AIAA Propulsion and Energy Conference: For Systems Engineers, It's all About the Architecture

©AIAA

AIAA Propulsion and Energy NASA exhibit.

Much like an architect can design the Parthenon or the Tacoma Narrows Bridge, so can an architect design a resilient system or one that breaks under strain.

That was the message from the panel, "Assuring Critical System Behavior in an Era of Increasing Complexity and Change," at the AIAA Propulsion and Energy Forum in Cleveland. The panel was moderated by John Thomas, a former president of the International Council on Systems Engineering or INCOSE.

The panel considered what is needed to create and maintain aerospace software and electrical systems that can meet today's changing demands and performance pressures.

Systems need to be adaptable and that requirement must be considered at the outset whenever possible.

"More and more since we add stuff to systems, we need to make sure we have [an] architecture that allows that," said Paul Nielsen, CEO of the Software Engineering Institute. Nielsen said a system can be designed from the start to accommodate new functionalities without the need to re-test the entire system.

Another goal, Nielsen said, must be resiliency. "If there's a fire in the city, the entire city doesn't shut down, they find a way to put out the fire and the city keeps operating," said. It must work the same way within a complex aerospace system, he suggested.

A system architect must accept that failures will happen and think strategically about that. "You design the system so the generator blows out and shuts the system down, before an overload destroys the components," said Don Burns, head of innovation systems engineering at Rolls-Royce.

Burns stressed that when many parts of a team work on a system, no one part of the team should consider its component the "master of the universe." A systemic viewpoint leads to a better overall system design.

The panel also advocated including many viewpoints and departments in the planning phase. Eric Gebhardt, chief technology officer and vice president of engineering for GE Oil and Gas, said that after the BP oil rig explosion and spill, his company assembled staff from its nuclear power and aviation wings to provide insight and advice for the design of blowout resistant systems.

Moderator Thomas and Nielsen pointed out that we live in an increasingly complex and connected world, which means each connection node represents a possible failure point. In light of that, the panel urged designers to pay attention to the weak points, and plan for ways to mitigate the damage that their failure can cause. Failure will happen -- you can't avoid it -- but you can minimize its effects.

Gebhardt pointed out that this type of planning is hard, because it's rare to design an entirely new system. "A lot of what we do now is brownfield systems and we get so few chances to engineer greenfield systems - meaning all we are doing is adding on to what is there."

By Duane Hyland, AIAA Communications

Please follow SpaceRef on Twitter and Like us on Facebook.