Dataset

airline-disruption-data-phase1b

Foot Traffic & Mobility Download 1 source

The airline-disruption-data-phase1b dataset is a tabular release hosted by publisher Dev123Hug456Face that compiles scheduled and actual flight-level operations data with delay, cancellation and diversion flags. It is distributed as a single Parquet split of roughly 12 million rows.

Open on Hugging Face

Data preview

Subset

Loading preview...

Dataset card

What this dataset contains

Each record corresponds to a single U.S. domestic flight and combines scheduling identifiers such as Airline, FlightNumber, OperatingAirlineCode and Tail_Number with route fields including Origin, OriginCityName, OriginState, Dest, DestCityName and DestState. Timing variables cover both scheduled and realised departure and arrival, with fields such as CRSDepTime, DepDelay, DepDelayMinutes, DepDel15, TaxiOut, TaxiIn, CRSArrTime, ArrDelay, ArrDelayMinutes, ArrDel15, AirTime and the timestamp FlightDate. Disruption outcomes are captured through Cancelled, CancellationCode and Diverted. Date parts are split into Year, Month, DayofMonth and DayOfWeek for calendar-based slicing.

Structure and fields

The release is a single train split of 12,038,708 rows stored in Parquet format, with a library footprint spanning Hugging Face datasets, Dask, Polars and the MLCroissant metadata spec. Numeric dtypes vary by field, with int8 flags for Del15, Cancelled and Diverted indicators, float32 measures for the delay and taxi durations, float64 for FlightNumber and int64 for the scheduled CRSDepTime and CRSArrTime stamps. String dtypes carry the airline, route and cancellation-code values, while FlightDate is encoded as a timestamp[ns]. The tagged size band is between 10 million and 100 million rows.

How analysts use it

Because OriginCityName entries in published samples are redacted, any city-level work has to rely on the Origin, OriginState and airport-code fields rather than full city names. Delay and cancellation histories by Airline, Origin and Dest support airline-operations benchmarking, route-level on-time performance scoring and AirTime-based turnaround modelling. Disruption flags (Cancelled, CancellationCode, Diverted) can be aggregated as a proxy for airline reliability stress, which is often used as an alternative read on passenger volume, hub utilisation and consumer travel friction in the U.S. domestic market.

Documentation & references 1