One is struck by two things regarding the “Most Wanted” aviation safety recommendations issued by the National Transportation Safety Board (NTSB): how basic they are, and how the Federal Aviation Administration (FAA) has engaged in seemingly endless foot-dragging in not implementing them. It is all the sadder because there is no doubt that the implementing the “Most Wanted” recommendations would materially enhance safety.

At an NTSB hearing 8 November, the manner in which the recommendations are placed on the “Most Wanted” list came under scrutiny. For example, there have been seven fatal runway incursion accidents in the last 17 years, in which a total of 112 persons were killed. By contrast, pointed out NTSB Member Deborah Hersman, there have been roughly 42,000 highways deaths per year. And of the incursion deaths, 46 occurred in one accident, the take-off crash in August 2006 of the Comair regional jet at Lexington, Kentucky (which was not strictly an incursion based on the circumstances of the crash), yet “no recommendations from this accident are on the ‘Most Wanted’ list,” she said.

“Incursions are always first” to be discussed at the annual “Most Wanted” hearings, and “how we prioritize the list is of concern to me,” Hersman said.

One of the NTSB staffers admitted that the incursion videos – the computerized animations of near collisions – receive such prominence because they are a way of dramatizing the risk (see Figure E).

More to the point, Elaine Weinstein, Director of the NTSB Office of Safety Recommendations and Communications, admitted that recommendations issued by the Board in the first year are not included in the “Most Wanted” list to give the FAA time to respond and, besides, recommendations receive extra attention by the FAA in the first year anyway.

The “Most Wanted” list has three basic attributes:

• The recommendations will impact safety overall.
• The subjects of the recommendations have a high level of public visibility and   interest, and
• The recommendations will benefit from special encouragement.

A February 2007 NTSB letter to the FAA on fatigue, one of the “Most Wanted” subjects, has gone unanswered.

“Does the FAA not have to respond to us?” Herman wondered.

Tom Haueter, the head of aviation accidents, said the Board has received positive responses to the air traffic controller recommendations issued after the Comair crash, and “putting recommendations issued in April on the ‘Most Wanted’ list would unfairly ding the FAA.”

This remark set off a chorus of support.

Joseph Osterman, the NTSB’s Managing Director, said, “They are recent, broad recommendations and positive responses have been received from the FAA.”

Vern Ellingstad, head of the NTSB’s research directorate, said, “I would be reluctant to beat up on the FAA.”

NTSB Chairman Mark Rosenker declared, “We don’t have to beat up on the FAA, and the merits of the issue stand alone.”

Most, if not all of the recommendations have languished on the NTSB “Most Wanted” list for years. Hersman was not persuaded that the list is getting its intended job done, and waiting a year for the FAA to act did not make much sense to her.

“It does not matter how old the recommendation is,” she demurred. “If it’s important, the vintage does not matter. I do not see this list as a punishment to the FAA, but as a tool. We are the ombudsman to the public, not the FAA.”

The FAA is certainly aware of the publicity surrounding the NTSB’s annual “Most Wanted” hearing. For example, the day of the hearing the FAA published a fact sheet on “Safer Flying in Icing Conditions,” one of the “Most Wanted” topics (see Figure A). The FAA proclaimed that an April proposed rule change would improve icing detection and activation of the ice protection system; what the FAA fact sheet did not say is that the notice of proposed rulemaking (NPRM) applies only to transport category aircraft, not corporate or general aviation (GA) planes. The proposed regulation can be interpreted in a way that manufacturers are not required to do anything new (see Air Accident Digest, May 2007, p. 21). As we wrote of that NPRM at the time

“It would seem that a better and safer design and engineering practice would be all-inclusive. The regulation could feature an ‘and’ as opposed to an ‘or’ principle. In this regard, it would seem that the safer practice would be to require 1) identification of conditions conducive to icing, 2) definition of visual clues of icing, and 3) an automatic ice detection system that visually and aurally warns the pilot of icing. Then, after initial activation, the system should be required to operate continuously and should be turned off by the pilots when, and only when, the conditions leading to icing and the visual clues of icing no longer exist.

