Skip to content
naulon
Not checkedNot checked for notices yet.

Transportation mode choice behavior with recommender systems: A case study on Beijing

Xiaoqian Sun, Sebastian Wandelt

Transportation Research Interdisciplinary Perspectives · 2021

doi:10.1016/j.trip.2021.100408 (opens in a new tab)

We email you when a new notice or a question about this paper appears.

Abstract

Understanding and predicting mode choice behavior in urban areas is an ongoing challenge, with several factors identified in past studies, e.g. built-environment, household statistics, trip properties, and many models being developed, e.g., regression and nested logit models. Existing research studies are predominantly designed around stated preferences surveys on small subsets of a population. The massive use of smartphones and route recommendation systems, however, offers the possibility of interacting with users, opening the potential to better understand and influence mode choice behavior, compared to sole offline analysis.This study explores the ability to predict travelers’ mode choice behavior in Beijing based on a collection of 300,000 recommended transportation alternatives from Baidu. The unique context of Beijing, with its enormous congestion and excessive penetration of smart phones, provides a unique view on actual transportation mode choice at a large scale; and behavioral changes induced by mobile communication technologies. We use machine learning techniques to identify the effects of driving variables, including transportation mode accessibility, weather conditions, alternative trip costs, and time of day. We find robust evidence supporting the observation that users preferably select the first-ranked alternative provided by the route recommendation system. This observation should be exploited further by transportation policy-makers to guide users towards greener and environmental-friendly transport modes.

Abstract licence: CC BY-NC

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.