor fault; (3) optimal sensor allocation for fulfilling prescribed requirements on monitoring and diagnosis capabilities; (4) identification of the maximum allowable level of data uncertainty under prescribed requirements on monitoring and diagnosis capabilities. This research will be demonstrated in a multistage complex manufacturing process as well as in Alzheimer?s disease study.
If successful, this research will (1) bridge the gap between data uncertainty studies in theoretical statistics and traditional quality engineering research that generally overlooked data uncertainties in developing quality improvement methodologies; (2) advance the state-of-art of the research in quality improvement by providing new enabling methodologies that facilitate process monitoring, detection, diagnosis, and control based on data with various types of uncertainties; (3) guide the design, selection, and allocation of sensors, as well as the data collection process, to guarantee that data uncertainties will not interfere with achieving quality improvement objectives. The research agenda will be integrated with an education plan, including new curriculum development, student internship opportunities, women and minority student recruitment, and K-12 programs.