“This approach would seem more in tune with the spirit and intent of the NTSB recommendations. It is also one well within the state of the art.

“After all these years, the NPRM in its present form is a prescription for inaction.”

This is the net result of the NTSB’s patient approach.

According to Osterman, the “Most Wanted” recommendations are deemed “so powerful” that Congress now wants a report from the Department of Transportation (DOT), the FAA’s governing agency, on the status of the recommendations.

Instead of a report on the status, two other approaches might be considered. One approach would be for Congress to mandate implementation of the NTSB’s “Most Wanted” recommendations (given that they are basic and broad improvements).

The other approach, a more subtle proposition with greater likelihood of success, may be a mandated response to Congress within 90 days of the NTSB’s “Most Wanted” hearing indicating one of the following:

1. Proposed adoption with a tentative timetable for action, and indications of any expected complications or delays (e.g., harmonization with foreign requirements, manufacturer re-design limitations, and so forth).
2. Proposed inaction specifying the detailed “reasons why.”
3. Proposed partial adoption – or adoption by stages or by timetable – with justifications and explanations.

This course might tend to highlight the acceptability (or otherwise) of any cost-benefit commercial undercurrents lying behind any FAA refusal to take action on any NTSB “Most Wanted” recommendation.

Right now, the “Most Wanted” recommendations languish, unexecuted or, as shown above with icing, acted upon in such a way as to be meaningless. Although Chairman Rosenker hailed the FAA and other agencies as “partners,” it is clear that NTSB actions, such as color-coding the recommendations (green for acceptable response and good progress, yellow for acceptable response but slow implementation, and red for unacceptable response), holding the “Most Wanted” hearing and exchanging letters have not had a material effect on the “partners” implementing the recommendations.

Member Hersman is right; a new approach is needed. The “Most Wanted” recommendations are presently coded red across the board for aviation, with the exception of one on fuel tank safety color-coded yellow – a strange designation given that the center fuel tank explosion of TWA 800 in July 1997 started the NTSB’s whole activity in this area, some ten years ago, not to mention that the FAA’s long awaited rulemaking on this issue was supposed to have been issued late this year and has slipped to February of 2008.
What follows is a brief summary of the NTSB deliberations; there are 19 recommendations in six categories (see details at Figure B).

Stop runway incursions and ground collisions of aircraft.

Since 1990 there have been seven fatal accidents involving 112 fatalities, and from May to September of this year there were seven incursion incidents involving 792 crew and passengers (see Figures C & D). Of interest, the FAA convened a “call to action” meeting in August 2007, at which airport safety was reviewed and the deployment of airport signage and markings was accelerated.

According to NTSB Vice Chairman Robert Sumwalt, “We were putting a lot of our emphasis on airport technology; we should be prodding the FAA to do more on cockpit-based technologies.”

He was thinking of cockpit systems like TCAS (Traffic Collision Avoidance System) and TAWS (Terrain Awareness & Warning System), which have significantly blunted the hazard of mid-air collisions and controlled flight into terrain (CFIT).

Member Kitty Higgins agreed with Sumwalt on the need for an in-cockpit system. “I’m really troubled that the FAA is moving ahead with on-airport systems, but not in-cockpit,” she lamented.

She pointed out that the FAA issued an NPRM 5 October on ADS-B (Automatic Dependent Surveillance, Broadcast) that calls for airplanes to broadcast a wealth of data for air traffic controllers (ADS-B Out), but the proposed rule defers ADS-B In technology for the next few years (see Aviation Safety & Security Digest). ADS-B In would enable an aircrew to receive direct warnings of a threatened incursion, but Higgins notes that neither the new A380 nor the forthcoming B787 will have ADS-B In installed.

“The FAA is issuing press releases congratulating themselves on reducing incursions, but they’re not pushing for in-cockpit technology earlier than 2020,” Higgins said of the 5 October NPRM.

