Skip to main content
ICTD 2024 Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Abstract for presentation (Poster or Podium) with a Paper in the Conference Proceedings

Transportation Safety

Predicting Bicycle-Involved Crashes in SCAG Region: A Machine Learning Analysis Using HSIS Data

Sunday, June 16, 2024
1:15 PM - 2:30 PM ET
Room: Galleria
Abstract:

Bicycling is one of the popular modes of transportation in the U.S. However, crashes involving bicycles continue to be a significant safety issue. This study used Health and Safety Information System (HSIS) data to examine the factors influencing the severity of bicycle-involved crashes in the Southern California Association of Governments (SCAG) region between 2013 and 2017. The study developed a Bayesian network model with strong consistency validation and low error rate, indicating the effectiveness of the model in analyzing crash data and providing valuable insight into improving rider safety. The study also created two scenarios to understand the impact of different variables on the probability of having a fatal crash. The first scenario showed that having proper lighting and reflective clothing could increase visibility and reduce the probability of a fatal crash. The second scenario showed the need for designing roads and infrastructure that can handle wet conditions and provide adequate drainage to reduce the probability of a fatal crash. The study’s results demonstrate the importance of designing effective infrastructure and highlighting the significance of proper lighting and visibility on bicycles to reduce the probability of a fatal crash. The findings can help policymakers and transportation engineers prioritize measures to improve rider safety. Furthermore, the study highlights the effectiveness of Bayesian network models in identifying the most significant factors contributing to bicycle-involved crashes. Overall, the study provides valuable insight into improving the safety of riders and demonstrates the effectiveness of Bayesian network models in analyzing and predicting bicycle-involved crashes.

Learning Objectives:

Attendees can expect to learn the following from this session:
  • Understand the importance of investigating bikeshare program usage and its relationship to sociodemographic variables, particularly among different racial groups.
  • Learn how to conduct statistical analysis using the latest National Household Travel Survey (NHTS) data, which is a comprehensive national transportation dataset in the U.S.
  • Gain insights into the findings of the study, including disparities in average bike trips among different racial groups, variations in bikeshare program participation by gender.