01763803

Component

https://annualmeeting.myt...nlineProgram/Details/15648

Predictive modeling is key to studying passengers' behavior in transportation research. Modelling the public transport system can be used to estimate present and future demand and users’ trend toward public transport services. Machine learning techniques have proven to be better at recognizing the patterns and relations in the data. While, the traditional techniques are aimed at forming casual relationships and are unable to recognize patterns in the data. This paper seeks to predict the ridership at stop level for the new bus routes using the socio-economic data, building data, and ridership data of the existing routes at stop level. Neural networks (NN), a machine learning method has been applied to build predictive models. Ridership of the existing routes has been used to train and validate the model performance, which is able to predict the public transport ridership of the new routes. This model can be used by public transport agencies and relevant government organizations to predict the public transport demand for new commuters before introducing any new changes in the public transit system.

This paper was sponsored by TRB committee AP050 Standing Committee on Bus Transit Systems.

TRBAM-21-02877

English

23p

2021

Transportation Research Board 100th Annual Meeting

Web

Figures; References; Tables

Bus routes; Data models; Machine learning; Mathematical prediction; Neural networks; Public transit; Ridership

Chattanooga (Tennessee)

Data and Information Technology; Economics; Passenger Transportation; Public Transportation

Transportation Research Board Annual Meeting 2021 Paper #TRBAM-21-02877

TRIS, TRB, ATRI

Dec 23 2020 11:11AM