Flight Operations

Flight Operations




Distracted

by Linda Werfelman. This article was originally published in the April 2009 issue of AeroSafety World magazine and is reprinted with the permission of the Flight Safety Foundation.

Five television news-gathering helicopters were manoeuvring to cover a police chase in Phoenix on July 27, 2007, when two of the aircraft—both Eurocopter AS 350B2s—collided over a downtown park as their pilot-reporters were describing the events occurring on the ground. The crash killed the two pilot-reporters and two news photographers and destroyed both helicopters. 

Mid-air collision between two helicopters

The pilots of two TV news helicopters
were busywith news-reporting duties
in the seconds before their
midair collision in Phoenix.

The U.S. National Transportation Safety Board (NTSB), in its final report on the accident, said that the probable cause was the failure of both pilots to see and avoid the other helicopter, and “contributing to this failure was the pilots’ responsibility to perform reporting and visual tracking duties to support their station’s ENG [electronic news gathering] operation.”

The report identified as a contributing factor “the lack of formal procedures for Phoenix-area ENG pilots to follow regarding the conduct of these operations.”

Visual meteorological conditions prevailed when the midair collision occurred at 12:46 local time, about 23 min after a police helicopter contacted air traffic control (ATC) to join the pursuit by police on the ground of a suspect accused of stealing a pickup truck, backing it into a police vehicle and then fleeing in the truck. Over the next 22 min, pilots of the five news helicopters also checked in with ATC and headed for the area of the police chase.

In accordance with informal procedures, the six helicopter pilots shared an air-to-air radio frequency to report their positions and their intentions. The accident helicopters—from Channel 3 and Channel 15—had audio video recording systems, and the information recorded by those systems was analyzed in the accident investigation. At 12:38, the Channel 15 pilot was recorded telling the other pilots that he was flying at 2 200 ft, and the Channel 3 pilot said that his helicopter was at 2 000 ft.

“According to the Channel 3 and 15 audio recordings, about 12:41:02, the Channel 15 pilot stated, ‘I’ll just kinda park it right here.’ About 12:41:18, the Channel 3 pilot broadcast, “OK, I’m gonna move.’ Between about 12:41:22 and about 12:41:26, the Channel 15 pilot stated, ‘where’s three?’ … ‘like how far?’ … and ‘oh jeez.’ The Channel 15 pilot then transmitted, ‘Three. I’m right over you.

Pilots of the two TV news helicopters had traded information about the positions of their aircraft several times, but investigators say the last exchange came about four minutes before their midair collision. 

Fifteen’s on top of you.’ Afterward, the Channel 3 pilot questioned which helicopter Channel 15 was over, to which the Channel 15 pilot responded, ‘I’m over the top of you.’ About 12:41:34, the Channel 3 pilot indicated that he was operating at 2 000 ft. About 12:42:25, the Channel 3 pilot stated to the Channel 15 pilot, ‘OK. … I got you in sight,’ to which the Channel 15 pilot responded, about three seconds later, ‘got you as well.’”

These comments—about 4 min before the collision—were the last in which the two pilots coordinated their helicopters’ positions or intentions. The video recordings from the helicopters showed that, during those 4 min, both helicopters continued to change position.

The report said that the suspect stopped the stolen vehicle about 12:46:05, and in a broadcast recording that began at 12:45:43, the Channel 3 pilot said, “Looks like he [the suspect] is starting to run. … Looks like he’s gonna try and take another vehicle … looks like they’ve got him blocked in there, but he did get ... ” The Channel 3 report then ended “suddenly, with an unintelligible word,” the NTSB said.

The Channel 15 pilot, in a live broadcast that began at 12:46:03, said, “He [the suspect] has stopped … now it’s a foot chase. Now he’s in another vehicle … doors open police … oh jee.” That report also ended suddenly, the NTSB said, and audio recordings from both helicopters indicated that the midair collision occurred about 12:46:18.

Both helicopters plunged to the ground in a city park, and the pilot of a third ENG helicopter told ATC there had been a midair collision.

