Harmonizing Prognostics and Health Management with Risk Analysis for Smart Manufacturing Systems

This thesis develops a methodology for the design phase of a Prognostics Health Management (PHM) system of a manufacturing process. The methodology (i) builds upon Hierarchical Holographic Modeling (HHM), Risk Filtering, Ranking, and Management (RFRM), and Fault Tree Analysis (FTA); (ii) provides scope and direction for a PHM system by identifying a prioritized set of targets that would most benefit from PHM capabilities; and (iii) tests the outcome of the developed methodology with two case studies with U.S. manufacturing companies. Currently, there are multiple methods to determine the major failure modes of a system following an accident or catastrophe. However, the proposed methodology in this thesis allows for a thorough analysis to be conducted even before a failure occurs in a manufacturing environment. Another important goal of this thesis is to demonstrate the compatibility and synergy between two seemingly different methodologies/processes: Risk analysis and PHM for manufacturing. More specifically, this research demonstrates that the theory, methodology, and current practice of the system-based risk analysis are harmonious and compatible with PHM. The compatibility between Risk Analysis and PHM is further solidified through an investigation and comparison between the Dynamic Roadmap for Risk Modeling, Planning, Assessment, Management, and Communication with the Major Design Components of Prognostics and Health Management. The major benefits to PHM from integrating Risk Analysis methodology are revealed through the two case studies in U.S. manufacturing industry.