Design-for-Six-Sigma synthesis of multi-stage assembly systems

Current trends in a manufacturing design can be characterized as time-based competitions which involve frequent product model changes and enormous investments in terms of time and cost. In recent years, these challenges have been addressed through various developments in Design-for-Six-Sigma (DFSS) approaches by building and testing prototypes to validate and improve a product design. However, DFSS-based approaches are limited in their scope in which they cannot ensure that critical product functions are within allowable specifications since the impacts of process parameters are not considered during building and testing of a new product prototype. This dissertation aims to address this challenge by proposing a set of generic design synthesis methodologies to optimize product dimensional quality in a multi-stage assembly system. The significance of this research can be summarized in two levels which are: (i) design task system synthesis; and (ii) individual design task development.;Design task system synthesis is to develop a framework to integrate multiple design tasks to optimize the design of an assembly system within the minimum computational time. The framework consists of two major elements which are: (i) Hybrid Design Structure Matrix to systematically describe the dependency between design tasks; and (ii) Task Flow Chain to represent a hierarchical model of tasks and to generate the optimum design task sequences. The individual design task developments contribute the uniqueness in creating a set of comprehensive process-oriented design tasks which are able to consider both product and process parameters in a multi-stage assembly process. The design tasks presented in this research involve: (i) automatic generation of a variation prediction model; (ii) process-oriented tolerance optimization based on the surrogate yield model; (iii) multi-fixture layout optimization based on integration of Genetic Algorithm and Hammersley Sequence Sampling; and (iv) part-to-part joint optimization in the fixture joint constraint system. The proposed design tasks aim to enhance the robustness of the assembly system subjected to the design constraints such as cost and workpiece locating stability. The applications of the proposed research are demonstrated and validated through various industrial case studies.