The Channel 3 pilot, who in September 2006 reported having 13 579 flight hours, received a commercial pilot certificate with a rotorcraft-helicopter rating on August 24, 1987. He also held a certified flight instructor certificate with a rotorcraft-helicopter rating and a second-class medical certificate. He was a backup pilot under contract to Channel 3 and a part-time employee; the station’s chief pilot said that the accident pilot had flown 79 flights and 124 flight hours for the station between January 2 and July 5, 2007. The accident pilot also worked full time for Westcor Aviation in Scottsdale, Arizona, as director of operations and a charter pilot; the operator said the accident pilot had flown 88 hr for the company in 2007.

The Channel 15 pilot, who had 8 006 flight hours—all in helicopters, including 907 hours in AS 350B2s—received a commercial pilot certificate with a rotorcraft-helicopter rating on December 7, 1990. He held a second-class medical certificate with a waiver for defective colour vision; the NTSB report said that the deficiency was not a factor in the accident. The pilot was hired in October 2005 by U.S. Helicopters, which had a contract to provide helicopter service to Channel 15. He flew an average of 45 hr per month for the station and did no other flying for U.S. Helicopters, the report said.

The Channel 3 helicopter had an ENG monitor near the instrument panel that displayed four scenes simultaneously: the station’s current broadcast, the video being recorded by the helicopter’s photographer and two other scenes selected by the pilot-reporter. The Channel 15 helicopter had a similar monitor that displayed one scene at a time.

The Channel 3 helicopter was equipped with an L-3 Communications SkyWatch SKY497 traffic advisory system that provided aural traffic warnings via the pilot’s headset, displayed traffic on a Garmin GNS 430 navigation unit and provided 20- to 30-s warnings of aircraft that were on a collision path.

“The system issued an aural alert when aircraft entered a cylinder of airspace surrounding the pilot’s aircraft that had a horizontal radius of … 1,216 ft [371 m] and a height of plus or minus 600 ft [183 m],” the report said. Manufacturer’s guidance said that after hearing an alert, the pilot should look for the traffic and comply with right-of-way procedures. The guidance material also noted that an alert is generated only when the collision threat is first detected and that it is possible for the alert to be “inhibited”.

Channel 3’s chief pilot told investigators that the system had been functioning when he flew the helicopter earlier on the day of the accident. He also said that, in situations in which “a lot of traffic (was) in close,” the volume of the aural alert was turned down to ensure that the pilot could hear radio transmissions on the communications frequency.

Channel 15’s helicopter had no on-board traffic advisory system, the report said.

In addition to their use of the shared air-to-air frequency and their scans of the TV display screens in the cockpit, the pilot-reporters monitored the Phoenix air traffic control tower frequency on another radio, communicated with their station news departments on a third radio and talked with their photographers over an intercom, the report said.

“Adequate” communication
Radar data showed that the Channel 15 helicopter had been between 2 000 and 2 200 ft and entered a climbing right turn in the seconds before the crash; the last radar return showed the helicopter at 2 300 ft. At the same time, the Channel 3 helicopter, which had been at 2 000 ft, turned right; the last radar return showed the helicopter at 2 100 ft.

As part of the investigation, NTSB representatives met with Phoenix ENG helicopter pilots, who said that communication between the accident pilots had been “adequate” during the police chase. They also noted that, at the time of the accident, all operators except one used pilot-reporters to fly their aircraft; the exception was a station that employed a reporter-photographer.

However, the pilots told investigators that they sometimes lost sight of other helicopters because the aircraft paint schemes “tended to blend in with the desert landscape and vegetation.” They recommended the use of high visibility paint schemes for main rotors and tail rotors, and light-emitting diode (LED) anti-collision lights to improve helicopter conspicuity. Neither accident helicopter had these features.

The chief pilot for Channel 3 told investigators that, since the accident, pilots of the ENG helicopters have had “a lot more” air-to-air communication, describing the location of their helicopters and acknowledging the positions of others.

“He also stated that, in a static situation, such as a building fire, no helicopters would change position until all of the pilots responded and that, in a dynamic situation, such as a car chase, the pilots would constantly communicate with one another and confirm each other’s positions,” the report said. “He further indicated that the pilots were providing more distance between each other’s helicopters and were asking the photographers more often to check clearances (separation) with other helicopters.”

The two accident pilots were experienced in helicopter operations in general and ENG operations in AS 350B2s in the Phoenix area in particular, the report said. Both also were experienced in simultaneously flying their helicopters and reporting.

