Risk-based technology portfolio optimization for early space mission design

The successful design, development, and operation of space missions requires informed decisions to be made across a vast array of investment, scientific, technological, and operational issues. In the work reported in this paper, we address the problem of determining optimal technology investment portfolios that minimize mission risk and maximize the expected science return of the mission. We model several relationships that explicitly link investment in technologies to mission risk and expected science return. To represent and compute these causal and computational dependencies, we introduce a generalization of influence diagrams that we call inference nets. To illustrate the approach, we present results from its application to a technology portfolio investment trade study done for a specific scenario for the projected 2009 Mars MSL mission. This case study examines the impact of investments in precision landing and long-range roving technologies on the mission capability, and the associated risk, of visiting a set of preselected science sites. We show how an optimal investment strategy can be found that minimizes the mission risk given a fixed total technology investment budget, or alternatively how to determine the minimum budget required to achieve a given acceptable mission risk.