Jet Propulsion Laboratories (JPL) was tasked with doing the conceptual designs of a mission to Jupiter, a project that would later be named Voyager. JPL was organized around the subsystems of a spacecraft--communications, power, structures, attitude control, computing and sequencing, and so on. In the systems division the job of a systems engineer was the overall architecture of the spacecraft and working out the coordination among the specifications of all the different subsystems.
The basic constraint was weight. It was expected that the Titan IIIC rocket would launch about 1,200 pounds into a trajectory toward Jupiter. Using a Saturn 1B would allow for a 3,000 pound spacecraft. Two designs were sketched out over a year, each with a different configuration based on weight. With 3,000 pounds to work with, the design work was relatively easy. The designers could essentially bolt together fairly well-understood subsystems. Consequently, the divisions would not have to coordinate very much because the design challenge was relatively low. But if they only had 1,200 pounds to work with, things were more difficult. Interactions began to play a big role.
Most of the work in systems design is figuring out the interactions, or trade-offs, as they were called. The moment you tried to optimize any one part, that choice immediately posed problems for other parts. The weight constraint made the whole thing a web of competing needs, and it all had to be considered together. Each part of the system had to be reconsidered and shaped to the needs of the rest of the system. A great deal of work went into trying to create clever configurations that avoided wasteful duplication. This work required learning about all of the subsystems and their possible interactions in order to imagine a configuration that might be effective. It was difficult to say the least, but it is also the beginning of learning strategy.
ACTION POINT: Understand the interactions involved when changing one part of a system.
No comments:
Post a Comment