“Many of the tasks that the pilots were performing during the accident flight—such as flying the helicopter, operating the radios and initiating communications—were well-learned skills that would have been performed without much cognitive or physical effort,” the report said. “However the two helicopters collided without either pilot detecting the impending hazard. Thus, even for experienced pilots, the ability to shift attention among competing task demands may break down under high workload conditions and can lead to a narrowing of attention on a specific task.”

A review of audio recordings showed that the accident pilots did not use the air-to-air frequency to report their positions as often as the ENG pilots participating in the post-accident interview had thought, the report said.

“It is difficult to determine the extent that the Channel 3 and [Channel] 15 pilots’ reporting duties contributed to the breakdown in each pilot’s awareness of the other helicopter,” the report said. “The additional tasks of directly observing activities on the ground and providing narration could have affected the pilots’ ability to maintain their helicopter’s position or track the other helicopter’s positions. From about 12:45:43 (Channel 3) and about 12:46:03 (Channel 15) to the time of the collision, the pilots were continuously reporting the events as they unfolded, which narrowed the pilots’ attention to the ground and away from other tasks, such as maintaining the helicopters’ stated position and altitude and scanning the area for potential collision hazards.

“Even with the limited evidence to determine the extent that the pilots’ ENG-related duties affected their ability to see and avoid the other helicopter, the circumstances of this accident demonstrated that a failure to see and avoid occurred about the time that a critical event of interest to the ENG operations (the carjacking) was taking place on the ground. … It is critical for ENG pilots to be vigilant of other aircraft during close-in operations and not to divert their attention to a non-flying-related task or event.”

The NTSB also cited a report filed with the U.S. National Aeronautics and Space Administration Aviation Safety Reporting System (ASRS)—one of 18 reports of near-midair collisions involving ENG helicopters—in which the pilot described how he inadvertently allowed his helicopter to descend toward a police helicopter because his “hectic” workload had distracted him from altitude awareness.

“The midair collision in [Phoenix] and the near-midair collisions described in … ASRS reports demonstrate the hazards involved in conducting ENG operations with multiple aircraft nearby,” the report said. “The safety board concludes that the Channel 3 and 15 pilots’ reporting and visual tracking duties immediately before the collision likely precluded them from recognizing the proximity of their helicopters at that time.”

Aftermath
After the accident, both Channel 3 and Channel 15 modified their flight operations. The Channel 3 news helicopter is now staffed by two pilots—one to handle flying and the second to handle news reporting. The Channel 15 helicopter pilot no longer has reporting duties; the helicopter carries a photographer to obtain video.

In February, the Helicopter Association International (HAI) approved a new Broadcast Aviation Safety Manual developed along the lines of many of the NTSB safety recommendations issued as a result of the accident investigation.

The 10 safety recommendations included a call for the U.S. Federal Aviation Administration (FAA) to require ENG operators to assign reporting duties to “someone other than the flying pilot, unless it can be determined that the pilot’s workload would remain manageable under all conditions,” and to require high-visibility blade paint schemes and high-visibility anti-collision lights on ENG aircraft.

Other recommendations said the FAA should develop standards for helicopter cockpit electronic traffic advisory systems to notify pilots of the presence of nearby aircraft, and require that the systems be installed in ENG aircraft; host annual ENG helicopter conferences to discuss relevant issues, and, based on those discussions, develop agreements specifying minimum horizontal and vertical aircraft separation requirements; and incorporate information from the HAI safety manual into an FAA advisory circular.

Other recommendations—superseding similar recommendations issued in 2003—call for requiring the installation of a “crash-protected flight recorder system” on new and existing turbine-powered, non-experimental, non-restricted-category aircraft that are not equipped with a flight data recorder and cockpit voice recorder and that are operated under U.S. Federal Aviation Regulations Parts 91, 121 or 135. The recorder should record cockpit audio, if a cockpit voice recorder has not already been installed, as well as “a view of the cockpit environment to include as much of the outside view as possible” and flight data, the NTSB said.

Since 2004, the NTSB has included similar recommendations on its “most wanted” list of transportation safety improvements.

This article is based on U.S. National Transportation Safety Board (NTSB) Accident Report NTSB/AAR-09/02, “Midair Collision of Electronic News Gathering Helicopter KTVK-TV, Eurocopter AS350B2, N613TV, and U.S. Helicopters Inc., Eurocopter AS350B2, N215TV, Phoenix, Arizona, July 27, 2007.” Jan. 28, 2009. 

