Smart Card Data in Venice

Raw data (data/raw) 1. Smart card datasets The raw validation datasets contain individual ticket validation records generated by ACTV systems during different temporal contexts: winter: period associated with the Venice Carnival (from 13 January to 14 March 2023); spring: period associated with the Easter holidays (from 4 April to 3 June 2023); summer-autumn: period associated with the Venice Film Festival (from 21 August to 20 October 2023). Each dataset represents ticket validations and includes the following columns: validation_datetime (timestamp): timestamp of the validation; serial (string): anonymised card identifier; stop (integer): stop identifier where the validation occurred; title (integer): internal ticket identifier; title_description (string): textual description of the ticket type. 2. Stop datasets Several auxiliary files describe the spatial structure of the transport network: stopsWater.csv: water transport stops with original identifiers, names, and geographic coordinates: stop_id (integer): unique identifier of the public transport stop; stop_name (string): name of the public transport stop; stop_lat (real): latitude coordinate of the stop; stop_long (real): longitude coordinate of the stop; stopsLand.csv: land transport stops with original identifiers and coordinates (same columns as stopsWater.csv); landKeyAreas.csv: a reduced set of five aggregated land key areas used to aggregate land transport stops: area_id (integer): identifier of the land area. area_name (string): name of the land area; area_lat (real): latitude of the land area; area_long (real): longitude of the land area. stopsLandMapped.csv: mapping between original land stop identifiers and aggregated land key areas: stop_id (integer): identifier of the public transport stop; area_id (integer): identifier of the land area the stop belongs to. Processed data (data/processed) The processed datasets are generated using the data processing pipeline provided in this repository. Each raw dataset is processed independently using the same pipeline to ensure comparability across periods. The processed datasets are characterised by the following attributes: validation_datetime (timestamp): timestamp of the validation; serial (string): serial number associated with the travel ticket; loc_id (integer): if the travel ticket validation occurred at a water bus stop, the value is the identifier of the public transport stop, otherwise it is the identifier of the corresponding land area; ticket_class (string): identifier of the ticket class associated with the travel ticket title; user_category (string): name of the user category the ticket class belongs to. File formats Processed data are distributed in the following formats: CSV (.csv): contains the processed validation records; JSON (.json): contains summary statistics and metadata describing the processing outcomes (e.g. number of records before and after cleaning).