Complexity analysis of the cost effectiveness of PI-led NASA science missions

For the last 20 years, NASA has allowed Principal Investigators (PIs) to manage the development of many unmanned space projects. Advocates of PI-led projects believe that a PI-led implementation can result in a project being developed at lower cost and shorter schedule than other implementation modes. This paper seeks to test this hypothesis by comparing the actual costs of NASA and other comparable projects developed under different implementation modes. The Aerospace Corporation's Complexity-Based Risk Assessment (CoBRA) analysis tool is used to normalize the projects such that the cost can be compared for equivalent project complexities. The data is examined both by complexity and by launch year. Cost growth will also be examined for any correlation with implementation mode. Defined in many NASA Announcements of Opportunity (AOs), a PI-led project is characterized by a central, single person with full responsibility for assembling a team and for the project's scientific integrity and the implementation and integrity of all other aspects of the mission, while operating under a cost cap. PIs have larger degrees of freedom to achieve the stated goals within NASA guidelines and oversight. This study leverages the definitions and results of previous National Research Council studies of PI-led projects. Aerospace has defined a complexity index, derived from mission performance, mass, power, and technology choices, to arrive at a broad representation of missions for purposes of comparison. Over a decade of research has established a correlation between mission complexity and spacecraft development cost and schedule. This complexity analysis, CoBRA, is applied to compare a PI-led set of New Frontiers, Discovery, Explorers, and Earth System Science Pathfinder missions to the overall NASA mission dataset. This reveals the complexity trends against development costs, cost growth, and development era.