A “technical issue” has caused UK National Air Traffic Services, NATS to impose air traffic flow restrictions[1]. They did not close UK airspace. This was not a repeat of the volcanic ash events of early 2010. Going from a fully automated system to a fully manual system had the dramatic impact that might be expected. The consequences, on one of the busiest weekends in the holiday calendar were extremely significant. Huge numbers of people have had their travel disrupted. Restricting the air traffic system ensured that aviation safety was maintained. The costs came to the UK’s air traffic handling capacity and that meant delays and cancelled flights.
Although the failures that caused the air traffic restriction were quickly resolved the time to recover from this incident meant it had a long tail. Lots of spoilt holidays and messed up travel plans.
It is normal for an Air Traffic Service (ATS) provider to undertake a common cause failure analysis. This is to identify multiple failures that may result from one event. So, the early public explanations coming from NATS of the causes of this major incident are surprising. Across the globe, contingency planning is a requirement for ATS. The requirement for the development, promulgation and application of contingency plans is called up in international standards, namely ICAO Annex 11.
So, the story that a single piece of flight data brought down the traffic handling capacity of a safety related system, to such a low level, is difficult to accept. It’s evident that there is redundancy in the systems of NATS, but it seems to be woefully inadequate when faced with reality. ATS comprise of people, procedures, and systems. Each has a role to play. Safety of operations comes first in priority and then air traffic handling capacity. What we know about even highly trained people and data entry is that human error is an everyday issue. System design and implementation needs to be robust enough to accommodate this fact. So, again attributing such a highly disruptive event to one set of incorrect data inputs does not chime with good practice or basic aviation safety management. It is concerning that one action can bring down a major network in this way.
EUROCONTROL would have had been sent a “rogue” flight plan in the same way as UK NATS. Brussels does not seem to have had the problems of the UK.
It is early days in respect of any detailed technical investigation. Drawing conclusions, whatever is said in public by senior officials may not be the best thing to do.
Calls for compensation have a good basis for proceeding. The holiday flight chaos across Europe comes down to one single failure, if initial reports are correct. That can not be acceptable. The incident left thousands stranded abroad with high costs to pay to get home.
Before privatisation, there was a time when the UK Civil Aviation Authority (CAA), ran the nation’s air traffic services[2]. It had a poor reputation at the time. I remember a popular newspaper cartoon saying – and now for some clowns from the CAA. They were entertaining delayed passengers.
UK NATS has done much good work to manage a safe expansion in air traffic and address many changes in technology, it would be a shame if this sad incident marks a decline in overall network performance.
NOTE 1: And this topical cartoon from the Daily Mail in April 2002: https://www.pinterest.es/pin/497577458805993023/
NOTE 2: A report on the incident is to be sent to the regulator, UK CAA on Monday, 6th September. Transport secretary to see Nats’ ATC meltdown report next week | Travel Weekly
NOTE 3: The likelihood of one in 15 million sounds like a low number but it’s not “incredibly rare” by any definition. Certainty when there are around 6000 flights a day in the UK. A duplicate error occurring is a basic error that could be anticipated.
[1] https://www.bbc.co.uk/news/live/uk-66644343
[2] https://commonslibrary.parliament.uk/research-briefings/sn01309/
John Vincent 