Transit Demand Estimation And Crowding Prediction Based On Real-Time Transit Data

With an increasing number of intelligent analytic techniques and increasing networking capabilities, municipal transit authorities can leverage real-time data to estimate transit volume and predict crowding conditions. We introduce a proactive Transit Demand Estimation and Prediction System (TraDEPS) – an approach that has the potential to prevent crowding and improve transit service, by measuring the transit activity (the number of passengers on the individual modes of public transportation and the demand on a route), and estimating crowding levels at a given time. This system utilizes a combination of real-time data streams from multiple sources, a predictive model and data analytics for transit management. The problem of transit crowding is translated into transit activity prediction, as the latter is a straightforward indicator of the former. This thesis delivers the following contributions: (1) A crowding prediction model. (2) A system supporting the methodology. (3) A feature which displays different crowding level conditions of a route on a web map.

Name thesis:degree_name

M Sc

Discipline thesis:degree_discipline

Computer Science

Grantor dc:publisher

The University of Western Ontario

Year dc:date.issued

2014

Author dc:creator

• Aro, Michael

Advisor dc:contributor.advisor

• Bauer, Michael A.

• Transit crowding
• Data Analytics
• Sensor
• Crowd
• Visualization
• Networks
• Big Data
• Presence
• Location
• Buses
• Passengers
• Urban Transit Prediction

Language dc:language.iso

en_ca

Handle dc:identifier.uri

https://hdl.handle.net/20.500.14721/35049

OAI identifier oai:identifier

oai:uwo.scholaris.ca:20.500.14721/35049