How Outdated Software Caused The Recent FAA System Meltdown
Steve Taplin is the CEO and cofounder of Sonatafy Technology, providing experienced nearshore software developers and engineers.
The recent Southwest Airlines meltdown hit hard, seeing thousands of passengers stranded at the airport due to a costly glitch. The crippled system haulted operations at several airports nationwide, with numerous cancellations and redirections.
Southwest airlines relies on a Federal Aviation Administration (FAA) crew-assignment system, SkySolver. The software is an off-the-shelf solution developed and released by a company known as Caleb and is now owned by General Electric Co.
SkySolver was customized to match crew members to flights; however, Southwest executives reported that it was overwhelmed by the volume changes in crew schedules. GE Aerospace, in a statement, explained that the application is a back-end algorithm that collects input from various systems and recommends solutions to resolve crew-related disruptions. Usually, airlines can supplement the algorithm with other software.
Unfortunately, most carriers still need help with home-grown solutions, which analysts say were built on outdated legacy mainframe computers. These machines may have offered robustness in the past; however, they considerably need to catch up to modern technologies. Thus, mistakes of this magnitude were bound to occur. The failure may have been the sign necessary for the airline to find the urgency in scaling up its modernization efforts.
Other high-profile software issues occurred with Ticketmaster Entertainment when their website crashed for the Taylor Swift concert last November. The company blamed cyberattacks from bots and unusually high website traffic and demand to be the cause of website application outages. This software crash caused regulators to investigate if Ticketmaster has a monopoly on the ticket industry and controls ticket pricing.
What Led To The FAA Systems Failure?
The crisis came after what seemed like a routine scheduled systems check. A senior FAA official gave his brief, stating that the fault resulted from an innocent mistake one of its engineers made during maintenance. The staff had unknowingly replaced one file with another. The system began acting up by showing signs of redundancy, also known as cascading failure, and eventually stopped. The glitch also affected the Notice to all Air Missions (NOTAM) system, which usually sends out flight hazards, real-time restrictions and other essential information to the personnel.
However, the struggle to modernize their computer operations has led to the loss of millions of dollars. Like several other carriers, Southwest has been maintaining its systems well. However, the meltdown triggers executives to face an important decision they have procrastinated on for years. Despite their reservations about cost, they understood the implications of delaying modernization due to the underlying issues. Had the new system been in place, redundancies would have caught the outage before it occurred, thus saving everyone all the trouble and keeping the airline more money, time and resources.
How Can Startups Help Airlines Prevent Such Tech Mishaps?
There is a lot of potential in modern technology that could save airlines and their passengers’ expensive snags. Cloud-based infrastructure and databases solve scalability issues almost seamlessly thanks to the distribution of computing resources across the internet. Airline operations technology is complicated due to the numerous real-time data points and constraints on a single system. This poses severe IT challenges that cannot be cushioned by regular maintenance only.
For a long time, the airline IT market has been reserved for a few large companies due to the nature of technology that seemingly underpins everything making it difficult for new entrants to thrive. Additionally, implementing new and better solutions is difficult as the stakeholders view modernization as a painful process concerning time and money.
Nevertheless, smaller airlines are taking on the challenge by adopting these technologies that everyone seems to be afraid of. Big Blue, a Barcelona-based analytics company, uses linear programming techniques to solve issues around aircraft assignment. It currently serves Spanish low-carrier Volotea, which has a fleet of 41 aircraft. The company’s chief executive, Pau Collellmir, once shared that the company was working on creating a more comprehensive platform that would offer solutions for several entities, including aircraft, maintenance, flights, crews and passenger itineraries sequentially, as most systems do. The company is also working on scaling its solutions to serve larger carriers such as Southwest.
Airlines can also invest in startups and partnerships with cloud providers like Amazon Web Services and Google Cloud to improve their technologies. They could also maximize artificial intelligence systems to unlock innovative solutions such as passenger personalization and predictive maintenance. Building this kind of relationship with active members of the tech ecosystem could direct the future of the airline IT market and create a competitive environment that motivates airline software companies to do better.
Bottom Line
In the wake of the nationwide snag, Southwest airlines canceled and delayed flights due to the system issue. Such faults usually translate into billions of dollars in losses from irregular operations. According to market research by Frost & Sullivan, digital solutions such as air-disruption management systems can reduce these costs by nearly half.
It is also worth noting that computer faults pose a high risk for cybersecurity and could lead to worse implications; therefore, modernization is gradually becoming a necessity rather than an option from stakeholders in the airline industry. It is becoming clear that digital modernization is essential and not optional for systems to keep up with real-world demands. Company leaders need to innovate at a faster rate and modernize legacy systems. Since this could be a massive endeavor, breaking up the project into manageable pieces based on the importance of the systems is often the best place to start.
