System architecture decisions under uncertainty : a case study on automotive battery system design

Abstract

Flexibility analysis using the Real Options framework is typically utilized on high-level architectural decisions. Using Real Options, a company may develop strategies to mitigate downside risk for future uncertainties while developing upside opportunities. The MIT-Ford Alliance has extended the techniques of flexibility analysis beyond high-level architecture to core product design decisions in future vehicle electrification. This thesis provides a methodology for a real-time support framework for developing novel engineering decisions. Risk is high in new product introduction. For hybrid and electric vehicles, market demand and technology forecasts have substantial uncertainty. The uncertainty is anticipated, as the high voltage battery pack hardware and control system architecture will experience multiple engineering development cycles in the next 20 years. Flexibility in product design could mitigate future risk due to uncertainty. By understanding the potential iteration of core technologies, the engineering team can provide flexibility in battery pack voltage monitoring, thermal control, and support software systems to meet future needs. The methodology used in this thesis has been applied in a Ford-MIT Alliance project. The Ford and MIT teams have valued key items within the core technology subsystems and have developed flexible strategies to allow Ford to capture upside potential while protecting against downside risk, with little-to-no extra cost at this early stage of development. A novel voltage monitoring technique and a unique flexible thermal control strategy have been identified and are under consideration by Ford. The flexibility methodology provided motivation and support for unique decisions made during product design by the Ford team.