This paper investigates time reliability at bus stops. Although it is typically evaluated from the transit provider's viewpoint, it must also account for passengers, as required in recent service-quality norms. Hence, data on both bus arrival (or departure) times and passenger arrivals must be collected and processed. Automated vehicle location (AVL) systems can collect bus data, but several challenges must be addressed to effectively use them. Passenger arrival data can be collected by surveys or direct observations and processed to derive patterns. This paper proposes two novel time reliability metrics: the percentage of passengers receiving regular service (PPR) and the percentage of passengers receiving punctual service (PPP) for regularity and punctuality evaluations, respectively. They are determined by a methodology that collects and handles AVL data, computes passenger patterns from passenger arrival data, and integrates AVL data and patterns. Experiments highlight the viability of the novel evaluation metrics using about three million of real-world AVL records. Their results are reported by straightforward Web tools. A comparison with traditional metrics shows that PPR and PPP provide a more careful evaluation by using the passenger as a normalization basis for their outcomes. In the new paradigm of demand-oriented services, the proposed metrics are crucial to quantify the ability of operators to serve passengers.

Rethinking Transit Time Reliability by Integrating Automated Vehicle Location Data, Passenger Patterns, and Web Tools

Benedetto Barabino
;
2017-01-01

Abstract

This paper investigates time reliability at bus stops. Although it is typically evaluated from the transit provider's viewpoint, it must also account for passengers, as required in recent service-quality norms. Hence, data on both bus arrival (or departure) times and passenger arrivals must be collected and processed. Automated vehicle location (AVL) systems can collect bus data, but several challenges must be addressed to effectively use them. Passenger arrival data can be collected by surveys or direct observations and processed to derive patterns. This paper proposes two novel time reliability metrics: the percentage of passengers receiving regular service (PPR) and the percentage of passengers receiving punctual service (PPP) for regularity and punctuality evaluations, respectively. They are determined by a methodology that collects and handles AVL data, computes passenger patterns from passenger arrival data, and integrates AVL data and patterns. Experiments highlight the viability of the novel evaluation metrics using about three million of real-world AVL records. Their results are reported by straightforward Web tools. A comparison with traditional metrics shows that PPR and PPP provide a more careful evaluation by using the passenger as a normalization basis for their outcomes. In the new paradigm of demand-oriented services, the proposed metrics are crucial to quantify the ability of operators to serve passengers.
2017
Automated Vehicle Location (AVL) raw data; bus reliability measure; service quality; web tools; Automotive Engineering; Mechanical Engineering; Computer Science Applications1707 Computer Vision and Pattern Recognition
File in questo prodotto:
Non ci sono file associati a questo prodotto.

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/11584/183980
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 22
  • ???jsp.display-item.citation.isi??? 19
social impact