Industries / Transport and vehicles / Aviation
Flight operations
Flight operations is the work of planning and flying each trip: choosing the route and fuel load, finding the way, and handing parts of the flying to the autopilot.
5
Milestones achieved, from 1965 to 2025.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1965
An airliner lands itself
On 10 June 1965 a British European Airways (BEA) Hawker Siddeley Trident made what the Farnborough Air Sciences Trust museum calls the first automatic landing by a civil airliner in scheduled passenger service. The system could fly the approach, the flare (the nose-up leveling just before touchdown), the touchdown and the roll-out. The first "blind" landing in scheduled service, made in weather too poor to land by sight, followed on 4 November 1966.
Credit
the Royal Aircraft Establishment's Blind Landing Experimental Unit, BEA, Hawker Siddeley and Smiths Industries
1983
GPS offered to airliners
After a Soviet fighter shot down Korean Air Lines Flight 007, which had strayed into Soviet airspace, the White House said on 16 September 1983 that "the United States is prepared to make available to civilian aircraft the facilities of its Global Positioning System." Sources disagree on how much this changed: Richard Easton argued in The Space Review (2018) that civil use was built into GPS from the start, citing a 1974 program plan and the TI 4100 receiver that civilians used before 1983.
Lessona crew that does not know it has drifted off course cannot correct it, and a position fix from outside the aircraft is a guard against that.
Credit
the Reagan administration (statement read by Deputy Press Secretary Larry Speakes)
1985
Twin jets cross oceans
US rules kept two-engine airliners within 60 minutes' flying of an adequate airport. In 1985 the FAA began allowing certain twins on routes up to 120 minutes from an airport under ETOPS (extended operations), set out in Advisory Circular 120-42; approval depends on proven engine reliability, counted as in-flight engine shutdowns per 1,000 engine hours. The limit rose to 180 minutes in 1988 "based on demonstrated safety record of these operations," and to 207 minutes for some flights in January 2000 (FAA's own account, 2003).
Credit
Federal Aviation Administration (FAA)
2011
Tablets replace the flight bag
On 1 December 2011 the FAA authorized American Airlines pilots on the Boeing 777 to use iPads as electronic flight bags in all phases of flight, including takeoff and landing. An electronic flight bag is a tablet or computer that holds the charts and manuals pilots used to carry on paper; AFP reported the paper bag can weigh 35 pounds (15.9 kg) or more, against 1.5 pounds (0.7 kg) for the iPad. FAA spokesman Les Dorr called American "the first major air carrier to get approval for operational use of the iPad as an electronic flight bag."
Credit
American Airlines, with FAA approval
2025
Solo cruise not shown safe
Aircraft makers had proposed extended minimum crew operations (eMCO): one pilot flies long stretches of cruise alone while the other sleeps. A risk assessment done for EASA, the European Union's aviation safety agency, by the Dutch and German aerospace research centers NLR and DLR (report dated June 2025) found "there will be situations in which an equivalent level of safety cannot be demonstrated." The main gaps were spotting a pilot who becomes incapacitated, cross-checking, and grogginess after waking (sleep inertia).
Lessonautomation that can fly the aircraft does not yet do the second pilot's job of watching the first.
Credit
NLR and DLR, for EASA
Watch for
Contrail avoidance at airline scale
Contrails are the thin ice clouds that form behind jet engines; the study's authors say they make up a large share of aviation's effect on the climate. In a randomized trial from 15 January to 13 May 2025 on eastbound North Atlantic flights, American Airlines dispatchers were offered flight plans that changed altitude to avoid air where contrails form. The 112 flights that flew the plan as intended formed 62.0% fewer contrails than the control group, with an estimated 69.3% less contrail warming and no statistically significant change in fuel use; across all 1,232 eligible flights the drop was 11.6%, because dispatchers often declined the plans, citing workload, turbulence and mid-flight climbs and descents (the authors' figures). An earlier Google-led trial (2024) reported 63.6% fewer contrails, checked by hand on satellite images.
Achieved whena group outside the trial team measures the reduction on its own data, or an airline flies avoidance routinely and an outside party verifies the result.
Claimed by
Google Research, Contrails.org (Breakthrough Energy), American Airlines, Flightkeys and Imperial College London, March 2026 (preprint, not yet peer reviewed)
Open
01
Contrails matched to flights
A public test set of real contrails seen in satellite images, each matched by people to the flight that made it, using public ADS-B tracks (ADS-B is the position report each aircraft broadcasts). Anyone could then score the automatic contrail-to-flight matching that avoidance trials rely on.
The briefCredit · unclaimed02
Holding fuel at world airports
A yearly public table, airport by airport, of the minutes arriving aircraft spend in holding patterns (oval loops flown while waiting to land) and the fuel they burn doing it, for the world's busiest airports.
The briefCredit · unclaimed
Open 01 · the brief
Contrails matched to flights
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public test set of real contrails seen in satellite images, each matched by people to the flight that made it, using public ADS-B tracks (ADS-B is the position report each aircraft broadcasts). Anyone could then score the automatic contrail-to-flight matching that avoidance trials rely on.
Why it is still open
EUROCONTROL researchers wrote in July 2025 that there is "a lack of comprehensive, human-labelled data containing temporally resolved, instance-level and flight-attributed annotations," and that getting such data on geostationary satellite images "remains very difficult at this stage." Google's OpenContrails set marks contrails but not their flights, and SynthOpenContrails links flights only to synthetic contrails. The 2025 American Airlines trial checked its own results with an attribution algorithm built by the trial team.
Built with what exists
GOES-16 satellite images (the images behind OpenContrails); OpenSky Network ADS-B history; pycontrails, an open-source (Apache 2.0) Python library for modeling aviation's climate effects, to predict where each flight could have made a contrail; an open-source labeling tool such as CVAT.
First step
Pick five clear days over the US East Coast that appear in OpenContrails, pull the OpenSky tracks for the same hours, and hand-match 50 contrails to time the work and fix the labeling rules.
Done looks like
At least 1,000 contrails each linked to one flight, labeled separately by two people with their agreement rate published, plus the score of at least one published attribution method on the set.
Open 02 · the brief
Holding fuel at world airports
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A yearly public table, airport by airport, of the minutes arriving aircraft spend in holding patterns (oval loops flown while waiting to land) and the fuel they burn doing it, for the world's busiest airports.
Why it is still open
Work with open data so far covers a few European airports. Olive, Sun, Basora and Spinielli (PLoS ONE, 2023) studied two months (October and November 2019) at five airports (Heathrow, London City, Dublin, Paris Charles de Gaulle and Amsterdam) and found holding had a larger environmental cost than point merge (aircraft fly set arcs until turned toward a merge point) or continuous descent (one smooth descent without level steps). ICAO defines a related measure, additional time in terminal airspace (KPI08), and a data group of 14 Asia-Pacific states and administrations computes it for their major airports from data the members share with each other. It is measured in minutes, not fuel, and covers only the members' airports. A search for a worldwide yearly figure built from open data found none.
Built with what exists
OpenSky Network ADS-B history; the open-source traffic Python library, which detects holding patterns; the OpenAP aircraft performance model for fuel; the paper's open data (doi.org/10.4121/20411868) as a check.
First step
Download the paper's open data and rerun the holding detection for Heathrow, October 2019.
Done looks like
The paper's Heathrow results reproduced from its open data, then a full year published for at least 50 airports on at least three continents, with code, so that another person can rerun any one airport and get the same minutes of holding.