Arendt, J., Malak, R., and McAdams, D., "Design with Uncertain Technology Evolution," SAE Int. J. Passeng. Cars - Mech. Syst. 5(1):677-687, 2012, doi:10.4271/2012-01-0912.
A major decision to make in design projects is the selection of the best technology to provide some needed system functionality. In making this decision, the designer must consider the range of technologies available and the performance of each. During the useful life of the product, the technologies composing the product evolve as research and development efforts continue. The performance evolution rate of one technology may be such that even though it is not initially a preferably technology, it becomes a superior technology after a few years. Quantifying the evolution of these technologies complicates the technology selection decision. The selection of energy storage technology in the design of an electric car is one example of a difficult decision involving evolving technologies. In the first year of production, a particular battery cell technology provides the best vehicle performance; however, a few years later, an alternate battery technology may have evolved to a point that its performance surpasses the previous best battery technology. Designing with technologies whose levels of performance evolve over time is poorly understood. The objective of this research is to create a computational method allowing designers to make decisions with a greater understanding of the impact of the decision on the design. In this research, it is assumed that the performance of a technology develops in the form of an S-curve; slowly at first, quickly during heavy R&D spending, and slowly again as the technology approaches its limits. A key concept in this research is that performance evolution of a technology translates to a shift in the corresponding performance Pareto optimal frontier. The assumed form of technology development allows the designer to apply the uncertainty of technology development directly to the S-curve model rather than applying the uncertainty to the performance thus better reflecting the change in performance over time. A case study of designing a battery pack with evolving battery cell technologies in an electric car powertrain demonstrates this method.