This paper presents a methodology and tool for designing a portfolio of Earth observing missions using small satellites with commonality of components across missions. The methodology is based on the theory of product platforms, which has been extensively used in mature industries such as automotive and consumer electronics, but has not been as fruitful in the space industry, mostly due to the small market and number of units. In order to determine the optimal level of commonality for a portfolio of space missions, the trade-off between cost and performance is considered. High commonality leads to large cost and risk reductions, but also reduces the range of performance that can be achieved by the portfolio of missions. Low commonality enables a wider range of performance at the expense of higher cost and risk. The methodology presented here takes this trade-off into account and allows a decision maker to determine the preferred level of commonality and a specific set of modules to reuse for a given portfolio of missions. The methodology formulates this problem as a dual-stage mixed-integer bi-objective optimization problem and uses a graph-based heuristic and a multi-objective evolutionary algorithm to solve it. This paper describes the methodology and illustrates its application to a portfolio of CubeSat missions based on existing and plausible mission requirements.
展开▼
