How accurate are the NTSB’s published general aviation accident rate statistics and related data? Not accurate enough, according to the Safety Board. Still not satisfied it is receiving realistic general aviation activity data–the basis of an accurate assessment of GA safety–the NTSB is once again urging the FAA to develop methods for obtaining better information about hours flown, fleet size, the purpose of flights and other data for GA operations, including fractional aircraft operations under FAR Part 91 Subpart K.
In recommendations sent to the FAA last month the NTSB pointed to the alleged failure of the agency’s General Aviation, Air Taxi Activity and Avionics (GAATA) Survey–currently the only source available for the Safety Board to determine accident rates, correlation factors and other safety statistics–to assess activity precisely. The NTSB also noted that the FAA has made little headway on improvements to data collection efforts recommended over the past five years, recommendations still open and reiterated last month.
One of these 2001 recommendations was to “Identify and implement methods independent” of the GAATA that can be used to check the accuracy of GA flight hour estimates. In its response, the FAA stated that it “agrees that improvements should be made to the GAATA Survey and noted that as of January 2002, it had already spent a year working to identify and prioritize specific improvements as part of the Safer Skies initiative. A published implementation plan was promised “pending resource commitments” for the process. At press time, that plan had yet to be published.
The other recommendation made in 2001 was to improve the accuracy of the FAA aircraft registry, used to determine the size and makeup of the general aviation fleet. The FAA responded that it was aware of problems, and in January 2002 stated that it was establishing a team (of lawyers) “charged with improving the currency of its civil aircraft registry by reviewing records of aircraft noting a sale, pending registration or undeliverable address.” This five-year-old recommendation also has an “open” status.
1999 Change Adds Complication
Further diluting the accuracy of activity data, in the NTSB’s opinion, was a major change the FAA made in 1999. Historically, the GAATA Survey had excluded all aircraft reported as destroyed, registered to dealers with a sale reported or registration pending, with known inaccurate owner address information or with missing make or model information. However, in 1999, the FAA expanded the assumed population of active aircraft to include all aircraft registrations that had become invalid within the last 10 years due to inaccurate address information, a reported sale or a pending registration.
When the FAA expanded the aircraft population, the estimated number of hours increased and the accident rate for 1999 was the lowest on record. However, the NTSB said the rate decrease is “questionable given that the definition of the active aircraft population was substantially different [from that used in] previous years.”
Lacking confidence in the numbers in the GAATA Survey and tired of waiting for the FAA to finish implementing the 2001 recommendation, the NTSB itself attempted to verify independently GA activity estimates by comparing FAA aircraft registry and annual flight-hour estimates with other indicators of flight activity, such as the number of GA accidents, the active pilot population and aviation fuel consumption for the years 1985 through 2002.
“Unlike the strong relationships observed between the number of active pilots, aviation fuel consumption and accidents, no significant correlations were observed between the annual GA flight-hour estimates reported in the GAATA Survey and the number of active pilots, barrels of aviation gasoline consumed or GA accidents,” the NTSB concluded. “These results strongly suggest that activity data derived from the GAATA Survey have not accurately portrayed the changes in GA activity during the studied period.”
Therefore, the NTSB in last month’s group of recommendations requested the FAA “develop, validate and document an unbiased method for generating and revising activity estimates based on Part 135 and Part 91, Subpart K, operator surveys or reporting.”
In its recommendations last month the NTSB also asked the FAA to mandate annual reporting requirements for Subpart K operators, echoing an August 2003 recommendation to require annual reports of activity data from Part 135 air-taxi operations. Nearly two years since that recommendation, the FAA is still reviewing it.
NATA: NTSB ‘Misses the Mark’
While the National Air Transportation Association (NATA) agrees there is a need for more accurate general aviation activity statistics, the FBO and air taxi trade group said the NTSB recommendation to improve data collection “misses the mark.” NATA contends that there are additional data resources available to the FAA that provide “substantial information with no additional burden on the industry.”
For example, the GAATA survey relies upon information from the aircraft registry to count the number of active Part 135 aircraft. “A more accurate, and readily available, figure is obtainable from the [agency’s] operations specification subsystem and includes the number of Part 135 operators and the number and type of aircraft operated.”
Additionally, “The NTSB does not yet appear to have a firm grasp on the fundamental elements that distinguish Part 135 on-demand operators and Part 91 Subpart K fractional ownership operations from Part 121 airlines,” NATA asserted. “The Board has requested that Part 135 passenger carriers report available seat miles. That type of data, while a key measure of Part 121 airlines, is irrelevant to Part 135 operators because the entire aircraft is most often engaged by just a single entity,” the association said.