Development and application of a methodology for minimizing manufacturing costs based on optimal tolerance allocation.

Abstract

With stiff global competition, manufacturers are seeking ways to improve the assembly quality at minimum costs. As the automotive assemblies are usually complex, comprising many components requiring specialized product and processing knowledge, determining an optimal way to improve the assembly quality becomes very difficult. It requires that the systems analysts have a good understanding of the functionality of the components and their relationship to the assembly, knowledge of manufacturing process capability and the associated cost for making the individual components to specified tolerances, and access to a good tolerance analysis and synthesis software tool. The unavailability of even one of these essential elements often leads to either across-the-board tolerance cuts or usage of over-simplified analysis tools. This research, through a detailed case study of the compression ratio of an engine, provides a methodology for integrating statistical tolerancing methods, actual process capability data, and investment cost versus tolerance relationships to minimize the manufacturing costs. First, the geometrical relationships, that tie the component tolerances to compression ratio, were determined. Then, the statistical tolerancing method, using Taylor series, was developed for compression ratio tolerance stack-up analysis. This approach was chosen for its ease in implementation of an optimization model. Next for Engine A, the actual process capability was studied for the major contributors to compression ratio variation. As a result, recommendations are made on ways to decrease the current costs of the existing engine, with the possibility of reducing both the within engine and between engine compression ratio variability. Finally, a method for developing realistic investment cost versus tolerance curve for a feature is presented; along with an algorithm to solve the mixed-variable, non-linear optimization model. The potential benefits to the engines analyzed were: a decrease in the current compression ratio tolerance by approximately one-half at no extra cost, an increase in the nominal compression ratio resulting in better fuel economy and more engine power, and a reduction in the operating costs for existing engines and investment costs for new engine programs.