A methodology for determining engineering costs and their effects on the development of product families

Abstract

The goal of most firms is to deliver products that satisfy customer needs. Delivering a variety of differentiated products allows firms to satisfy the broadest range of customers. There is, however, a fundamental tension between this product differentiation and product cost. The use of product platforms allows a firm to reduce this tension, offering variety while also benefiting from the economics of mass production for shared components. The selection of components and subassemblies for platforming can have wide ranging effects on both product performance and cost. This thesis addresses the latter by presenting a methodology to assess product development costs, the amount of part sharing in a product family, and the effects of platforming on development, fabrication, and assembly costs for product families. Ordinal metrics are presented to assess the performance of product families. The methodology of process-based cost modeling, used to estimate product fabrication and assembly costs, is also posed. A method for determining the allocation of costs for parts and subassemblies shared among product family variants is outlined. A process-based cost model of the automotive product development process is presented. This model uses product part and subassembly characteristics to estimate the engineering effort required at various stages of the development process. Product development cycle time is also estimated. Linear regression analysis is used to determine which part and subassembly characteristics affect engineering effort. Additional development costs, such as computer hardware and software, overhead, and physical prototypes are also taken into account. Two automotive body architectures are analyzed to determine the cost savings from

(cont.) platforming: a tubular architecture (low tooling cost; high variable costs) and a traditional unibody architecture. Of the two designs, the tubular architecture is found to have less cost savings from platforming, even though the tubular architecture shares more parts. The higher fraction of variable costs and lower tooling cost, reduce the opportunity for sharing in the tubular architecture. Two instrument panel (IP) beam designs are also compared to study the effects of parts consolidation. A magnesium die cast IP beam is compared to a functionally equivalent steel IP beam that contains over six times as many parts. Development costs are found to make parts consolidation an attractive option, while this consolidation is found to reduce the opportunity for sharing and thus cause the magnesium IP beam family to be a more costly option than the steel design. Across all case studies, the fabrication and assembly of products are found to account for the majority of product cost; the majority of cost savings from platforming however, come from reduced development and assembly costs. Additionally, production volume and product lifetime are found to have large effects on which product architecture performs better. A significant perspective result, which emerged, was that the part sharing metric weighted by fabrication investment is found to be the most effective predictor of cost savings of the metrics tested.