Skip to main navigation Skip to search Skip to main content

Exploring bus service demand in emerging business formats: A text mining analysis

  • Chunqin Zhang
  • , Waner Li
  • , Jiachen Shou
  • , Hongbin Ma
  • , Junyan Xiang
  • , Martin Skitmore*
  • , Xian Liu
  • , Donglei Rong
  • , Wenbin Yao*
  • *Corresponding author for this work

Research output: Contribution to journalArticleResearchpeer-review

Abstract

This study examines passenger demand characteristics for bus services and offers policy optimization recommendations derived from these insights. An analysis of 6403 valid social media comments from January 2023 to December 2024, utilizing Gibbs Sampling Dirichlet Multinomial Mixture topic modeling and the Jaccard distance metric, identified four core themes: “travel experience,” “operational efficiency,” “economy,” and “environmental impact,” which were further categorized into twelve sub-themes. The findings reveal that passengers prioritize “travel experience” (encompassing driver attitude, cleanliness, seat comfort, and congestion) and “operational efficiency” (including schedule density, punctuality, route optimization, and peak-hour traffic management), followed by “economy” (ticket pricing and the senior free-ride policy) and “environmental impact” (low-carbon features and battery recycling). Based on the findings, several policy recommendations are put forward, including the enhancement of the in-vehicle environment, the implementation of comprehensive driver training programs aimed at improving the passenger experience, the optimization of scheduling and the promotion of real-time bus tracking systems to boost operational efficiency, the adjustment of the fare structure to ensure greater equity, and the refinement of management practices for new energy buses to support environmental sustainability.
Original languageEnglish
Article number104010
JournalTransport Policy
Volume179
DOIs
Publication statusPublished - Apr 2026

Fingerprint

Dive into the research topics of 'Exploring bus service demand in emerging business formats: A text mining analysis'. Together they form a unique fingerprint.

Cite this