About

One recorded flight, analysed in depth

flightanalyser turns the recorded track of a single flight into a structured post-operation analysis: how it was flown phase by phase, how directly it was routed, what the light conditions were on board, and how it compares with other flights on the same route.

It works from recorded ADS-B surveillance data, and it says where that data is incomplete.

Who it is for

Anyone who wants to understand how a particular flight was operated, using figures that can be traced back to the data. The use is always informational.

  • Aviation enthusiasts

    See where a flight's block time went, whether it held before landing, and which side of the cabin had the sun.

  • Students

    Work through real flight phases, step climbs, holding patterns and routing allowances, with every threshold stated.

  • Analysts

    Place one sector against up to 60 others on the same route from the last 30 days, with sample sizes and exclusions reported.

  • Journalists

    Check what a recorded track shows about a specific flight, and what it cannot show, before quoting a figure.

How it works

Every flight passes through the same pipeline, whichever way it was loaded. One function analyses an upload, a lookup and the sample, so the three can never disagree about method.

  1. Step 1: Track

    One flight, loaded by flight lookup, route search, CSV upload or the bundled sample. Each becomes the same list of timestamped positions with altitude, groundspeed and track.

  2. Step 2: Validation

    Rows outside physical bounds are dropped and points sorted by epoch time; a file holding several flights is split into sessions. Reported speeds are checked against movement between positions, and gaps are bridged with flagged synthetic points.

  3. Step 3: Phases

    Nine phases from idle to taxi-in, and nine events that anchor every duration, chart and map marker. Thresholds use height above the measured ground elevation, not raw altitude.

  4. Step 4: Metrics

    Observed: Seen directly in the recorded track

    Durations, distances, maximum and cruise altitude, groundspeed percentiles, climb and descent rates, and sampling statistics.

  5. Step 5: Analysed figures

    Analysed: Worked out by flightanalyser: an estimate

    Cruise profile, holding, operational efficiency, daylight and data confidence. The route benchmark and operating pattern run only when requested.

Observed, analysed and reported

Every figure carries its provenance, because the kind of figure decides how much weight it can bear.

Observed: Seen directly in the recorded track

Read from the recorded track, or computed from it by arithmetic.

  • Times, durations and distances along the track
  • Maximum altitude and cruise altitude
  • Groundspeed percentiles, climb and descent rates
  • Gap count, longest gap and median sampling interval

Analysed: Worked out by flightanalyser: an estimate

Inferred by flightanalyser from the track and reference values. An estimate, not an operator record.

  • Efficiency score, grade and findings
  • Holding patterns and step climbs
  • Cabin-side sun, best view side and red-eye status
  • Benchmark percentiles and verdicts
  • Airports resolved by proximity to the track ends

Reported: Supplied with a flight lookup

Supplied with a flight lookup, and shown as received.

  • Registration and Mode-S address
  • Operator, and livery where it differs
  • Aircraft type code
  • Origin, destination and runways

Uncertainty is part of the result

A confidence score — high, medium or low — summarises gaps, interpolation, irregular or sparse sampling, and velocity data that contradicts the positions. Analysis notes then name each specific consequence, for example:

Analysis note7-minute gap during cruise; cruise levels and any step-climb timing across it are interpolated.

Data sources and attribution

Where each part of an analysis comes from, and on what terms.

  • Flight lookups

    Flight data from Flightradar24.

    Searches, tracks, gates, runways, airspaces, the aircraft's day, departure traffic, the route benchmark and the operating pattern are fetched from Flightradar24 when you look a flight up. CSV uploads don't use it.

  • CSV uploads

    Your own flight track CSV export.

    Parsed in your browser. The file is never sent to the server.

  • Airport database

    About 8,800 airports with IATA codes: names, ICAO codes, positions and elevations.

    Time zones are derived from each airport's coordinates and, near a border, can occasionally be the neighbouring zone.

  • Aircraft reference

    More than 80 commercial types: manufacturer, engines, typical seats, MTOW and range.

    Compiled from manufacturer specifications and type-certificate data sheets.

  • Maps

    Coastlines, borders and runways for the route map.

    Drawn from public-domain map outlines and the runway database; no map service is involved.

  • Sample flight

    The bundled demonstration.

    Synthetic: a simulated London–New York sector written in the seven-column CSV format. Not a real flight.

For information only

flightanalyser is for general information only. Its figures are estimates computed from recorded ADS-B surveillance data, which can be incomplete, delayed or wrong, and its analysed values rest on general reference figures rather than an operator’s own data.

It is not suitable for flight planning, dispatch, or any operational or safety-critical use. Its findings describe what the track shows, not why, and are not a judgement of any crew, airline or airport.

Look at a flight

Load a flight by lookup, upload a CSV export, or open the sample.