Focus on CRM

As mentioned in ASL 1/2012, we are currently running a series of selected articles dedicated to crew resource management (CRM) awareness. Our second feature article on CRM is entitled “CRM Assessment: A Pilot’s Perspective” and was written by Captain David McKenney, Vice Chairman (Human Factors) for the IFALPA Human Performance Committee.

CRM Assessment: A Pilot’s Perspective

by Captain David McKenney, Vice Chairman (Human Factors) for the IFALPA Human Performance Committee. Originally published in ICAO Training Report, Vol. 1, No. 1 – July/August 2011. Reprinted with permission.

The International Federation of Air Line Pilots’ Associations (IFALPA) has long recognized that relying solely on a pilot’s technical knowledge and skills is not sufficient to safely operate complex aircraft in today’s flying environment. Crew Resource Management (CRM) was developed over 30 years ago to help address this issue.

As Captain David McKenney of the International Federation of Air Line Pilots’ Associations highlights, IFALPA supports CRM as a training program and as an adjunct to traditional technical training approaches. The pilot’s federation also suggests that industry and regulators should focus their efforts on producing guidance that encourages more effective CRM training approaches and on developing tools to measure CRM results across the entire culture within the airline.

Originally portrayed primarily as a conflict resolution skill, CRM has evolved today to define a set of skills that supports pilot technical and decision-making flying capabilities. It does this by providing them with the cognitive and interpersonal skills needed to address human error by managing resources within an organized operational system.

CRM is normally defined as a management system which makes optimum use of all available resources, including equipment, procedures and people, to promote safety and enhance the efficiency of flight operations. IFALPA believes CRM can improve the proficiency and competency of individual pilots and flight crews as a whole, especially when it is implemented as an error management strategy.

Flight crews need specific skills and strategies to assist them in coping with the dynamic demands of piloting and in reducing errors. IFALPA supports integrating CRM into flight crewmember training as a tool to minimize the consequences of human error and to improve flight crew performance.

Industry recognizes CRM as a “best practice” when fully integrated into initial licensing and recurrent training programmes, including Multi-Crew Pilot Licensing (MPL) and Advanced Qualification Programmes (i.e., AQP, ATQP).

When first introduced, a cornerstone in the acceptance for CRM training was the assurance that it would not include evaluation. Much of the value and strength of CRM is based on this principle. IFALPA believes the introduction of any checking or jeopardy assessment process has the potential to destroy such benefits and negatively affect safety. To understand the issues, one needs to review what CRM training is and how it is implemented.

Threat and error management
Fifth generation CRM places a good deal of emphasis on behavioural trends and Threat and Error Management (TEM). One of the underlying principles of this fifth generation approach to CRM is the premise that human error is inevitable and should be normalized within the system (Helmreich, 1997).

“Flight crews need specific skills and strategies to assist them in coping with the dynamic demands of piloting and in reducing errors. IFALPA supports integrating CRM into flight crewmember training as a tool to minimize the consequences of human error and to improve flight crew performance.”

Pilots should be taught the limitations of human performance and be trained to develop skills to detect and manage error. For this error management approach to succeed in any organization, the organization itself must first recognize and communicate their formal understanding that errors will occur and also adopt and strongly reinforce a nonpunitive approach to error reporting.

CRM as a culture
CRM is not just aircrew-centric; it does not start and stop with the captain or crew. Effective CRM must be embedded within the cockpit and safety culture of the airline while addressing airline specific items (i.e., carrier-specific operations and procedures) and needs to be practiced and accepted at all levels of the organization to positively affect operational safety.

To be truly effective, CRM must be embedded in the airline’s Safety Management System (SMS), which should provide for open advocacy and feedback. Each carrier must therefore develop a CRM program that is tailored to their specific culture and pilot demographics and understand that no single CRM program or approach is suitable for all operations and all airlines.

This lack of “one-size-fits-all” characteristic has made it difficult for the industry to adopt a single and universal CRM program with standardized terms, definitions and application methodology.

Integrating CRM into flight crewmember training
Recognizing that safety depends on the coordination of key people in the entire system and not just on the actions of pilots, CRM training should be implemented by carrier flight operations personnel who possess pertinent knowledge of the culture, policies, procedures and training of that particular air carrier. Evidence shows that a joint CRM course for flight crews, cabin crews, and dispatchers can improve the level of understanding and cooperation across the entire team.