Member Steven Chealander, a former airline pilot, agreed, declaring flatly, “We need layers of protection right up to the cockpit.” He noted that runway status lights have been tested for years at Dallas-Ft. Worth International Airport, but they have not been mandated by the FAA for wholesale installation, even though they would provide a direct warning to the flight crew. It should also be noted that Honeywell’s RAAS (Runway Awareness Advisory System), now in the market, would provide both take-off and landing alerts to the flight crew regarding the runway, distance remaining, and so forth. While RAAS does not warn of the presence of another aircraft, it does enhance the aircrew’s situational awareness and goes so far as to warn against taking off or landing on a taxiway, which has occurred more than once.

The Board voted 5-0 to retain this item on the “Most Wanted” list and to color-code the area red to indicate the FAA’s non-responsiveness.

FIGURE C

FIGURE E

Recreation of runway incursion at San Francisco, 26 May 2007. Source: NTSB

Eliminate flammable fuel/air vapors in transport category aircraft.

The NTSB’s Bob Swaim pointed out that the recommendation originally stemmed from the 1996 loss of TWA flight 800 to a center wing tank explosion, but there have been other accidents involving dangerous fuel/air vapors. For instance, the center wing tank exploded on a China Airways B737 at Okinawa this past September, when the spilled fuel from other tanks caused a ground fire (see Air Accident Digest, September 2007, P. 8). Resultant heating caused the center wing tank to explode (see Figure F).

Moreover, according to Swaim, the FAA’s effort to reduce in-tank ignition sources has proven to be inadequate. Over 120 airworthiness directives (ADs) have been issued to address issues such as chafed wires, potential arcing, and so forth. One such AD was incorporated in a Transmile B727, yet the airplane experienced a short circuit in the fuel tank anyway in May 2006 (see Figure G).

The FAA has drafted a final rule addressing the reduction of flammable vapors, but it has been delayed repeatedly. The rule would establish new design standards for fuel systems and would also necessitate retrofitting of flammability reduction systems (e.g., inerting) on existing airplanes. The rule would affect more than 3,200 U.S. airliners and set the standard for safer fuel systems worldwide.

The rule was to have been issued in September, but has been pushed back to February 2008. Apparently, the cost-benefit calculation in the draft final rule poses a hurdle. According to an FAA letter of 23 October 2007, “We continue to work on the details of the rulemaking action to determine how it can be imposed consistent with our mandate to demonstrate that the additional benefits of this requirement are sufficient to justify its costs.”

As Swaim said, “Staff is concerned that the cost is overstated and the benefit understated.”

For example, the inerting system on which the cost was calculated features air separation modules that take bleed air from the engines and separate out the oxygen, leaving nitrogen-enriched air to be routed to the fuel tanks.

A newer system, described elsewhere on this website, features a catalyst-based approach that is simpler, cheaper, weighs less, and provides inerting for the descent phase, which the separation modules do not do.

Given that the original recommendation is now 11 years old, Chairman Rosenker asked why this issue should not be downgraded from yellow to red. “I am generally disappointed and disturbed” at the delay, he said.

“It’s like groundhog day, we just keep dealing with it,” groused Member Hersman, referring to the movie Groundhog Day, in which the protagonist keeps living the same day over again until he gets it right.

“What will we be discussing when this issue comes up next year? Hope is not a strategy,” she said.
Swaim replied, “Staff recommends keeping it on the list as yellow. However, we’re greatly concerned about the slippage in time and the mis-stated cost-benefit calculation.”

The Board voted 4-1 against color-coding this item red, Member Hersman being the sole exception.

FIGURE F

A fire fed by fuel spilled from the wing tanks heated the center wing tank, causing the fuel/air vapors therein to explode. Source: NTSB

FIGURE G

Reduce dangers to aircraft flying in icing conditions.

