Integrated design and analysis of product quality and tooling reliability

System reliability is a vital factor to ensure quality and productivity in manufacturing processes. Most existing system reliability research is based on an analysis of component failures and their interdependency.; The downtime of a manufacturing process is caused by both tooling component failures and product quality out of design specification. In this thesis, the interdependency between product quality and tooling component reliability is represented by a newly defined concept: “QR-Co-Effect”. Without considering the QR-Co-Effect, a system reliability analysis could be biased. Traditionally, various techniques such as tolerance design and maintenance have been applied to improve either product quality or tooling reliability separately. The QR-Co-Effect model provides a link between tooling reliability information and product quality. Through this link, the effectiveness of those traditional techniques can be improved. Furthermore, in multi-station manufacturing processes, propagation of the QR-Co-Effect along stations (the “QR-Chain” effect) occurs.; In this thesis, a theoretical basis and a methodology are developed for modeling and analysis of integrated quality and reliability, with an emphasis on the QR-Co-Effect. The thesis is organized into 5 inter-related discussions: (1) QR-Co-Effect modeling for a fixture station through a Monte Carlo simulation procedure; (2) QR-Chain modeling: an analytic solution and an easy-to-evaluate upper bound is derived; (3) Optimal tooling degradation rate design based on the QR-Chain model: within the context of automotive body assembly processes, a nonlinear optimization problem is formulated with optimality studied; (4) Quality-Oriented-Maintenance: three feasible multi-component maintenance policies are adopted by considering the loss due to product quality; and (5) Unified decision-making framework for tolerance design and maintenance: two optimization problems are formulated and analyzed based on availability of quality loss function. In addition, several case studies are conducted to illustrate the application of the developed methodology.; Our results should have a strong impact on industrial practice. The lack of communication between and integration of quality and reliability improvement groups in most manufacturing firms is generally regarded as a major obstacle to lowering costs. In particular, our analysis of data collected from the automotive industry shows a significant QR-Co-Effect. As a result, a substantial economic impact from this research is foreseeable.