Flight Data Monitoring Comes into Its Own
Flight Data Monitoring (FDM) is a proactive system for capturing, analyzing, and visualizing data recorded during routine commercial and general aviation flights. Its roots go back to 1939, when France’s François Hussenot and Paul Beaudouin developed the “Type HB” flight recorder to capture basic readings such as altitude and speed during test flights. This FDM solution was rather unique, as it used a mirror-reflected thin beam of light hitting a moving reel of film to record flight data. (The film reel was 8 meters long and 88 millimeters wide.) The mirror tilted in real-time to reflect changes in measured parameters over time, which was traced on the film like a seismograph’s track.
The concept of the Type HB recorder evolved into modern FDM devices, one of which is the well-known ‘black box’ flight data recorder. But FDMs have progressed far beyond this point to become the foundation of sensor-based inflight operations monitoring.
Today, the wealth of this data, and the modern ability to analyse it in great detail, have made FDM a key fleet management tool for aircraft operators large and small. It has also inspired the development of many cutting-edge FDM products and services.
FDM was pivotal in resolving the following situation: At the time of going to press, November 28th, Airbus announced that approximately 6,000 A320 Family aircraft would require immediate precautionary software updates, after analysis of a recent JetBlue flight revealed that intense solar radiation may corrupt data critical to the functioning of flight control systems. Airbus has worked proactively with the aviation authorities to request immediate precautionary action from airline operators via an Alert Operators Transmission (AOT) in order to implement the available software and/or hardware protection, and ensure the fleet is safe to fly. This AOT will be reflected in an Emergency Airworthiness Directive from the European Union Aviation Safety Agency (EASA).
In this article, we will examine the trends and challenges shaping modern FDM, with the help of four companies that specialize in this technology. They are ERGOSS, Fliant, FLYHT, and Scaled Analytics.
ERGOSS specializes in solutions for the processing and analysis of massive aircraft data. The company’s main SARA-FDM module was originally designed as an accident analysis solution, and was used as such in investigations with the French BEA and the American NTSB.
SARA-FDM’s initial version evolved into a complete FDM solution that is being used by airline groups such as AVIANCA and AIRFRANCE-KLM, among others.
“SARA solutions can process all known types of flight data recorders, regardless of the recording frequency, without any resampling of parameters,” said Fabrice Tricoire, CEO of ERGOSS. “All types of civil aircraft (and military transport and business aircraft) can benefit from SARA, from the smallest to the largest. We even process COMAC C909 (ARJ21) aircraft and have worked on SUKOI SuperJet 100s.”
Fliant is building a unified flight data analytics platform that, among other use-cases, supports FDM/FOQA. Its FDM solution is called FlightVue FDM. This is a cloud solution that automates the entire process from data collection to reporting and stakeholder engagement. It works together with another product called FlightAdvise to provide automated FDM de-briefings to pilots.
“Our platform was built to support a multitude of aviation time-series data sources,” said Iliya Maksimov, Fliant’s Co-founder and Business Leader. “In the FDM/FDA context, this is usually data downloaded from FDR/QAR devices in the ARINC 717/767 binary formats.”
FLYHT provides three types of Automatic Flight Information Reporting Systems (AFIRS) that support airline FDM. AFIRS 228 and AFIRS Edge+ units are avionics data and connectivity systems that support data recording and harvesting of sensor data from aircraft. AFIRS UpTime and AFIRS Gateway are services that enable management and automatic distribution of data transmitted by the AFIRS products. AFIRS ELA or Edge Logic Application runs in the AFIRS avionics units enabling real-time FDM onboard aircraft.
“The AFIRS 228 system provides a QAR and ELA (real-time FDM) capability on over 1,000 aircraft and is certified on DHC-8s, ATRs, CRJs, ERJs, B737s, A320s, A330s, B747s, B757s, B767s and B777s,” said William Cecil, FLYHT’s Senior Director, Aircraft Data & Edge Solutions. “The new 5G AFIRS Edge systems are certified for Boeing 737NG, MAX and A320CEO, NEO with A330 and Embraer jet STCs planned in early 2026.”
Finally, Scaled Analytics provides flight data analysis solutions that help aircraft operators improve safety, efficiency and compliance. Its flagship product, Sky Analyst FDM, is a Flight Data Monitoring platform designed specifically for civilian and business aviation (fixed wing and helicopter). It automates the collection and analysis of flight recorder data, helping operators identify trends, monitor exceedances, and proactively manage risks and safety departments.
“We support any aircraft with data stored in ARINC 717, 767, 429, and CSV format,” said Dion Bozec, Founder and President of Scaled Analytics. “We use our own proprietary software for decoding the data and programming events.