The NTSB’s Dan Bower noted that the board has issued 18 recommendations and one safety alert in the past two years (see Figures J, K, L). The problem, he said, is that these recommendations address aircraft in icing on a case-by-case basis, whereas the unrequited “Most Wanted” recommendations address all airplanes.

The FAA issued a final rule in August, Bower said, but the action remains to certify all airplanes – not just new designs – for icing.

“We have to look at why this is taking so long before we send out all these recommendations” for specific airplanes, said Hersman.

The Board voted 5-0 to leave this area color-coded red.

FIGURE J

FIGURE K

FIGURE L

Path of the upset. The airplane lost altitude as it was flying into rising mountains, increasing the hazard of a sudden descent due to icing. Source: NTSB

Improve aviation audio and data recorders and require cockpit video recorders. The FAA issued an NPRM in February 2005 with a possible final rule in 2008. However, the proposed rule is only partially responsive to the NTSB’s recommendations (see Figure M). It requires retrofit of a 2-hour cockpit voice recorder (CVR) capability, but it only requires an independent power supply on new aircraft, and there is no mandatory requirement for dual redundant CVR’s and flight data recorders (FDR’s), nor is there a requirement for cockpit video recording.

In addition to which, there is “no tangible FAA activity” regarding additional FDR parameters for the Boeing 737, which were deemed necessary because of the older model B737s history of rudder reversals.

All of the proposed compliance dates in the recommendations have long since passed.

The Board voted 5-0 to leave the issue color-coded red.

FIGURE M

Improve crew resource management (CRM).

Since March 1998 all scheduled airlines (Part 121) are required to train their aircrews in and practice CRM.

“We need an equivalent level of safety for a charter operation,” declared Vice Chairman Sumwalt.

An NPRM filling this need for Part 135 is slated to be issued in 2007; obviously, time is running out.  According to an NTSB staffer, the proposed ruling will not require CRM as comprehensive for Part 135 operators as it is for Part 121 operators. As such, the proposed ruling may not fully address the factors in three crew-involved accidents in 2004 and 2005.

Member Chealander declared, “We find CRM issues all the time in our accident investigations. We need to push this hard; it’s vitally important to public safety.”

The Board voted 5-0 to keep this issue on the “Most Wanted” list, color-coded red for noncompliance.

Accidents involving on-demand charter operations, and why the NTSB thinks crew resource management is critical for these operations. Source: NTSB

Reduce accidents and incidents caused by human fatigue. “We have four recommendations in the aviation mode, basically closing the loophole allowing crews to exceed hours of service for repositioning [non-revenue] flights,” noted NTSB staffer Dr. Jana Price.

There are no duty time limits for aircraft maintainers, and the FAA has no rulemaking proposed (see Figure N).

Member Hersman said, “We already have pilots and mechanics on the ‘Most Wanted’ list, and I’m asking that we add controllers.”

The problem, as the staff saw it, is that recommendations addressing controller fatigue were issued in April, well within the last year and therefore not appropriate for inclusion in the “Most Wanted” list.

The Board vote was 4-1  to include controllers; Member Chealander was opposed.

All five members voted to keep the issue on the “Most Wanted” list, color-coded red to indicate the FAA’s non-responsiveness. It should be mentioned that air traffic controllers are FAA employees, and therefore the agency can set fatigue limits, as it were, without going through the extended rulemaking process. An opportunity has been lost here to have the FAA set the example with its own workforce.

FIGURE N

The NTSB’s letter to the FAA went unanswered, raising questions about how responsive the agency will be to concerns about air traffic controller fatigue. Source: NTSB

Recall that Rep. Jerry Costello (D – Il.), chairman of the House Aviation Subcommittee vowed at a 6 June hearing that the FAA will not be allowed to lapse into regulatory torpor and will face continual pressure to act on the “Most Wanted” safety recommendations (see Air Accident Digest, June 2007, P. 11). It is not clear what that continual pressure constitutes, but a Congressionally-mandated response within 90 days along the three-part lines suggested above might well break the FAA’s obfuscations and the dilatory logjam.