Air carriers develop CRM programs that promote the integration of practical flight management skills with traditional technical skills. CRM awareness and error management training is most beneficial when the training curriculum is individualized for each pilot, tailored to each airline’s unique culture and includes the added realism of Line Oriented Flight Training (LOFT).

Lack of regulatory guidance
While CRM has evolved over the past 30 years, regulatory measures have not kept up. A lack of standardized CRM terms, definitions, application methodologies and guidance is continuing to impede CRM standardization across the industry.

Different CRM application methodologies relating to awareness training and error management strategies are currently used. For many years, the industry provided guidance material that centred on the benefits of flight crewmembers’ awareness of CRM, often referred to as “soft skills”. The biggest benefits to teaching soft skills were the resulting improvements in attitudes, perceptions and teamwork. Although training in the soft skills is useful to pilots as recognition and perception training, it only represents one of the issues confronting flight crews.

The error management methodology uses standardized procedures, flight management skills and specific error prevention techniques for the management of safe flight by flight crewmembers. Currently, there is no governing regulatory documentation for error-management techniques, although IFALPA strongly supports training in this area. As a result, CRM courses among airlines vary widely, some only teaching awareness training while others stress both awareness training and threat and error management.

Subjective evaluation criteria
IFALPA stands firmly against any CRM evaluations for flight crewmembers, individually or as a crew in any jeopardy event and most especially when the evaluations in question utilize only subjective criteria.

Little, if any, qualitative evaluation criteria exist for CRM and there is no universally accepted methodology for identifying unsatisfactory pilot CRM performance. Regulators have allowed operators with different corporate cultures a great deal of flexibility in introducing CRM training, resulting in a wide spectrum of quality, quantity and effectiveness levels in CRM courses across the industry.

Vague terms such as “Captaincy”, “Airmanship”, “Followership”, and “Synergy” lack any formal or recognized definition within the CRM concept. These worthwhile attributes are presently beyond the ability of any expert to evaluate objectively, much less a check airman unskilled in the meaning of these terms.

Specifically, evaluation of the effectiveness of non-technical training skills is very subjective and extremely variable. There is no universally accepted definition of the CRM concept or category of CRM terms within the air carrier industry. IFALPA is concerned because flight crew CRM evaluators lack adequate standards and guidance material.

CRM evaluation exposes a crewmember’s certificate and career to unsubstantiated jeopardy when no objective industry definitions or standards of CRM skills exist. In one case, an air carrier terminated pilots based on CRM performance alone, although CRM has not matured sufficiently enough for evaluators to effectively evaluate a flight crewmember’s performance.

Industry experience has shown that it is difficult to train and calibrate instructors/evaluators to successfully identify markers that would lead to an overall “grade” or “consistent grading”. This is in part due to these markers not being adequately defined and therefore not observable.

Unintended consequences of evaluating CRM
There has been no demonstrated case that improved safety results from introducing jeopardy assessment/checking of CRM. In fact some CRM experts within the aviation industry believe the unintended consequences of evaluating CRM could actually reduce current safety margins. IFALPA agrees in its published IFALPA Policy on CRM, which states in IFALPA Annex 6, Part I that:

IFALPA believes that to introduce jeopardy assessment or checking of CRM at this point would fundamentally change the facilitator/instructor and flight crew relationship and potentially block or reverse the many benefits to be gained from CRM training, including the possibility of having a negative impact on safety. Jeopardy assessment or checking CRM may result in crews producing acceptable CRM behaviour in the simulator but have little real impact on the safety culture of the airline.

For CRM training to genuinely impact the safety culture in aviation, CRM must be wholeheartedly embraced by pilots without the threat of any punitive action. To this end, IFALPA supports open feedback and discussion between facilitators/instructors and flight crew on CRM topics. This feedback should however be non-numerical (e.g., Enhanced – Standard – Detracted) and focus on reinforcing good skills and discussing areas of improvement. IFALPA recognizes that a high level of trust and openness must be present for such discussions to be effective.

