A loss of separation incident occurs, on average, once every three days in Australian airspace, says the Australian Transport Safety Bureau (ATSB).
In a report published on 18 October, the safety regulator notes that although 90% of the incidents pose little or no risk of collision, there are still about six incidents annually “where an elevated risk of collision exists”.
It also published investigation reports on two loss of separation incidents, where it found that air traffic controllers' (ATC) high workloads and the lack of an automated air traffic conflict detection system for aircraft not subjected to radar or surveillance broadcast, have played a part.
The first report relates to an incident on 18 January 2012 which involved a Tigerair Airbus A320, registered 9V-TAZ, and an Etihad Airways Airbus A340, registered A6-EHH.
The A320 was flying on the Singapore-Perth route while the A340 was on its way to Abu Dhabi from Sydney when the loss of separation occurred. The A320 was southbound at 35,000ft (10,700m) when the crew of the A340 was given block clearance to operate between 34,000ft and 36,000ft.
The A340 was at 34,500ft when its crew observed opposite direction traffic at 35,000ft on its traffic collision avoidance system. They initiated an immediate descent to 34,000ft, before the ATC even noticed the issue.
The second report relates to an incident which occurred between a Virgin Australia Boeing 737, registered as VH-VUV, and a Qantas Airways 737, registered as VH-VXM, on 8 November 2011.
VUV was on the Perth-Brisbane route while VXM, which was not equipped with an automatic dependent surveillance – broadcast (ADS-B) equipment, was heading to Melbourne from Port Hedland.
The loss of separation occurred when VUV was 10.9nm (20.2km) from Ceduna and VXM was also within 11nm of the vicinity. The arrival times for the two aircraft overhead Ceduna indicated that they were 2min apart, which was less than the required 10min separation standard.
ATSB found that traffic conflict between the two jets lasted for over 60min, but was not picked up by either controllers on duty, both of whom were experiencing high workloads exacerbated by their limited experience.
The controller who later detected the incident was also “stunned” and did not manage the compromised separation recovery situation effectively.
ATSB also found that Airservices Australia did not have a system to effectively manage the fatigue risk associated with allocating additional duties. The controllers in the incident had in the previous seven weeks on two occasions been given additional duty that resulted in a fatigue audit score above 80.
In both incidents, ATSB found that Airservices’ processes for monitoring and managing controllers’ workloads did not ensure that newcomers had sufficient skills and techniques to manage the high workload situations they were exposed to.
“Although the air traffic services provider has been working on the issue for several years, there was still no automated air traffic conflict detection system available for conflictions involving aircraft that were not subject to radar or ADS-B surveillance services,” it adds.
In response, Airservices says the first stage of a flight plan conflict function has been deployed in Brisbane upper airspace, with further roll out planned in Melbourne for 2014. It has also updated its fatigue risk management system.
Rostering arrangements have also been amended to ensure more controllers are scheduled during peak periods, and a group set up to determine a suitable workload model to be used by ATC shift managers to monitor and forecast ATC workload on a sector by sector basis.
However, the ATSB is not satisfied that Airservices has adequately addressed the issue of limited formal guidance for controllers on how to determine appropriate consolidation periods on one sector before they are transitioned to start training on another sector.