Trends in FDM
According to Iliya Maksimov, flight data monitoring has been a “compliance checkbox” for many airlines for decades. “Only a few of them, usually large operators, have really used FDM in an open and non-siloed way to drive safety improvements and engagement,” he told Aerospace Innovations. “However, thanks to advances in cloud technologies and AI, together with the growing requirements of regulators to engage a wider audience inside the airline with FDM results, the industry is slowly opening FDM/FDA programs and engaging more internal and external stakeholders.”
So what are the trends driving FDM today? “One of the biggest trends today is the democratization of data,” replied Dion Bozec. “It’s becoming easier and more affordable for operators of all sizes to access and analyze flight recorder information. That’s certainly a good thing for aviation safety overall, as more organizations are now able to adopt data-driven safety practices that were once limited to major airlines. At the same time, we’re seeing a growing focus on integration and usability. Operators want systems that not only process flight data but also connect directly with their Safety Management System (SMS).”
According to William Cecil, FDM has been traditionally seen as a flight safety function that supported airline SMS. However, this is changing. “More recently, airlines are utilizing FDM technology and data to reduce costs,” he said. “As well, FDM is increasingly used by airlines for operational efficiency, aircraft health management and predictive maintenance use cases.”
As for Fabrice Tricoire’s take on FDM trends? “A major trend in recent years — pioneered in 2014 by ERGOSS with SARA-PWI — has been to put cockpit flight crews at the center of the flight data usage loop,” he told Aerospace Innovations. “As pilots now have the ability to review all their flights, deviations, trends, and statistics, they feel truly involved and are major players in flight data analysis. As well, there’s a trend in FDM toward automation. This is a good thing as long as it is limited to repetitive tasks with low-added value, or the detailed analysis of user behaviour in response to clearly erroneous results or spurious alarms that are known to operators.”
Cybersecurity Just One of FDM’s Challenges
There are many challenges confronting flight data monitoring today. A case in point: “Cybersecurity is a great concern nowadays,” said Maksimov. He then quipped, “you know your company is gaining traction when you start receiving targeted hacking attempts.”
Becoming serious once again, Maksimov continued by saying, “when it comes to FDM, it’s important to always keep in mind that the flight data is highly sensitive data — both for the business and for the individuals involved. In fact,
flight data should be considered personal data as it represents a recording of the pilots’ actions. From a business perspective, it’s a recording of the airline’s core activity. Now imagine someone gaining access to Coca-Cola factory surveillance footage and seeing how they mix the ingredients. This is the same for an airline when it comes to the security of their flight data.”
Unfortunately, legacy FDM solutions were built in a different epoch on the cyber-risk timeline, said Maksimov. As a result, “the way that data is being secured in such systems can be ridiculous in some instances. We know about several cases where operator FDM programs were crippled by ransomware attacks. This is something that can have a direct impact on safety standards and can directly affect an airline’s safety oversight if the operator cannot access their data.”
“Another challenge is the speed of processing and analysis,” Tricoire said. This is a problem that ERGOSS has addressed by virtue of being a web-based FDM platform. “To illustrate this performance, a major operator that conducts intensive operations in hubs requires SARA to report any alerts to the Maintenance Control Centre (MCC) within five minutes of the aircraft sending the data upon arrival,” he explained. “This allows the MCC to make strategic decisions regarding the reassignment or exchange of aircraft in its fleet of nearly 500 aircraft, thereby avoiding the hassle and costs associated with AOGs.”
Scaled Analytics’ Bozec sees two more challenges affecting FDM. “First, the term ‘FDM’ is being used for programs that are clearly not flight data monitoring,” he said. “We’ve already started seeing customers come to us because they were with a vendor who was supposed to have offered them an FDM program, only to find that it did not meet EASA requirements. Another emerging challenge is that a small number of aircraft manufacturers have started restricting access to detailed parameter documentation that was once widely available to operators. This trend can make it harder for operators to implement independent FDM programs while limiting competition, especially among smaller operators.”
FLYHT’s Cecil sees four more challenges facing FDM today. These include expanded data harvesting to support new use cases — especially aircraft health management and predictive maintenance — and custom data recording encoding that requires corresponding custom decoding for all the services that consume the data. This second challenge “makes it difficult for service providers to onboard new customers,” he noted. As well, it can be difficult to access the necessary high-speed wireless data connections to move this data around, at a time when cellular technology obsolescence is forcing airlines to move from 2G and 3G to 4G and eventually 5G transmission equipment.
AI’s Role is Uncertain
Given the massive amount of data consumed by FDM systems, one would expect FDM vendors to be solidly in favour of AI to process and analyse this data quickly. But that’s not the case: “As with AI in other sectors, the aviation industry is still figuring it out,” said Maksimov. “With Flight Ops and Safety being the most conservative units in an airline, AI seems to be propagating very slowly.”
Fliant shares the airline industry’s conservative view of AI in FDM. “We are very careful when we talk about AI being part of the process,” Maksimov told Aerospace Innovations. “This doesn’t mean we don’t see the use cases though. But those use cases don’t start with building a ‘Minority Report’ (with a reference to the old movie) reality where artificial intelligence is highlighting safety-related occurrences way before they happen. No, the implementation of AI in FDM starts deep down within the data — from verifying the quality of the data and fixing sensor issues, through optimization of how the data is being onboarded all the way to detecting anomalies that do not fall under the strictly-defined exceedance trigger logics.”
Tricoire is similarly cautious about using AI in FDM. “We must be wary of excessive incorporation of AI, which will only serve to lower the accuracy of analyses if we are not careful and place too much trust into algorithms versus actual brain thinking,” he warned. “Our solutions should focus instead on analyzing massive amounts of data to detect clusters of weak signals that are insidiously building today’s deviations for tomorrow. SARA is there to draw analysts’ attention to these ‘insidious signals’.”
Bozec is a bit less conservative in his view of AI and FDM. “Scaled Analytics has a number of AI projects in development that I cannot comment on, but in the general sense, it is safe to say that FDM will benefit considerably through advances in AI,” he said. “For instance, a lot of routine data analysis can be automated using AI. Data quality checking can be done by AI algorithms as well.”
“We always prefer to work with official data documentation,” Bozec added. “However, for older aircraft — particularly those in regions where documentation may have been lost or is no longer supported — modern AI tools can help operators interpret available data more efficiently and establish a baseline for analysis. This makes it possible for them to begin monitoring flight data responsibly, even when full manufacturer documentation isn’t accessible.”
Cecil is more upbeat about AI than his colleagues. “Machine Learning (an aspect of AI) is playing a big role in predictive maintenance,” he said. “It is being used to develop predictive models that can ‘listen’ to the data being harvested from the aircraft, identifying failure precursors, and providing sufficient advance notice of component failures so that unscheduled maintenance can be converted into scheduled before a failure impacts normal operations.”
All the above being said, “AI is expensive, at least for now and when done right,” said Bozec. “So the real test will be to see how well the market will support the costs associated with AI-supported FDM and if those costs justify the benefits. I think they do, but we don’t know if the airlines share this opinion.”
What’s Coming Next in FDM
Despite their differing views, the experts interviewed for this story agree that great advances are coming to FDM in the years and months to come.
“Three advances we can expect are: FDM will cover use cases relevant to ground operations while the engines are off, FDM will consume multiple channels of aircraft system data in parallel which will remove the need for custom QAR encoding and enable faster onboarding for FDM service providers, and FDM will adapt to support consumption of data streamed real-time from the aircraft while in flight,” said Cecil.
“I expect to see FDM fully integrated with SMS, plus more advanced event detection engines,” Bozec said. “I see those as advances that will genuinely help operators improve safety and efficiency. They might not be as exciting as flight animations with blades of grass blowing in the wind, but we see these as genuine improvements that will help operators get more insights from their safety data (FDM and SMS). I also see more advanced FDM tools being adapted to bring more benefits to maintenance departments, so more of the organization can benefit from the analysis of flight data. This is something we’ve already begun working with our customers on and I’m pretty excited for the future on the aircraft maintenance side of things.”
As for Tricoire’s predictions? “Future advances — some of which we are already working on — concern the plurality of information sources to complement the use of raw aircraft data,” he replied. “In addition to information such as METARs, ACARS, and ADS-B, which is becoming increasingly widespread, we can also consider integrating information from training sessions to better understand individual pilot trends, statistics and reactions. Last but not least, we are moving towards very high levels of immersion in interactive flight replay at increasingly optimized costs, making virtual reality accessible to every pilot and analyst, who can now find themselves in a quasi-realistic environment for a better understanding of the root causes of deviations from SOPs.”
“At Fliant we see a bright future for FDM programs globally,” said Maksimov. “From a regulatory environment standpoint we see EASA focusing more and more on what next-generation FDM will be while they break the silos and empower FDM programs to do more. We expect to see a lot of movement on the vendor side related to how data is being shared with pilots, maintenance and other stakeholders while protecting its privacy. We will also see more and more advanced capabilities that not only detect exceedances but identify precursors and predict events. Combining this with the current hardware capabilities like wireless data transfer (and even live data transfer) we are also looking at a future where FDM happens on the go and even in real-time.”
In summary, flight data monitoring is doing great things today, and poised to do even more in the future. In this scenario, everybody wins: The airlines, the customers, the regulators, the FDM vendors, and the general public.
By James Careless
