Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Unified Occupancy on a Public Transport Network through Combination of AFC and APC Data

Amir Dib, Noëlie Cherrier, Martin Graive, Baptiste Rérolle, Eglantine Schmitt

TL;DR

The unified occupancy method is introduced, a geo-statistical model to extrapolate occupancy to every course of a public transportation network by combining AFC and APC data with partial coverage, and is evaluated on real data from several public transportation networks in France.

Abstract

In a transport network, the onboard occupancy is key for gaining insights into travelers' habits and adjusting the offer. Traditionally, operators have relied on field studies to evaluate ridership of a typical workday. However, automated fare collection (AFC) and automatic passenger counting (APC) data, which provide complete temporal coverage, are often available but underexploited. It should be noted, however, that each data source comes with its own biases: AFC data may not account for fraud, while not all vehicles are equipped with APC systems. This paper introduces the unified occupancy method, a geostatistical model to extrapolate occupancy to every course of a public transportation network by combining AFC and APC data with partial coverage. Unified occupancy completes missing APC information for courses on lines where other courses have APC measures, as well as for courses on lines where no APC data is available at all. The accuracy of this method is evaluated on real data from several public transportation networks in France.

Unified Occupancy on a Public Transport Network through Combination of AFC and APC Data

TL;DR

The unified occupancy method is introduced, a geo-statistical model to extrapolate occupancy to every course of a public transportation network by combining AFC and APC data with partial coverage, and is evaluated on real data from several public transportation networks in France.

Abstract

In a transport network, the onboard occupancy is key for gaining insights into travelers' habits and adjusting the offer. Traditionally, operators have relied on field studies to evaluate ridership of a typical workday. However, automated fare collection (AFC) and automatic passenger counting (APC) data, which provide complete temporal coverage, are often available but underexploited. It should be noted, however, that each data source comes with its own biases: AFC data may not account for fraud, while not all vehicles are equipped with APC systems. This paper introduces the unified occupancy method, a geostatistical model to extrapolate occupancy to every course of a public transportation network by combining AFC and APC data with partial coverage. Unified occupancy completes missing APC information for courses on lines where other courses have APC measures, as well as for courses on lines where no APC data is available at all. The accuracy of this method is evaluated on real data from several public transportation networks in France.
Paper Structure (17 sections, 4 equations, 6 figures, 2 tables)

This paper contains 17 sections, 4 equations, 6 figures, 2 tables.

Table of Contents

  1. INTRODUCTION
  2. RELATED WORK
  3. FRAUD MODELING TO UNIFY OCCUPANCY FROM TICKETING DATA
  4. General principle
  5. Data preprocessing
  6. Ticketing data
  7. Counting cells data
  8. Fraud computation for stations with some counting cells measures
  9. Fraud computation for stations with no counting cells measure
  10. EXPERIMENTS AND RESULTS
  11. Experimental setup
  12. Performance evaluation
  13. Real courses examples
  14. Impact of counting cells coverage
  15. Fraud map interpretation
  16. ...and 2 more sections

Figures (6)

  • Figure 1: Boardings ($Y_i$), alightings ($Z_i$) and occupancies ($O_i$) over a network line.
  • Figure 2: Summary of unified occupancy method. All courses go through O/D reconstruction, then courses with APC data are used to compute average $R_i$ for a subset of the stations. Geospatial regression provides $R_i$ for the stations without counting cells measures. In the end, the $R_i$ are used to compute the unified occupancy for courses without APC data.
  • Figure 3: Occupancies of a course in Nevers in function of the serviced stops.
  • Figure 4: Occupancies of another course in Nevers in function of the serviced stops.
  • Figure 5: Performance of occupancy reconstruction on one line initially fully equipped and of which counting cells were sequentially removed.
  • ...and 1 more figures