Department of Mathematics

General Motors

Proposed Project for the MSU Industrial Math Students

Problem 1: Substructuring the cost evolution of technology products: Laptop Case Study

The cost and cost maturity of technology in products often depends on the development and technical improvement of several individual subcomponents. For example, a laptop computer decomposes into many subcomponents such as memory chips, batteries, CPUs, software, etc.. Each of these subcomponents have evolved through very different technological pathways and have their own evolutionary cost curves that can be observed over recent history. This project is intended to illustrate how various cost evolution mechanisms for the subcomponents can be used to approximate the cost evolution trajectory for laptop computers products as a whole. Which technological mechanisms and production improvements lead to cost-down behavior in each subcomponent? And furthermore, which subcomponents contribute most to the cost evolution of the laptop product as a whole and which components or features are used to maintain or improve the competitiveness of a laptop brand? Can particular technical features of each type of subcomponent and their expected cost evolution curves be used to approximate cost evolution behavior of products similar, but not identical to laptop computers?


Problem 2: Descriptive ID's for vehicles and their subcomponents

The assignment of a VIN, Vehicle Identification Number, adheres to an international standard and uniquely identifies a vehicle. However, a VIN does not identify the major subcomponents and distinguishing characteristics of a vehicle. During and after manufacturing a vehicle, engineering, marketing and other corporate functions require unique vehicle identifications that distinguish the vehicle by its hardware, software and other special characteristics such as color. For this project develop an algorithm to assign each vehicle a unique ID from which one can discern its major hardware and software content. The algorithm should consider using the ID to group components by function and by combinatorial inclusions and exclusions. The references in the ID should be obvious to nonexperts. The ID should be flexible enough to evolve and be backwardly compatible over time.