It’s no surprise he speaks about staffing and tech in the same day. An FAA-commissioned study in 2023 found that the top issues with air traffic safety are inadequate controller staffing, and outdated technology. My time working with the FAA proved these observations to be true as well.
As a human factors engineer at Raytheon, I worked on the NextGen Weather system for the FAA from 2016-2018. My job was to study controllers in their workplace, and design our software features according to their needs and preferences. But I wound up being a mediator between the end-users and FAA-aligned consultants.
“The FAA”—the administrative employees who allocated budget for and oversaw our project, along with their technical consultants from academia—didn't particularly care what the controllers thought. The controllers likely only had a seat at the table because of the negotiations of their union, National Air Traffic Controllers Association (NATCA).
We were trying to build streamlined, controller-focused software; the FAA wanted to add consultant-requested bells and whistles that would delay the project, and ultimately present distractions to controllers through unnecessary buttons and options. The power these consultants wielded was perplexing. It was a terrible dilemma in which to be placed.
So when you see reports about “outdated technology,” it’s not always because the FAA doesn’t have funding, or waits too long. Some of the technology spends unnecessary years in development, while people argue in a conference room. While consultants hurl insults at the Subject Matter Expert air traffic controllers and engineers, and the latter threaten to quit. I’d never seen anything like it, and I’ve sat in joint meetings with customers from every service in the military.
The FAA did not respect the controllers or the engineers trying to serve them. But it didn’t have to be this way. All the FAA needed to do was listen to the controllers, and let us build accordingly. But the FAA couldn’t get out of its own way in the development of great tech.
The controllers were brilliant and reasonable. There was a whole culture and way of talking, similar to military culture. You were glad this crew had the safety-critical work of air traffic under control. I never met a bad or even “ok” controller. Everyone who makes it out of the grueling 2-year training process is a rockstar.
But they looked tired. On some occasions they couldn’t get shift coverage to meet at our offices for the FAA-scheduled product evaluations. That was a pivotal time in the ongoing staffing issue that had plagued the FAA for decades, and would continue to, up to the present day.
The illegal air traffic controllers’ strike in 1981 led to President Reagan’s firing of 11,359 out of the FAA’s 17,000 controllers. They were quickly replaced with a massive “post-strike” cohort of young controllers who would retire around 2007-2017. Over the years, the FAA would fail to keep staffing high enough to stay ahead of the curve.
By the beginning of my time with the FAA in 2016, then-President of NATCA, Paul Rinaldi, testified to Congress that the FAA had a “staffing crisis.” The controller workforce had actually been shrinking every year since 2011, with only 10,667 total certified controllers at that time. Controllers were pulling overtime to make up the difference, working 6-day weeks, task-loaded to saturation. For years on end. This is why they looked rough.
That the FAA had introduced the social justice experiment known as the Biographical Qualification (BQ) in 2013, justified by pseudo-science research out of the FAA’s human factors engineering division, only exacerbated the broader staffing issue. Any arbitrarily-promoted, unqualified candidates would only fail the Air Traffic Selection and Training (AT-SAT) test or wash out of the 4-month FAA Academy, wasting time and training seats. The FAA had 10,593 certified controllers as of 2023, a 34-year low signaling a larger problem.
The FAA’s internal dysfunction resulting in inadequate staffing also plays out in the lopsided state of its technology. To gaze upon an air traffic facility, at least around 8 years ago, was a crapshoot. Some looked like the War Room of Dr. Strangelove, outfitted with slick wall-to-wall displays. Others resembled a ‘90s LAN party with monitors crammed into every nook and buckets of gummy bears in the corner. I could not account for the disparity.
Our project was going to provide state of the art aviation weather data, like a kicked-up version of Weather Underground. To my horror, some of the controllers were mostly excited because they wouldn't have to turn around anymore. Some of the more cramped air traffic facilities did not have enough room to accommodate the growing number of dedicated information displays. At these sites, some monitors had been re-arranged to be positioned behind you; including, inexplicably, the weather.
So for years, these controllers had been looking ahead at the radar screen with the planes, then turning around to get a good look at the weather, then turning around again to look back at the planes and approximate from memory. For whatever reason, our new system would include moving the weather display back to the same vicinity as the radar display.
But we had an even easier solution for the cramped-displays problem. We also made the radar display showing the planes, called STARS. We could overlay a simplified version of the weather map right on the radar interface. Like having a weather layer on Google Maps. The technology wasn’t there before, but we could do it now. The controllers could still have a dedicated weather display for occasional detailed review, placed wherever. The controllers wanted this. The FAA said no.
We offered to make the weather variable transparency, or toggle on and off, or come in different color schemes. The FAA said no. “Weather on STARS!” the controllers and engineers romantically chanted up and down the hallways. We went ahead and built Weather on STARS on our own time, and showed it at the NATCA conferences. The FAA again said no.
It’s as though the FAA was trying to spend more time and money making a more complicated technology environment for the controllers. I left my project long before it was done; originally a 2-year contract, it went several years over schedule and millions of dollars over budget, possibly not deployed even today. There’s plenty more I could say about why, but that’s boring. When you think of the FAA, think of how some controllers had to turn around and memorize the weather while other controllers were told they were wrong by third party consultants.
In defense engineering there’s this concept that when there’s a perceived emergency, stuff gets done. When there is no perceived emergency, stuff never finishes. In 1981, the FAA recognized a staffing emergency and solved it. Why hasn’t staffing been a continuous emergency since then, the same way keeping planes separated holds emergency-level concern for controllers every moment of the day? Why isn’t the timely delivery of ego-free, air traffic-specific technology an emergency?
The FAA needs to think bigger than it has generationally done, to both create and fill a larger pipeline of recruits. And it needs to think holistically about technology rather than worrying about the stove-piped needs of the bureaucratic offices of technology managers. The FAA needs to become truly controller-focused, to become a great steward of air traffic and the flying public.
