Date of Award
5-2013
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Legacy Department
Industrial Engineering
Committee Chair/Advisor
Cho, Byung Rae
Committee Member
Shappell , Scott A.
Committee Member
Greenstein , Joel S.
Committee Member
Melloy , Brian J.
Abstract
Many industrial firms seek the systematic reduction of variability as a primary means for reducing production cost and material waste without sacrificing product quality or process efficiency. Despite notable advancements in quality-based estimation and optimization approaches aimed at achieving this goal, various gaps remain between current methodologies and observed in modern industrial environments. In many cases, models rely on assumptions that either limit their usefulness or diminish the reliability of the estimated results. This includes instances where models are generalized to a specific set of assumed process conditions, which constrains their applicability against a wider array of industrial problems. However, such generalizations often do not hold in practice. If the realities are ignored, the derived estimates can be misleading and, once applied to optimization schemes, can result in suboptimal solutions and dubious recommendations to decision makers. The goal of this research is to develop improved quality models that more fully explore innate process conditions, rely less on theoretical assumptions, and have extensions to an array of more realistic industrial environments. Several key areas are addressed in which further research can reinforce foundations, extend existing knowledge and applications, and narrow the gap between academia and industry. These include the integration of a more comprehensive approach to data analysis, the development of conditions-based approaches to tier-one and tier-two estimation, achieving cost robustness in the face of dynamic process variability, the development of new strategies for eliminating variability at the source, and the integration of trade-off analyses that balance the need for enhanced precision against associated costs. Pursuant to a detailed literature review, various quality models are proposed, and numerical examples are used to validate their use.
Recommended Citation
Boylan, Gregory, "ROBUST PARAMETER DESIGN IN COMPLEX ENGINEERING SYSTEMS:" (2013). All Dissertations. 1081.
https://tigerprints.clemson.edu/all_dissertations/1081