On 7 October 2017, the crew of a Boeing 787-9 (G-ZBKF) being operated by British Airways on a scheduled international passenger flight from London Heathrow to Delhi as BAW143 was on approach to runway 29 at destination in night VMC when it became obvious that a landing clearance was not going to be given because a departing Boeing 777-300 (HS-TKC) being operated by Thai Airways on a scheduled international passenger flight from Delhi to Bangkok as THA316 was still on its takeoff roll from the same runway. The crew-initiated go around of the 787-9 resulted in the two aircraft coming to within 0.2 nm laterally and 200 feet vertically.
A Serious Incident Investigation was carried out by the Indian Aircraft Accident Investigation Bureau (AAIB) in accordance with the Aircraft (Investigation of Accidents and Incidents) Rules 2017. Recorded ATC radar and RTF communication data and landline communications between the APP and TWR positions involved were available and relevant QAR data was supplied by British Airways. No details of the experience levels of the pilots or controllers involved were obtained.
It was found that the 787 had been given a closing heading to join the 29 ILS LLZ by APP radar. The Captain of the aircraft subsequently recalled that ATC speed control had been used but the RTF transcript did not include any such speed control instructions. Once fully established, the aircraft was transferred to TWR South approximately 10nm from touchdown as number 2 to land. At that time, FDR data showed that the 787 was reducing though 185 KCAS and the Airbus A321 7 nm in front was, according to ATC data, maintaining 140 knots. By the time the 787 had 7nm to run, it was 6.5nm behind the A321 ahead and at 183 KCAS. The A321 landed and, with the 787 showing on the radar display as having 6 nm to run, the departing 777 was cleared to line up at an intersection 110 metres from the beginning of the runway (the Investigation noted that the actual separation was slightly less “due to the displaced landing threshold and the intersection line up”). This clearance did not include confirmation that an immediate departure would be required when cleared.
Less than a minute later, with the 787 4nm from touchdown at 160 KCAS, the TWR controller cleared the 777 for takeoff but again did not add a requirement for an immediate departure. The 777 “took some time on the runway before commencing its takeoff roll” and when it was still not airborne as the 787 approached the runway threshold, the 787 commenced a go around from 200 feet due to the occupied runway. The TWR controller asked the 787 to expedite climb to 3,600 feet and, in coordination with the APP arrival radar controller, instructed it to turn left onto heading 180°. He then instructed the 777 to stop climb at 2,600 feet and transferred both aircraft to the APP arrival radar frequency. The action to prevent any further conflict was successful. The surface weather shortly before the conflict was recorded as 3000 metres visibility in haze with no significant cloud and it was concluded that it had not been a contributory factor in the conflict.
The 787 is 0.7 nm from touchdown at 160 knots and the 777 is at 140 knots on its takeoff roll. [Reproduced from the Official Report]
It was noted that the AIP India entry for the airport and the ANSP MATS Parts 1 (General) and 2 (Delhi IGI) variously included a number of requirements aimed at avoiding conflicts of the type under investigation at Delhi including that:
- pilots of departing aircraft shall complete all mandatory pre-departure checks before entering the active runway for departure so that the aircraft is in a position to take off immediately upon receipt of takeoff clearance [AIP]
- controllers observing a delay in a departing aircraft commencing its takeoff roll after being cleared to do so shall cancel that clearance and instruct the aircraft to vacate the runway immediately at the nearest (available) taxiway [AIP]
- the approach arrivals controller shall be responsible for providing the desired landing interval between arrivals and that the nominal spacing between arrivals when there is a departure in between two arrivals on runway 29 is 7nm although this may be increased when the speed differential between the preceding and succeeding arrivals is high [ANSP MATS Part 2]
- controllers shall not normally permit a landing aircraft to cross the runway threshold on its final approach until the preceding departing aircraft has crossed the end of the same runway or has started a turn or until all preceding landing aircraft are clear of the runway [ANSP MATS Part1]
- radar controllers are required without exception to instruct all arriving jet aircraft within 20 nm of touchdown to fly at no more than 180 knots (or 170 knots if necessary), then when on a closing heading for final approach (or from 12 nm from touchdown if straight in) to reduce to 160 knots and then between 10 nm and 5nm to be 160 knots. [ANSP MATS Part1]
It was noted that none of these requirements had been complied with and also observed that landline coordination between TWR and APP radar controllers had been effected in “non standard (local) language”.
The Investigation also considered without any explanation that the 787 crew should, in the absence of the mandatory ATC-instructed speed control, have automatically flown at the 787-9 “approach speed” contained in the “FAA Reference Approach Speeds for Boeing Aircraft” dated 30 March 2016. These figures were reproduced in full in the ANSP MATS Part 1 and were acknowledged as having been taken from data issued by Boeing Airport Compatibility Engineering and claimed to be thereby “laid down in SOP”. By clear implication, this was considered by the Investigation as if it was an aircraft operator SOP which would be automatically applied in the absence of ATC speed control instructions. The fact that this indicative guidance on ‘normal’ approach speeds had not been followed led the Investigation to formally determine that the “failure of the crew” to follow it was a “Contributory Factor” to the conflict (see below).
The Probable Cause of the event was determined as “non adherence to Standard Operating Procedures by the TWR controller”.
Three Contributory Factors were also identified as follows:
- Failure of the APP radar controller to provide adequate inter-arrival spacing and to apply speed control methods to arriving aircraft.
- Failure of the 777 flight crew to adhere to the published departure procedures.
- Failure of the 787 flight crew to adhere to the published speed control norms.
Two Safety Recommendations were made as follows:
- that Delhi International Airport should ensure that its TWR controllers are reminded of the importance of situational awareness in such situations and (directed) that they should not use a non-standard language during (land line) coordination.
- that Delhi International Airport should ensure that its APP controllers are made aware of the need to monitor inter-arrival spacing and use speed control techniques and always use the standard language during inter-unit coordination.
The Final Report was completed on 20 November 2018 but not published until 13 June 2019.