Design for Manufacturing and the Life Cycle
Successful new product development requires the ability to predict, early in the product development process, the life-cycle impacts of alternative product design concepts. Ignoring downstream issues (or producing poor estimates) leads to poor decisions and product designs that cause unforeseen problems. These products must be redesigned. Accurate predictions allow a product development team to create a superior design that performs satisfactorily in all ways. This reduces the number of redesign iterations, the time-to-market, and the development costs. Consequently, manufacturing companies (and solution providers) have developed many design decision support tools that form the class of Design for Manufacturing and Life Cycle methodologies.
Design for Manufacturing (DFM) and Design for Assembly (DFA) are two of the most common and popular techniques in this class of tools. Traditionally, DFM and DFA evaluate the materials, the required manufacturing processes, and the ease of assembly. That is, DFM and DFA study the feasibility and cost of manufacturing the product at the operation level. However, continued research has led to a broader portfolio of methods and techniques to address a wide range of lice cycle concerns including manufacturing system performance, life cycle cost, and environmental issues. We anticipate that integration of big data and cloud infrastructure will enable global design teams real-time analysis of their design decisions using Design for Manufacturing and Life Cycle methodologies. Our own research is focused on minimizing the total life cycle cost and environmental impact of high quality plastic products to thereby improve the quality of life for all people.
Product Architecture Designed for Global Supply Chains
Extensive Simplex Method
The Extensive Simplex Method is presented for use in general decision making problems including engineering design, manufacturing process and quality control, and other applications. The system relies on a function matrix that relates decision variables to performance variables. The system utilizes both global and local linearization of non-linear functions, after which the Extensive Simplex Method is used to derive the set of all feasible decisions based upon the specification limits for the performance variables and the control limits on the decision variables. Beyond current Six Sigma best practices, the described system explicitly considers both modeling uncertainty and uncontrolled variation. The specification limits may be automatically tightened by the confidence intervals and variation limits to ensure feasibility to a desired level of confidence and robustness, as consistent with other Design for Manufacturing and Life Cycle methodologies.
Three sets of feasible decisions are established including 1) the global feasible set that establishes the extreme limits of feasibility by allowing all the decision variables to vary simultaneously within their range of the control limits, 2) the local feasibility, which shows the immediate feasibility for each decision variable holding other decision variables at their current value, and 3) the controllable feasibility for each performance variable holding other performance variables at their current value. The system provides a perspective view of 1) the function matrix, 2) a historical view of the decision variables which may be used in a manner similar to statistical process control X-Bar charts, 3) a historical view of the performance variables which may be used in a manner similar to statistical quality control charts, 4) a set of decision windows showing the joint feasibility of all pairs of decision variables, which may be used in a manner similar to process windows, and 5) a set of performance windows showing the joint feasibility of all pairs of performance variables, which may be used in a manner similar to Pareto Optimal graphs.