Description

On 10 February 2010 a Boeing 737-300 being operated by KLM on a scheduled passenger flight from Amsterdam to Warsaw, Poland took off in normal night visibility from a taxiway parallel to the runway for which take off clearance had been given and accepted - runway 36C. Because of the available distance and the absence of obstructions, the take off was completed uneventfully as was the subsequent flight.

Investigation

An Investigation was carried out by the Dutch Safety Board. There was concern that the Cockpit Voice Recorder (CVR) had continued to run after the event so that the communications relating to it had been overwritten, despite the existence of an operator procedure describing the option to trip the controlling circuit breaker during a flight. Recorded flight data was, however, available.

It was established that en route from the terminal gate to the runway, the aircraft had been de-iced on an apron specifically designated for that purpose. ATC had then given an instruction to taxi to the departure runway 36C via taxiway ‘A’ which was against prescribed direction of travel (see the diagram below) It was noted that there were two parallel taxiways adjacent to runway 36C, ‘A’ and ‘B’, with taxiway ‘A’ located furthest to the east. Whilst the aircraft was taxiing to the beginning of the runway, ATC suggested that if ready on reaching, the aircraft could take a shorter route to the take-off runway via entry W8 and this offer was accepted. Soon afterwards, the aircraft turned right on W8 and whilst between taxiway ‘A’ and taxiway ‘B’ received ‘line up and wait’ and then take off clearances in quick succession. It turned right again onto taxiway ‘B’ and after a short pause began a standing start take off.

At first, the runway controller failed to realise what had happened but when they realised, they decided not to order the aircraft to stop the take-off because it had already gained too much speed and there was no risk of collision ahead of it. Once the aircraft was safely airborne and in the climb, ATC advised of the taxiway take off and were informed by the crew that they were unaware of this.

Subsequently, it was found that another aircraft had been taxiing on Taxiway ’B’ in the normal direction and had been approaching the left turn onto the section which the KLM aircraft was using for take off to taxi along it in the opposite direction. As the KLM aircraft passed at a high ground speed from taxiway ‘B’ to the continuation of the parallel taxiway designated taxiway ‘D’ in front of the other taxiing aircraft, the latter aircraft had been 280 metres away from the intersection, a distance which was estimated to have been equivalent to approximately 30 seconds of taxi time.

Air traffic control informed the crew of the incident while the aircraft was climbing. The crew informed air traffic control that they were unaware they had taken off from a taxiway.

An illustration of the taxi route taken by the aircraft reproduced from the Official Report

Performance calculations carried out as part of the Investigation found that in the prevailing circumstances, a distance margin of 340 metres would have been available in the event that the aircraft had used the runway from intersection W8 and then rejected the take off at the calculated V1 speed. The total available length of the parallel taxiway as actually used from the W8 intersection was greater than the runway distance available, so that the margin using the taxiway would have been slightly larger. However, the taxiway was only half as wide as the runway.

Since the event had occurred in a complex airport environment to a crew based there and who were thus open to complacency, the Investigation considered various aspects of this ‘human factors’ context including the following:

• If pilots who are broadly familiar with a complex airport do not habitually validate their actual ground position against a chart then when the limits of actual familiarity are approached in good visibility, the risk of errors which might be a more obvious risk in poor visibility is raised.
• Lighting at Amsterdam meets all International Civil Aviation Organisation (ICAO) standards but some runway entries and exits do not have green centreline lights, as in the case of entry W8.
• The thin layer of snow which was present on W8 impeded crew awareness of their position.
• The lights of Runway 36C were inconspicuous at the location where the navigation error occurred but the taxiway lights were clearly visible.
• The fact that green taxiway centreline lights are all illuminated at all times the lighting system is in use rather than selected to correspond to a cleared taxiway route as applies at some other airports with complex taxiway networks.

The Investigation noted that neither the Airport Operator nor the ANSP had carried out any risk assessment of taxiway use. The instruction to taxi to runway 36C via taxiway ‘A’ meant that taxi was against the prescribed direction of travel as promulgated in the AIPs and by definition the aircraft would then have to cross taxiway ‘B’. Whilst ATC may override normal taxiway directional if this is deemed “necessary” , it was considered that the crew error could not have occurred if instructions to taxi via taxiway ‘B’ had been given in the absence of any necessity to do otherwise.

A review was also carried out of the prevailing arrangements for the transfer of departing traffic from Ground Control to Runway Control. It was noted that there was a requirement that “an aircraft may be transferred from the ground controller to the runway controller if there is no longer any room for error on which taxi route the aircraft should follow”. However, it was concluded that the application of this procedure was open to interpretation by individual controllers. In the particular case of tactical deviations from the promulgated directional use of taxiways, it was considered that ATC should specifically monitor aircraft position and use progressive instructions to avoid misunderstanding and possible error, whereas in the investigated case, take-off clearance had been issued to the aircraft before it had crossed taxiway ‘B’

The Conclusions of the Investigation were that the event was attributable to the flight crew’s lack of awareness of the aircraft ground position and that following factors played a role in this:

• The flight deck crew’s workload had increased after they had accepted the shorter route. As a result the crew had to enter changes in the flight management computer and had less time to visually check the aircraft’s position at the airport from the cockpit.
• The crew were not using a ground movement chart as they felt they were sufficiently familiar with their home base, Schiphol;
• The pilot in command was distracted by communications between the air traffic controller and a Boeing 747 taxiing in front of the aircraft that had taken a wrong route.

It was also concluded that the failure of ATC to intervene and prevent the occurrence could be attributed to the following:

• The air traffic controller was forced to shift his attention to another aircraft and assumed that the PH-BDP crew would follow his instructions correctly.
• After having received take-off clearance, the aircraft was no longer monitored until an air traffic control officer in the air traffic control Tower saw it take off from the taxiway.

Safety Action taken by KLM was noted as was the absence of similar action by the other parties involved.

Five Safety Recommendations were made as a result of the Investigation as follows:

1. that (airport operator) Amsterdam Airport should prepare a risk assessment of air traffic taxiing near take-off and landing runways in collaboration with air traffic control and implements the outcomes in its procedures, unless the risk assessment shows otherwise
2. that (airport operator) Amsterdam Airport changes the infrastructure so that all taxiways made available to air traffic control have green centreline taxi lights indicating the route(s) to be followed only.
3. that (ANSP) Air Traffic Control the Netherlands (LVNL) prepares a risk assessment of air traffic taxiing near take-off and landing runways in collaboration with the airport and implements the outcomes in its procedures;
4. that (ANSP) Air Traffic Control the Netherlands (LVNL) ensures - until such time as the risk assessment has been completed and the resulting outcomes have been implemented - that entries without green centreline taxi lights are no longer used during darkness if an aircraft is required to taxi across a taxiway.
5. that the European Aviation Safety Agency (EASA) and the Federal Aviation Administration (FAA) (should) increase the minimum recording time of the cockpit voice recorder (Cockpit Voice Recorder (CVR)) in order to better safeguard recorded data for the purpose of incident and accident investigation.

The Investigation was completed on 21 December 2011 and the English language version of the Final Report: Take Off From Taxiway / Amsterdam Airport Schiphol is now available.

Related Articles

• Runway Identification
• Runway Lighting
• Wrong Runway Use
• Situational Awareness
• Complacency