AINsight: Small Operators Can Improve with Big Data
Using data proactively can enhance safety, even for smaller operators.

Last week, we built a “roadmap” to make Part 91/135 operations safer. That blog expanded on a theme first highlighted by NTSB’s Office of Aviation Safety director John DeLisi at the Bombardier Safety Standdown in October. The premise is simple: if the goal is to reduce fatal accidents in business aviation, then why not follow the roadmap laid out by the Part 121 airlines?

The basis of this plan is sound and the results are impressive; by employing a model of an integrated safety system, scheduled airlines in the U.S. have evolved into the world’s safest mode of transportation. To curb fatal accidents in business aviation, the NTSB is now pushing for a wider adoption of safety management systems (SMS) and flight data monitoring (FDM).

Today, every major U.S. airline employs a robust modern safety system. This integrated approach, anchored by SMS, has many components with each element contributing in its own unique way. Fueling this system are a number of underlying programs such as FDM, aviation safety action programs (ASAP), and line operations safety audits (LOSA).

This week, we’ll explore many of the attributes and a few of the nuances of an FDM program. In this limited space, we can’t go deep into the nuts-and-bolts of establishing an FDM program, but will provide some guidance to ease the implementation for the business aviation operator.

By definition, “Flight data monitoring is the systematic, proactive use of digital flight data from routine flight operations to improve aviation safety within an intrinsically non-punitive and just safety culture.” FDM programs assist an operator in identifying, quantifying, assessing, and addressing operational risk. More detailed information on FDM programs is readily available online. In the U.S., FAA advisory circular AC 120-82 is the go-to document. A better, more useful, document for the smaller operator is UK CAA’s CAP 739. This document is the basis for much of the guidance presented in this article. Helicopter operators should check out the HFDM Toolkit

FDM has a long history. The first airlines began dabbling with in-flight data in the 1960s to validate autoland performance. For offshore helicopters supporting oil and gas, FDM has been a contractual requirement for more than a decade. In 2004, a number of corporate operators participated in the FSF/NBAA “C-FOQA” trial. Since then, the acceptance of FDM—in particular by larger Part 91 flight departments—has been growing and is viewed as a best practice.

In the U.S., the term FOQA—flight operational quality assurance—is often used interchangeably with FDM. However, in the eyes of the FAA, only those programs “approved” by the agency are considered a FOQA program. The main motive to pursue FAA approval is the enforcement incentives for an operator and its employees as codified under 14 CFR 13.401(e).  

One of the main objectives of an FDM program is to allow operators to compare their standard operating procedures (SOPs) with those actually achieved in everyday line flights. A feedback loop that should be part of an SMS will allow timely corrective action where safety may be compromised by significant deviation from SOPs. FDM is especially good at identifying new hazards, putting into context a given risk or identifying ineffective risk-mitigation strategies.

For a smaller organization, the challenges with implementing an FDM program relate mainly to a limited infrastructure, mixed fleets, recorder equipage, and a smaller workforce. According to the UK CAA, this is expected, “as with other aspects of operational safety there are issues with the applicability of large operator type systems into smaller organizations.”

None of these issues, however, is insurmountable. FDM programs are scalable and have been successfully implemented in smaller flight ops organizations. As an example, a number of outstanding Part 91/135 operators have operational “FAA-approved” FOQA programs, including Arkansas Children’s Hospital (two Sikorsky S-76Ds), Jack Henry and Associates (four Embraer Phenom 300s), and SevenBar Aviation (six Agusta 109s).

One challenge for a smaller operator is limited resources and infrastructure, especially as it relates to information technology (IT) tasks. Within an FDM program, there are several opportunities to “offload” some duties by outsourcing, primarily as it relates to the technology “grunt” work (IT functions, processing, and validating data). CAP 739 specifically addresses this issue: “Small operators (fewer than 10 aircraft) may find that it is commercially more appropriate to contract analysis services to a third-party.” In the U.S., there are a number of companies that offer these services.

Another challenge relates to small and/or mixed fleets. Statistical variance is an issue with small operators that log fewer flights hours. Such operators are cautioned not to act on small samples of aggregate data, while larger organizations may act on monthly FDM reports. It may be more appropriate for a smaller operator to act on quarterly reports where statistical variance is less of an issue.

One of the primary tenets of an FDM program is confidentiality. For the small operator, it’s imperative that its FDM program has controls in place on the conditions and sensitivity of FDM information. Confidentiality challenges for the small operator include non-scheduled ops, one-off flights and other flight profiles where it may be too easy to identify a particular crew.

The number of progressive business aviation operators adopting FDM is promising. Many of these operators are now sharing data with ASIAS. This is great news. Reading into the signals from the NTSB—and to a lesser extent Canada’s TSB—formal recommendations for FDM programs may be in the works for on-demand charter operators; I’m curious why it took this long. 

Stuart “Kipp” Lau
Contributor - Safety
About the author

Pilot, safety expert, consultant, and aviation journalist Stuart “Kipp” Lau writes about flight safety and airmanship for AIN.

See more by this author