Paper ThAT13.5
Dong, Xiaomeng (Old Dominion University), Xie, Kun (Old Dominion University)
Inferring Risky Driving Behavior Shifts Pre-, During-, and Post-Pandemic with Hidden Markov Models
Scheduled for presentation during the Poster Session "Traffic prediction and estimation III" (ThAT13), Thursday, September 26, 2024,
10:30−12:30, Foyer
2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada
This information is tentative and subject to change. Compiled on October 7, 2024
|
|
Keywords Data Mining and Data Analysis, Simulation and Modeling, Other Theories, Applications, and Technologies
Abstract
The COVID-19 pandemic has had a notable impact on transportation safety. Risky driving behaviors were observed more frequently during the pandemic, resulting in higher likelihood of severe crashes. However, there remains a dearth of research investigating the post-pandemic effects on driving behaviors and transportation safety. This study uses open data from the state of Virginia to examine shifts in risky driving behaviors from 2018 to 2023, spanning the periods pre-, during-, and post-pandemic. Factor analysis was employed to measure latent variables representing aggressive and inattentive driving behaviors. Additionally, Hidden Markov Models (HMMs) were used to infer shifts in hidden states related to these risky driving behaviors. The HMMs revealed four hidden states pertaining to risky driving behaviors. It was found that these states shifted to the highest risk at the beginning of the pandemic in early 2020, which persisted for nearly one year until the 2020 holiday season. Results also suggested that post-pandemic, inattentive driving behaviors became more frequent compared to pre-pandemic levels, while aggressive driving behaviors returned to the pre-pandemic frequency.
|
|