Artificial intelligence has been mobilised by a collective industry effort to address the long-running issue of NOTAM complexity, and lies at the heart of a model developed to simplify them.
NOTAMs contain a variety of information for pilots, including essential safety-critical data.
But cockpit representatives are dismayed over the sheer quantity of NOTAMs with which pilots are presented, particularly given that the information within them is not filtered or listed by priority, and appears in a format which is difficult to read. A substantial proportion of NOTAM information is also outdated but has not been removed from the distribution system.
ICAO highlighted the problem two years ago, initiating a campaign intended to start reducing the burden, focusing particularly on removing old NOTAMs.
The NOTAM problem has been a notable concern for an industry collective called OpsGroup, which is supporting the NOTAM Alliance – an assortment of airlines, pilots, dispatchers and other end-users – in a concentrated effort to re-arrange raw NOTAM information, using artificial intelligence, and generate a simple, user-friendly briefing for pilots.
“The linchpin of this effort is creating an open-source blueprint that everyone is able to use,” says the group.
“We don’t want to create a closed-use, commercial product. We want to solve NOTAMs for everyone.”
It states that it will create an implementation guide which will allow all aircraft operators to use the process.
“Until this year, we were all convinced that the only way to fix NOTAMs was to change the way NOTAMs are issued by NOTAM officers,” the group says.
“But artificial intelligence has launched a cannonball of change into the aeronautical information sphere.”
Tests with artificial intelligence tools has shown that they can understand a NOTAM with a high level of reliability.
This has enabled the group to design a prototype system to gather raw NOTAMs, use artificial intelligence to apply tags from a list of 50 key phrases that pilots find most valuable – such as ‘runway closed’ – and provide an easy-to-read, plain-language summary.
The NOTAMs are then sorted, with the most critical and applicable ones being prioritised and presented clearly, while the remainder – dubbed ‘dark NOTAMs’ – are relegated to an appendix.
Some 300 individuals – including 175 pilots, and representatives of airlines, dispatchers, flight-planning providers, ICAO, OpsGroup and other organisations – participated in a five-day focused drive in May to shape the model.
“We’ve built something that works,” says the group. “But there is a next step – getting people to use it.
“The heavy lifting it done. Now we want to see it adopted beyond our core group of airlines and operators.”
It stresses that the model is not a front-end app, for a mobile phone, but intended for airline operational control centres or flight-planning offices. Work is continuing to refine the development and improve its efficiency.