Actualizing Flexible National Security Space Systems.

摘要

The National Space Policy of the United States of America (2010) calls on the National Security Space (NSS) community to continue to "develop and apply advanced technologies and capabilities that respond to changes to the threat environment." This national policy reinforces a trend within the NSS community that places a greater emphasis on the ability of NSS systems to operate in an uncertain future. This dissertation informs policymaking by presenting three essays that address some of the challenges associated with improving the flexibility of NSS capabilities. In addition, each essay includes an application of the decision logic that would be required to optimally support the next generation of flexible space systems.;The first essay explores the hypothesis that the NSS community can be more risk-tolerant when launching small satellites that are inexpensive and can be quickly replaced following a launch failure. The ability to bundle multiple payloads on a single launch vehicle is a decision unique to small satellites that adds an extra dimension to the launch risk calculation. While bundling multiple small satellites on a single launch vehicle spreads the initial launch cost across multiple payloads, this strategy also makes a launch failure more costly. This essay develops an analytic framework for constructing optimal small satellite launch strategies for a range of risk preferences. Given the available fleet of launch options, targeted small satellite launch demand, and nominal small satellite attributes, risk-adverse decisionmakers would always seek to minimize the expected cost when choosing how best to launch a set of small satellites. Simple modifications to existing technologies would have the largest impacts on cost and risk. These modifications include the ability to easily dual manifest on a medium-cost, medium-reliability launch vehicle and a secondary payload adaptor that can support three 360-kilogram payloads. Developing and fielding a low-cost, low-reliability launch option offers the greatest reductions in financial risk but would significantly increase the mission risk associated with launching small satellites.;The second essay examines the potential role for ridesharing small additional payloads on launch missions purchased in support of large and expensive primary payloads. The U.S. Air Force and its commercial partners have developed a suite of secondary payload adaptors that can accommodate a wide range of potential payloads. While increasing the number of rideshare opportunities could make access to space more affordable for small satellites, challenges persist. One concern is that by increasing the complexity of the launch vehicle integration process, ridesharing auxiliary satellite systems could increase the risk exposure to the primary satellite system. In this essay, I show that, historically, ridesharing additional payloads has reduced the reliability of a launch attempt by 1.25 percentage points. Given available data, any increase in the delay of launches is difficult to detect and probably insignificant when compared to the overall development timeline of an NSS satellite system. This essay concludes with an enterprise-level assessment of the expected savings associated with a generic rideshare opportunity. Given reasonable estimates of the cost and reliability of secondary launch options, it is only cost-effective to rideshare additional payloads with primary satellite systems that cost less than ;The third essay examines how uncertainty and flexibility would be incorporated into the systems analysis process for new NSS systems. While the Department of Defense (DoD) has traditionally used a cost-centric framework for systems analysis, the academic community of practice has suggested that a value-centric framework would be more appropriate for flexible space systems. This essay finds that the value-centric framework aims to identify a solution that is optimally aligned with the preferences of the decisionmaker, while the cost-centric framework is a convenient method for identifying a solution that satisfactorily addresses the preferences of the decisionmaker. The promise of identifying an optimal solution is appealing, but practical challenges dampen the usefulness of the value-centric framework in the context of the DoD decision environment. Specifically, the incommensurable nature of military effectiveness makes it very hard to identify a consistent value model for DoD systems. This essay concludes with a discussion of how the cost-centric framework can be modified to better incorporate uncertainty into the design and selection of flexible NSS systems.*;*This manuscript was formatted assuming that the reader would have access to a color copy. Interested readers who obtain a copy that is difficult to read may contact the author at Kenneth.Grosselin gmail.com for a color copy.