Besides IFALPA, individual pilots are also concerned about the negative implications of “evaluating” CRM skills. Evaluations can lead to a mistrust of the program, especially if the evaluation of these skills is done in an arbitrary and capricious manner. If we evaluated CRM today, it would be done in an “opinion-oriented” fashion. This could lead to evaluation controversies and mistrust of the system by the pilots, resulting in possible negative safety implications.

Another undesirable result of attempting to evaluate CRM would be the unwillingness of pilots to be themselves during evaluation and training. What is much more likely is that they would act the way they perceive the check airman wants them to act in order to achieve a passing grade. This would result in a misrepresentation of the crew’s CRM skills and most likely some undiscovered deficiencies in a crew’s performance, primarily because the evaluator wouldn’t have established a realistic representation of how the crew conducts CRM during normal line operations and thus could not have provided meaningful feedback.

Introducing jeopardy assessment after 30 years of effective CRM training completely undermines the fundamental principles of fifth generation CRM. The success of an effective fifth generation CRM program that focuses on threat and error management requires the formal understanding that errors will occur and that companies must adopt a non-punitive approach to error. Introducing assessment/checking of CRM skills would introduce the possibility of failure which could be perceived by many pilots as punitive.

Since effective CRM must be embedded within the safety culture of the airline, and since it similarly needs to be practiced and accepted at all levels of the organization to positively affect operational safety, it is difficult to independently assess/check only one single element (in this case the pilots) on their company culture skills when these are actually dependent on multiple personnel performing multiple tasks across the entire company.

To evaluate only one aspect of a company’s CRM system would do little to increase the safety of the entire system. Further complicating the issue is that evaluation would be based mostly on subjective evaluation criteria that have already proven very difficult to use as a basis for training and calibrating instructors/evaluators.

“Just because crews can demonstrate effective crew coordination while being assessed under jeopardy conditions does not guarantee they will actually practice these concepts during normal line operations. Industry studies show that line audits, where crews are observed under non-jeopardy conditions, provide more useful data.”

Just because crews can demonstrate effective crew coordination while being assessed under jeopardy conditions does not guarantee they will actually practice these concepts during normal line operations. Industry studies show that line audits, where crews are observed under non-jeopardy conditions, provide more useful data (Helmreich, Merritt, & Wilhelm, 1999). Data from such audits demonstrates that changes in pilot behaviour result from CRM training that includes LOFT and recurrent training (Helmreich & Foushee, 1993), which is consistent with participant feedback.

Summary
IFALPA supports CRM as a training program and as an adjunct to traditional technical training programs. IFALPA recognizes the substantial benefits arising from training of non-technical skills and supports the continued instruction and reinforcement of CRM on a regular basis. CRM can improve the proficiency and competency of individual pilots and flight crews as a whole, especially when it is implemented as an error management strategy and is not checked/assessed by any method that could result in a failure.

Instead of jeopardizing the safety record of an already successful CRM program by introducing CRM skill checks that have no demonstrated safety benefits, industry and regulators should instead focus their efforts on producing comprehensive guidance on how to properly train CRM and measure its effectiveness across the entire culture within an airline. This would include developing training guidance on: how to effectively teach error management skills; specific error prevention techniques; integrating CRM training into scenario-based training; integrating flight management skills with technical skills; helping pilots develop decision-making skills; and lastly, training pilots on how to properly manage resources in today’s complex aircraft/airspace system.

Pilot CRM skills have been used in many high-profile “saves”, such as the UAL 232 complete hydraulic failure in 1989, or more recently the US Air 1549 landing in the Hudson River. More important to overall industry safety is the fact that nearly a half million pilots successfully use their CRM skills day-in and day-out, safely completing nearly 100 000 daily flights without ever having had jeopardy assessment of their CRM skills.

About the author
Captain David McKenney is a B-767 pilot for United Air Lines and is the Vice Chairman (Human Factors) for the IFALPA Human Performance Committee. He also serves as a human factors and training expert for the Air Line Pilots Association, International (ALPA). Prior to his airline career, Captain McKenney was a Computer Science Professor at the U.S. Air Force Academy. He also served as Co-chair of the 2010 FAA Industry Stall/Stick Pusher Working Group and is Co-chair of the United States PARC/CAST Flight Deck Automation Working Group. Captain McKenney has accumulated over 16 000 hr in 35 years of military and civilian flying and has additionally served as a flight instructor and check airman.

Date modified: