If you wish to contribute or participate in the discussions about articles you are invited to join SKYbrary as a registered user


TCAS RA Survey

From SKYbrary Wiki

Article Information
Category: Safety Nets Safety Nets
Content source: SKYbrary About SKYbrary


The EUROCONTROL Safety Improvement Sub-Group (SISG) identified TCAS RA Not Followed as one of the Top 5 ATM Operational Safety Priorities in 2017.

There was no reliable and representative data on the number of TCAS RA not followed. The results from different sources ranged from 1% to 35%. Therefore, a survey on the matter among pilots was conducted. It was voluntary and received 3800 responses from 90 countries.

The aim of the survey was to gather and analyze statistical data to better understand the factors that lead to TCAS RA non-compliance and the impact of the operational environment on such occurrences.

The time period covered by the survey was 5 years ending in 2017.

A dedicated study combined the results of the survey with theoretical analysis of relevant generic scenarios and possible safety barriers.

Study Findings

General Statistics

  • About 36% of the participants reported to have “encountered a TCAS RA situation in the past 5 years”.
  • About 5% of the participants reported experiencing a TCAS RA that was not followed either at all or not as commanded.
  • About 15% of all the pilots that experienced TCAS RA participated in a TCAS RA not followed event.
  • About 12% of the participants reported having 3 or more TCAS RAs during the 5-year period.

The statistical analysis that took into account the number of events (i.e. more than one event per person) concluded that the percentage of RAs that are not followed is likely to be around 11%.

Reasons for TCAS RA Not Followed

A total of 18 reasons (out of 20 possible) were stated for not following a TCAS RA.

  • The most reported reason was "Decision not to follow due to visual acquisition and/or avoidance of the conflicting traffic" (45% of all non-follow events).
  • The second most reported reason was "Short duration RA (RA terminated before a response could be taken)" (15% of all non-follow events, i.e. 3 times less frequent than the first one).
  • The third most reported reason was "Proximity to the ground" (11% of all non-follow events, i.e. 4 times less frequent than the first one).

Reason Analysis

  • The Top 3 reasons account for 70% of all responses (45 + 15 + 10).
  • ICAO documents state that all RAs should be followed unless the safety of the aircraft is at risk. Three reasons (proximity to terrain, priority of other alerts and performance issues) with a total of 14% fall within this definition.
  • Three other reasons are considered valid for non-compliance (short duration, false/incorrect RAs and closely spaced runways). These account for 24 % of the answers.
  • About 38% (14 + 24) of RAs Not Followed might be for legitimate reasons.
  • The "Decision not to follow due to visual acquisition and/or avoidance of the conflicting traffic" is not considered a valid reason and account for about 70% of the non-legitimate reasons. This is the prevailing answer for all flight levels.
  • The level of inability to follow RAs due to their short duration is constant across all altitude bands.
  • About 46% of RAs not followed are below 3000ft. Most of these involve a decision not the follow the RA after visual acquisition of the apparent intruder.

Pilot TCAS Training

  • Pilots who, in training, are not pre-warned that a TCAS RA is coming, are disposed to react better in live situations.
  • No one type of TCAS RA is trained more than 80% each time. The less common, but perhaps more demanding RAs such as Crossing Climb/Descend are only trained around 30% of the time.
  • About 5% of pilots involved in TCAS RA Not Followed events report no TCAS training within the previous 5 years.

Study Recommendations

The dedicated study made 5 recommendations:

  • that IATA, Pilot Associations, Aircraft Operators and Regulators review the findings of this study and consider undertaking operational safety analysis and improvement activities for “TCAS RA Not Followed”.
  • that IATA, Pilot Associations, Aircraft Operators and Regulators consider actions to support an increased active use of FDM in the monitoring of TCAS RA compliance and the provision of feedback to training organisations and flight crew involved.
  • that European ANSPs and the EUROCONTROL Safety Improvement Sub-Group (SISG) monitor occurrences involving “TCAS RA Not Followed” to determine changes in frequency and severity.
  • that all European stakeholders monitor and support the development of tools and procedures that may assist in the prevention and/or mitigation of “TCAS RA Not Followed” events.
  • in particular, that all European stakeholders promote and emphasise the requirement and importance of following TCAS RA commands despite an apparent intruder being visually identified and monitored (subject to the overriding safety of the aircraft).

Accidents and Incidents

  • A310 / B736, en-route, Southern Norway, 2001 (On 21 February 2001, a level bust 10 nm north of Oslo Airport by a climbing PIA A310 led to loss of separation with an SAS B736 in which response to a TCAS RA by the A310 not being in accordance with its likely activation (descend). The B736 received and correctly actioned a Climb RA.)
  • A319 / PRM1, en-route, near Fribourg Switzerland, 2011 (On 10 June 2011 an ATC error put a German Wings A319 and a Hahn Air Raytheon 390 on conflicting tracks over Switzerland and a co-ordinated TCAS RA followed. The aircraft subsequently passed in very close proximity without either sighting the other after the Hahn Air crew, contrary to Company procedures, followed an ATC descent clearance issued during their TCAS ‘Climb’ RA rather than continuing to fly the RA. The Investigation could find no explanation for this action by the experienced crew - both Hahn Air management pilots. The recorded CPA was 0.6 nm horizontally at 50 feet vertically.)
  • A320 / B738, en-route, near Córdoba Spain, 2014 (On 30 October 2014, a descending Airbus A320 came close to a Boeing 737-800 at around FL 220 after the A320 crew significantly exceeded a previously-instructed 2,000 fpm maximum rate of descent assuming it no longer applied when not reiterated during re-clearance to a lower altitude. Their response to a TCAS RA requiring descent at not above 1,000 fpm was to further increase it from 3,200 fpm. Lack of notification delayed the start of an independent Investigation but it eventually found that although the A320 TCAS equipment had been serviceable, its crew denied failing to correctly follow their initial RA.)
  • A320 / B738, en-route, north northwest of Sofia Bulgaria, 2016 (On 4 June 2016, a Boeing 737-800 instructed to climb from FL340 to FL380 by the controller of one sector in Bulgarian upper airspace came into sufficiently close proximity to an Airbus A320 under the control of a different sector controller to trigger co-ordinated TCAS RAs. Separation was eventually restored after the 737 followed its RA despite the A320, which had already deviated from its clearance on the basis of a prior TCAS TA without informing ATC, ignoring their RA. The Investigation found that the root cause of the conflict had been inadequate coordination between two vertically separated ATC sectors.)
  • A320 / E195, vicinity Brussels Belgium, 2018 (On 23 February 2018, an Embraer 195LR and an Airbus A320 on SIDs departing Brussels lost separation after the 195 was given a radar heading to resolve a perceived third aircraft conflict which led to loss of separation between the two departing aircraft. STCA and coordinated TCAS RA activations followed but only one TCAS RA was followed and the estimated minimum separation was 400 feet vertically when 1.36 nm apart. The Investigation found that conflict followed an error by an OJTI-supervised trainee controller receiving extended revalidation training despite gaining his licence and having almost 10 years similar experience in Latvia.)
  • A320/B738, vicinity Delhi India, 2013 (On 2 September 2013, a B737 crew were not instructed to go around from their approach by ATC as it became increasingly obvious that an A320 departing the same runway would not be airborne in time for a landing clearance to be issued. They initiated a go around over the threshold and then twice came into conflict with the A320 as both climbed on similar tracks without ATC de-confliction, initially below the height where TCAS RAs are functional. Investigation attributed the conflict to ATC but the failure to effectively deal with the consequences jointly to ATC and both aircraft crews.)
  • B752 / B752, en-route, north of Tenerife Spain 2011 (On 20 November 2011, a problem in reading the altitude labels on the ATC radar control display led to a Finnair Boeing 757 being cleared to make a descent which brought it into proximity with a Thomas Cook Boeing 757 in day VMC. Co-ordinated TCAS RAs were generated onboard both aircraft but when the Finnair aircraft failed to respond to its Climb RA and continued descent, the other aircraft, which had responded correctly to its initial RA, received a further RA to reverse their descent to a climb. The Finnair aircraft reported retaining visual contact with the other aircraft throughout.)
  • C525 / B773, vicinity London City UK, 2009 (On 27 July 2009, a Cessna 525 departing from London City failed to comply with the initial 3000 feet QNH SID Stop altitude and at 4000 feet QNH in day VMC came into close proximity on an almost reciprocal heading with a Boeing 777-300ER. The 777, on which line training was being conducted, failed to follow any of the three TCAS RAs generated. Actual minimum separation was approximately 0.5nm laterally and estimated at between 100 feet and 200 feet vertically. It was noted that the Cessna had been given a stepped climb SID.)
  • DH8A/DH8C, en-route, northern Canada, 2011 (On 7 February 2011 two Air Inuit DHC8s came into head-to-head conflict en route over the eastern shoreline of Hudson Bay in non radar Class ‘A airspace when one of them deviated from its cleared level towards the other which had been assigned the level 1000 feet below. The subsequent investigation found that an inappropriate FD mode had been used to maintain the assigned level of the deviating aircraft and noted deficiencies at the Operator in both TCAS pilot training and aircraft defect reporting as well as a variation in altitude alerting systems fitted to aircraft in the DHC8 fleet.)
  • DH8D / DH8D, vicinity Sudbury ON Canada, 2016 (On 14 October 2016, two Bombardier DHC8-400s received coordinated TCAS RAs as they came into opposite direction conflict near Sudbury, an uncontrolled airport, as one was descending inbound and emerging from an overcast layer and the other was level just below that layer after departing. Both aircraft crews ignored their RAs and their respective visual manoeuvring brought them to within 0.4nm at the same altitude. The Investigation noted that the conflict had occurred in Class ‘E’ airspace after the departing aircraft had cancelled IFR to avoid a departure delay attributable to the inbound IFR aircraft.)
  • E55P, Blackbushe UK, 2015 (On 31 July 2015 a Saudi-operated Embraer Phenom on a private flight continued an unstabilised day visual approach to Blackbushe in benign weather conditions. The aircraft touched down with excess speed with almost 70% of the available landing distance behind the aircraft. It overran and was destroyed by impact damage and fire and all occupants died. The Investigation concluded that the combination of factors which created a very high workload for the pilot "may have saturated his mental capacity, impeding his ability to handle new information and adapt his mental model" leading to his continuation of a highly unstable approach.)
  • F100 / EC45, vicinity Bern Switzerland, 2012 (On 24 May 2012, a Fokker 100 descending visual downwind to land at Berne and an EC145 helicopter transiting the Bern CTR (Class 'D' airspace) VFR came within 0.7 nm horizontally and 75 ft vertically despite early traffic advice having been given to both aircraft. The Investigation attributed the conflict to the failure of the F100 crew to follow either their initial TCAS RA or a subsequent revised one and noted that although STCA was installed at Berne it had been disabled "many years before".)

... further results

Related Articles

Further Reading