I arrived at JPL about 7 months ago, and since then I have learned to speak the engineering language of Requirements, a foreign language to me as I studied physics at college. Here at JPL, most of our missions are designed to accomplish science goals. In other words the objective of most missions is to contribute to scientific communities and the public by providing one-of-a-kind data sets achievable only by having a one-of-a-kind satellite or rover (for example). In my section, Mission System Concepts, we are often tasked with developing a Science Traceability Matrix (STM) for our mission concept. By doing so, we can map the technical requirements for the subsystems to some science requirement we want to accomplish, and show that our design is complete and really does what it is supposed to.
I have recently been working with a team of scientists and systems engineers to write a STM for an instrument concept. I naively thought that this would be a straightforward task. I thought that the scientists would say what they want to measure and then we all decide what that requires of the instrument. Definitely not an easy task! First of all, scientists always want as much data as they can get. While I would like to believe that all mission or instrument designs are driven by science goals, they are mostly driven by cost (and schedule). More data and capability almost always means more money. So, the real task of developing an STM is working with the scientists to decide what science must really be accomplished in order for it to be worth building the instrument or mission. I have quickly learned that requirements are not so much “flowed” down from science to measurement to instrument. It is more of a cyclical process because of the fact that every instrument/mission has finite technical and fiscal resources. (Unfortunately, I have to be vague about the “instruments” and “missions” I work on because they are still in concept phase and are competition sensitive.) One general example that comes to mind is telescope size. The bigger a telescope aperture, the more light collected, the more information collected. Bigger the better! Let’s say, however, that due to cost and technical feasibility, we can only have a telescope with a 1.0 m diameter mirror. Even though scientists would like the data-set attainable with a 1.5 m telescope, and they could come up with many science objectives achievable with that data-set, they will have to put up with less capability. So, it now has to be decided what science objectives can be accomplished with the 1.0m telescope, whether they would want to put in their time working on such a mission, and whether the scientific community would appreciate and use such data.
So, for hours I sat trying desperately to jot down anything I heard that seemed to resemble a requirement. Even with a group of very capable and focused scientists, by the end of the 3-hour meeting we had about 2 or 3 science requirements!