What caused the crash of Comair Flight 5191?

An interesting article in today's Lexington Herald-Leader summarizes the points raised by various groups about what caused the crash of Comair Flight 5191 on August 27, 2006. The views were included in documents submitted to the National Transportation Safety Board (NTSB) as contributions to its investigation of the accident at Lexington, Kentucky in which 49 of the 50 souls on board perished.

While the Herald-Leader article provides a good quick reference, using bullet points, here's some expansion on the newspaper summary, based on a reading of the actual documents.

The document submitted by Comair, Inc., the company operating the accident aircraft, acknowledges that the conduct of its flight crew contributed to the accident, but states that "it would be simple but inaccurate to conclude that the only cause of this tragic accident was a mistake by Comair's well-trained and experienced flight crew." In support of that argument, the document comments at length on the airline's internal safety management process.

The Comair document cites inadequate runway surveillance by the air traffic controller on duty at the time of the accident as a contributing factor. In addition, Comair notes that the Jeppesen and NACO (National Aeronautic Charting Office) charts used by its pilots had incorrect information, and that a NOTAM about the taxiway used that morning also was inaccurate.

A document submitted by NATCA (the National Air Traffic Controllers Association), states flatly, "The probable cause of the accident was Comair 5191 crew's failure to maintain situational awareness while taxiing for departure as well as failure of the crew to ascertain that the runway they were taking off from was the assigned departure runway." In support of their view, they include in their document excerpts of the pre-departure conversation between the two pilots, and between the flight crew and the air traffic controller, as transcribed from the aircraft's Cockpit Voice Recorder (CVR). The conversations, NATCA contends, show that the crew did not maintain the 'sterile cockpit' requirement while they were taxiing, and that this probably compromised their situational awareness.

NATCA cites the FAA's failure "to properly staff the Air Traffic Control tower" as a contributing factor. NATCA notes that there was only one controller on duty in the tower at Lexington at the time of the accident, while FAA directives actually required that there be two controllers on duty on that shift -- one for tower functions, and one for radar functions. NATCA's document cites a list of 31 discrete activities carried out by the lone controller in the tower in the 23 minutes immediately preceding the accident. They use this list to illustrate the task load placed on this single individual, noting that 14 of the 31 activities should have been carried out by a radar controller.

A third document was that submitted by the the Director of Operations of Lexington's Blue Grass Airport, the site of the accident. This document also cites the CVR transcript as evidence for the "loss of situational and location awareness by the flight crew" and violation of sterile cockpit procedures. This led to "the positioning of the airplane incorrectly on Runway 26." The airport management's document goes on to cite "at least eight examples of inattention and lack of focus by the flight crew, especially the First Officer, all of which illustrated the lack of professionalism by the pilots."

In their own defense, the airport management's document states that the airport taxiways, markings, lights and signage all complied with safety standards and certification requirements of the FAA. They point a finger of blame at the FAA and Jeppesen for publishing and distributing inaccurate charts and diagrams of the airport runways and taxiways, while mentioning that a NOTAM issued by the airport did provide "correct information to supplement the existing charts and diagrams."

The fourth document in this set was submitted by the Air Line Pilots Association (ALPA), the union representing the flight crew of Comair Flight 5191. At 127 pages in length, ALPA's submission is the lengthiest. Here is the Overview from that document (reparagraphed for easier reading).

As a party to this investigation, the Air Line Pilots Association, International (ALPA) has identified numerous safety concerns which will be addressed in this document.

Some of the areas of concern are as follows: First, numerous changes to the airport’s layout were not accurately presented to the crew.

Second, some of the Notices to Airman (NOTAM) reflecting the changes to the airport were not made known to the accident crew.

Third, ALPA has identified several human factor concerns relating to situational awareness including fatigue and workload management.

Fourth, our investigation has identified areas of concern regarding air traffic control policy and procedures which led to inadequate staffing, fatigue, and lack of controller vigilance.

Finally, poor coordination between the air traffic controller and the airport crash and rescue personnel delayed the first responders. Additionally, lack of proper emergency locator equipment caused further delay when personnel were not immediately able to accurately locate the wreckage location.

The document goes on to present an extensive analysis, with numerous illustrations, and ALPA's recommendations regarding each of the points. In all, the ALPA document's findings detail 35 discrete items that they found contributed to the accident.

The ALPA document concludes that the flight crew were given misleading information from taxiway signage and lighting cues, which led them to believe that their aircraft was in the correct position for takeoff on Runway 22, when in fact they had entered Runway 26. In addition, ALPA cites a "deficiency in the FAA/NFDC/Jeppesen chart revision process," such that the charts available to the crew "did not accurately reflect what they would be encountering that morning during taxi operations."

The ALPA document also cites understaffing at the Lexington air traffic control tower on the morning of the accident, and acknowledges that the controller "was operating in a fatigued state." But unlike the other documents, which focus on the FAA's failure to ensure that the tower was adequately staffed, the ALPA document focuses on the actions of the controller himself, stating:

Available evidence indicates the controller did not maintain vigilance ensuring the correct taxi route and runway were used. After the controller cleared the aircraft for takeoff, he almost immediately turned his back to address administrative duties. Had he maintained an increased level of vigilance related to controlling this aircraft, he may have noticed that the aircraft had entered the incorrect runway. As part of his duties, the controller was responsible for issuing ATIS broadcasts at LEX. Critical NOTAM information which would have enhanced the situational awareness of the crew was omitted from the ATIS broadcasts.

This is an unusually long post for this blog, I know. If you have read this far, you must have a keen interest either in this specific accident, or in the aviation accident investigation process. In either case, I encourage you to read all of the documents mentioned in this post.

Many thanks to the Herald-Leader for publishing the links to the documents.