As the world rotorcraft community looks at new ways to attack the helicopter accident rate, the FAA and industry partners are spooling up the Aviation Safety Information Analysis and Sharing System (ASIAS) for rotorcraft. Rotorcraft ASIAS began in late 2014 and currently 30 operators, 13 of them from law enforcement, are participating in the program, in which operator flight recorder data gets confidentially downloaded, converted, validated, checked for consistency, de-identified, then uploaded into a central database currently administered by the Helicopter Association International (HAI). Then standard measures, metrics, data analysis and queries run against the data, and format processes can be run on events to generate studies and spot safety trends. The FAA does not get access to individual operator data.
Airlines have successfully used the ASIAS tool and a similar system has been developed under the Partnership to Enhance General Aviation Safety, Accessibility and Sustainability (Pegasas) for general aviation.
Cliff Johnson, FAA research lead for rotorcraft ASIAS, said the program promises to be particularly useful in light of the comparatively static rate of fatal helicopter accidents. “Flight data trend monitoring can be used as a proactive tool, not just after an accident or incident has occurred. You can use it to prevent one from happening in the first place. It is about what you do with the data and how you use the data. It’s about how we help you improve safety in your daily operations. We solicit participating operators to provide data through the research effort to help them improve safety. We can get smarter about data, how we bring it in, and develop some new safety analysis tools.” Johnson assured operators that the “FAA doesn’t touch any of the data” and that the program is based on “voluntary, non-punitive data sharing.” He said operators would have access not only to their own data but also would be able to compare it to that from similar operators flying similar aircraft to see how they measure up.
Keith Cianfrani, HAI flight data monitoring specialist, reports that the program is going well among operators who have already signed up for ASIAS. Cianfrani calls it a “Fitbit” for the aircraft and a particular flight. “Participants look at ASIAS and say it could be part of an SMS [safety management system] program. It fits a cog in the wheel to enhance safety.”
Johnson said the ASIAS team is beginning to develop performance models and metrics around parameters involving the most common causes of rotorcraft accidents: loss of control, unstabilized approaches and helipad overruns. “In your typical FDR [flight data recorder] report you are not going to get every single parameter on the helicopter but you can synthesize the parameters. Almost one-third of accidents involve some sort of loss of control, whether because of autorotation, settling with power or some type of dynamic upset,” Johnson said. “The modeling allows us to mine for the data and determine what differentiates a good autorotation from a bad one. All these are physics-based models and can help us stay out of these events if used as a predictive tool. The same with vortex ring state [settling with power]. We can model rotor downwash to see where we get into the state. We also looked at dynamic rollover, tipover accidents in high crosswinds, and we would like to be able to calibrate that data.”
Johnson said the ASIAS team is also looking at cockpit audio and video data. “Some of the recorders are not going to record everything. But if the recorder fails you can look at the video data and using image processing techniques you can translate the image into digital data, such as flight instruments, and also look at engine and background noise, including alarms. You can use an algorithm to read the gauges and image processing to gather the data, including that from engine torque and attitude indicators. You can also analyze the spectrum and frequency of audio data.”
The FAA is currently flying data analysis flight-test missions in its S-76A to define and validate some of the ASIAS techniques that have been devised. The S-76 is outfitted with three separate flight data recorders. It is also being used to test low-cost mobile devices such as the general aviation airborne recording device, developed by the FAA and Mitre as a low-cost flight data management device alternative.
Johnson said average participating operators can use ASIAS to “address risks in their own operations.” He gave the example of an operator using ASIAS data to plot flight trend proximity to obstacles and weather as an example. But globally, he acknowledged that the program will need more participants and more data to generate meaningful information. “To do this and really make it accurate you need large amounts of data from various operations. If you just have data from onesies and twosies here and there, patterns really don’t show up.”