Conference Paper: Dependency Network Analysis: Fostering The Future Of Space With New Tools And Techniques in Space Systems-Of-Systems Design And Architecture

Title: Dependency Network Analysis: Fostering The Future Of Space With New Tools And Techniques In Space Systems-Of-Systems Design And Architecture

Publication Date: 9/23/2013

Conference: The 64th International Astronautical Congress, Beijing, China


Advances in technology, causing increasing size and complexity of systems, are constantly urging system engineers to develop innovative tools and techniques to better address new challenges. The development of modern technologies towards advancing space systems capabilities has been equally met with increasing risk and cost. In order to minimize the impact due to the failure of a single launch or to budget cut, space exploration has been mainly characterized by a series of monolithic missions, largely independent from one another. Recently, new paths have been proposed to deal with new issues in space systems engineering: ESA developed a facility to apply the Concurrent Engineering method to the design of future space missions, thus recognizing the importance of integrated development, parallelization of tasks, and top-down approach, when dealing with systems having partial dependencies to each other. NASA has begun to interface spacecraft on different missions towards a common goal (Mars Reconnaissance Orbiter provides mapping of Mars, but also served as communications relay during the landing of Curiosity rover), thus adding flexibility and reliability to space exploration. Approaches aimed at increasing sustainability of space missions, and reliability and scalability of space systems (modularity, on -orbit assembly, network of intermediate exploration missions) have also been undertaken. Based on such considerations, this paper introduces new tools and techniques to address architecture and design of future space missions and systems. Recognizing the importance of interdependency between components in space systems, and among complex systems, we propose techniques to perform and exploit analysis of dependencies, both operational and developmental. We identify metrics to assess and measure the system value, with respect to desired features. Operational dependency analysis is then used to evaluate and compare different architectures and designs, both in the final stage of development, and when only partial development has been achieved. The paper then shows how the developmental dependency analysis is used to drive parallel development of component systems in order to increase flexibility and partial capability, and to reduce risk. The techniques are scalable and can be applied to a variety of scenarios.