Fare evasion is a critical threat for Transit Agencies (TAs) and/or Public Transport Companies (PTCs) everywhere, especially in Proof-of-Payment Transit Systems (POP-TSs). The research on fare evasion risk is limited and based on econometric models restricting time characterization to a single period. This paper aims to enhance the use of fare evasion risk over several periods for possible real-time deterrence against fare evasion. The paper moves from an existing framework, identifying the factors of fare evasion and risk exposure in terms of frequency (or probability) and severity (or vulnerability), and adopts Artificial Neural Networks (ANNs) to shed light on the intricate nexus between these components, estimating the fare evasion risk for every (segment of a) route. Next, the risk index is evaluated for each time period of interest. The predictions are ranked and represented by time-dependent dashboards to recognize routes with high-risk evasion that require deterrence strategies. Some real-time strategies are simulated from fare inspection logs, passenger surveys, and probability distributions on data collected in three years. In conclusion, this research provides actionable insights for TAs/PTCs in dealing with fare compliance and can be integrated into any bus transit management system.

Toward real-time deterrence against fare evasion risk in public transport

Di Francesco, Massimo
Secondo
;
Zanda, Simone
Ultimo
2024-01-01

Abstract

Fare evasion is a critical threat for Transit Agencies (TAs) and/or Public Transport Companies (PTCs) everywhere, especially in Proof-of-Payment Transit Systems (POP-TSs). The research on fare evasion risk is limited and based on econometric models restricting time characterization to a single period. This paper aims to enhance the use of fare evasion risk over several periods for possible real-time deterrence against fare evasion. The paper moves from an existing framework, identifying the factors of fare evasion and risk exposure in terms of frequency (or probability) and severity (or vulnerability), and adopts Artificial Neural Networks (ANNs) to shed light on the intricate nexus between these components, estimating the fare evasion risk for every (segment of a) route. Next, the risk index is evaluated for each time period of interest. The predictions are ranked and represented by time-dependent dashboards to recognize routes with high-risk evasion that require deterrence strategies. Some real-time strategies are simulated from fare inspection logs, passenger surveys, and probability distributions on data collected in three years. In conclusion, this research provides actionable insights for TAs/PTCs in dealing with fare compliance and can be integrated into any bus transit management system.
2024
Fare Evasion; Risk Prediction; Risk Management; Artificial Neural Network
File in questo prodotto:
File Dimensione Formato  
Articolo.pdf

accesso aperto

Tipologia: versione editoriale (VoR)
Dimensione 6.6 MB
Formato Adobe PDF
6.6 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11584/424463
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact