Before CAE RISE, There Was SOQA

More than a decade ago, CAE leveraged the flight operational quality assurance (FOQA) process, which analyzes flight data to improve pilot performance, to create SOQA – Simulator Operations Quality Assurance, which compares simulator-derived data with FOQA data. CAE tapped the capabilities of modern full-flight simulators (FFSs) to offer SOQA evidence-based training tools as a new product.

CAE’s first SOQA implementation was for the US Department of Defense, using Flightscape (a commercialization of a Transport Canada program) in late 2009 to compare military FOQA and military FFS data in a SOQA program designed to prevent accidents and incidents. The program used a Boeing 737 simulator at CAE SimuFlite in Dallas.

The basic idea behind SOQA is that simulators are repositories of valuable information, as are the recorders on aircraft that log information used in FOQA programs. SOQA offers a quantitative bridge between a safety management system (or any safety program) and training, explained Captain Lou Nemeth, then-chief flight safety officer for CAE. Typically, he said, when a safety issue is raised or as he put it, a latent unsafe indicator is discovered, “90 percent of the time the mitigation strategy is ‘we need to do more training.’”   

A key element of both FOQA and SOQA is that the data can help identify areas where pilots are flying the simulator differently from the way they fly the airplane. Another advantage of SOQA is that the computer is better than an instructor at spotting pilot deviations, and allowing the computer to do its job frees up the instructor to focus on human factors and other hard-to-measure behaviors.

“SOQA will help the instructor detect things that in the past have gone undetected,” said Nemeth. A simulator instructor might not, for example, notice that a student nearly scrapes a wingtip during an engine-out takeoff. But the computer records this exceedance. And the data is used to drive an animation of the training event so the student and instructor can easily see that undetected event and figure out a way to prevent that.

“With SOQA in the environment,” said Nemeth, “it alerts the instructor to the things that heretofore have gone undetected.” As another example, he pointed out, “If a pilot exceeded maneuvering speed in the simulator, it may very well have gone undetected.” And that kind of flying behavior can cause serious consequences in the real world. “It’s training based on evidence instead of training based on prescriptive regulations,” he said.

Another benefit of SOQA is the ability to create animations of accident events and replay them for pilots during training. CAE used an animation of a 737 nose-low, high-bank situation as a tool to poll instructors on how they would teach pilots to handle the airplane if that happened.

As SOQA moves into the airline and business aviation fields [I wrote in 2011], someone will have to pay for the service, including equipping simulators for the necessary data gathering and replay capability as well as analysis. CAE used its own simulators. The use of SOQA will benefit both initial and recurrent training, according to Nemeth, and the desire for SOQA should come from safety officers who see the benefit of mitigating safety threats through use of SOQA data.

“I think SOQA will be popular with airlines and business jet operators that see safety as a strongly held value in their organizations and believe in a systems approach versus a prescriptive regulatory approach,” Nemeth concluded. “This is the way forward. Prescriptive regulatory training is going to end.”

Excerpted from The Robot in the Simulator, Artificial Intelligence in Aviation Training by Rick Adams, FRAeS

Available in paperback or PDF – NEW BOOK – AI in Aviation Training – AVIATION VOICES