Concept design plays a central role in project success for space missions, as the product of concept design effectively locks in the majority of system lifecycle cost. It involves a concurrent investigation of requirements and multiple mission characteristics such as flight dynamics, design, performance, concept of operations, technology, verification approach, launch and ground interfaces, cost, schedule, and risk.
Done well, concept design can provide an executable system-level design baseline for project teams in preliminary design and later project phases. Not done well, concept design can lead to several undesired outcomes including cost overruns, schedule delays, the need for redesign or multiple redesigns, fluid technical baselines, and contract disputes or cancellations.
The extraordinary leverage concept design has on system lifecycle cost presents a business case for conducting concept design in a credible fashion, particularly for first-of-a-kind systems that advance the state of the art and that have high design uncertainty. A key challenge, however, is to know when credible design convergence has been achieved for such systems.
Using a space system example, this work describes a process suited for conducting comprehensive concept design and design-to-cost trade studies for such systems. Aspects discussed are: a) what concept design is and why it is important, b) the level of convergence needed for the concept design product in terms of customary technical and programmatic resource margins available in preliminary design, c) techniques for designing the mission-level trade space, and d) challenges in determining credible design convergence.
This work illustrates a systematic trade study process for exploring significantly different mission concepts over a few discrete design “cycles” with the objective of credibly converging the technical, cost, and schedule characteristics for a single baseline mission concept design to the first order before entering preliminary design. Trade study cases bound the trade space, and the final (baseline) solution is deduced by interpolation. More like a root-finding algorithm than like the successive refinement approach typically used in preliminary and detailed design, it enables broad coverage of the trade space in minimum time. It also helps illuminate unexpected findings, including major “unknown unknowns” that otherwise may have remained hidden until preliminary design or later phases wherein their discovery could induce significant impacts.
Key in this process is focusing at the first order level for sizing and performance, and deferring second and third order considerations to later phases. Equally important is recognizing that early team system performance expectations and early team cost estimates tend to be optimistic. Cost estimates tend to increase as teams learn more about both the design and the work breakdown structure used for costing. Apparent only in hindsight, this effect often significantly moderates team performance expectations and advises against selecting a design baseline from early cycle results. Team learning typically tapers off for most designs after three cycles when truly bounding trade study cases have been evaluated.
The techniques described here have been used at NASA Goddard Space Flight Center for the development of mission concept designs.