MONITORING DYNAMIC SYSTEMS: THEORETICAL DEVELOPMENT OF THE SELF-STARTING CUSUM FOR NON-ZERO SLOPE BEHAVIOR

摘要

Engineering Managers do not have an adequate tool for monitoring the progression of a living system. Most living systems undergo Non-zero slope behavior, which makes it difficult to discern special cause from common causes of variation. The features of the Self-Starting CUSUM chart provide a fertile base for the development of a tool that addresses this issue. This research paper proposes a modification to the estimators of the Self-Starting CUSUM chart to develop a tool for engineering managers who are interested in monitoring increasing and decreasing trends in systems. A historical account of the logical developments leading to this idea is presented. Then a subsequent explanation of the model is detailed. Finally, a discussion of possible impact areas and conclusions about the feasibility, and applicability of the methodology, concludes this work. The efforts of this paper intend to help engineering managers gain more predictive power over living systems.