| Decision technology for manufacturing management is a well-researched field with a vast literature. Virtually all of the results in this entire field rest explicitly or implicitly on a traditional business paradigm: The primary criterion for manufacturing performance is the control of costs, sometimes supplemented by another criterion concerning the achievement of on-time delivery.; The rise of fast-moving, technology-based industries such as semiconductors presents a challenge to this traditional paradigm. Prices decline rapidly due to rapid technological progress and obsolescence, and the rates of decline vary across different products. There is an intrinsic value of speed in manufacturing deployment and execution that is not captured under the traditional paradigm.; Using semiconductor manufacturing as a concrete subject, this thesis re-visits the spectrum of decision technologies for manufacturing management and revises these technologies to incorporate analyses of the economic value of speed.; The first step is to develop mathematics that quantifies the impacts on sales revenues from differences in manufacturing speed. Classical economic effects of supply vs. demand are shown to play a much more minor role than technological obsolescence in determining price trends of semiconductors. Empirical analysis is developed for selected product types to quantify the rate of price decline that is independent of classical supply-demand economics.; Next, we consider investment and engineering efforts related to manufacturing such as ramping up new plants and improving manufacturing yields. Models are developed to quantify both cost and revenue cash flows given price trends and performance metrics such as timing of capacity additions, timing of introductions of new process technology, yield ramp curves, manufacturing cycle times, etc.; The thesis then turns to the planning and scheduling cycle. Consideration of declining revenue is incorporated into production planning, release scheduling and dispatching. Generally, the impact of including price trend consideration lessens as the time frame and scope of the scheduling decision is reduced.; Finally, the thesis considers benchmark analysis of manufacturing and how to incorporate speed into efficiency analysis. New objective functions for data envelopment analysis are formulated and applied to industry data to demonstrate improved comparisons of manufacturing efficiency that evaluate differences in speed. |
