Please rotate your device to landscape mode to view the charts.

Transportation resilience under Covid-19 Uncertainty: A traffic severity analysis

Journal: Transportation Research Part A: Policy and Practice

Date: 2024

Author: Qiao Peng, Yassine Bakkar, Liangpeng Wu, Weilong Liu, Ruibing Kou, Kailong Liu

Abstract:
Transportation systems are critical lifelines and vulnerable to various disruptions, including unforeseen social events such as public health crises, and have far-reaching social impacts such as economic instability. This paper aims to determine the key factors influencing the severity of traffic accidents in four different stages during the pre- and the post Covid-19 pandemic in Illinois, USA. For this purpose, a Random Forest-based model is developed, which is combined with techniques of explainable machine learning. The results reveal that during the pandemic, human perceptual factors, notably increased air pressure, humidity and temperature, play an important role in accident severity. This suggests that alleviating driver anxiety, caused by these factors, may be more effective in curbing crash severity than conventional road condition improvements. Further analysis shows that the pandemic leads notable shifts in residents' daily travel time and accident-prone spatial segments, indicating the need for increased regulatory measures. Our findings provide new insights for policy makers seeking to improve transportation resilience during disruptive events.

Link: Google Scholar

Background and Context

Research Focus

This study examines how the Covid-19 pandemic affected traffic accident severity patterns in Illinois, USA from 2016-2023, focusing on identifying key factors influencing crash severity across different pandemic stages.

Methodology

The research uses Random Forest machine learning models combined with explainable AI techniques to analyze 69,971 traffic accidents across four distinct time periods: pre-pandemic, early pandemic, mid-pandemic, and post-pandemic stages.

Data Collection

Traffic accident data was collected from multiple sources including Bing Maps and MapQuest APIs, covering 27 different variables related to meteorological elements, weather conditions, transportation environment, traffic signs and event timing.

Shift in Traffic Accident Numbers During Covid-19 Stages

  • Shows the total number of traffic accidents across different pandemic stages
  • Peak in accident numbers occurred during Stage 3 (mid-pandemic)
  • Post-pandemic numbers show return towards pre-pandemic levels

Change in Accident Severity Distribution Over Time

  • Illustrates how the distribution of accident severity levels changed across pandemic stages
  • Notable increase in Level 3 (severe) accidents during early and mid-pandemic stages
  • Significant decrease in Level 4 (extremely severe) accidents during the pandemic

Impact of Key Factors on Accident Severity By Stage

  • Shows how different factors influenced accident severity before and during the pandemic
  • Human perception factors became more important during the pandemic
  • Traditional factors like traffic signals became less influential during Covid-19

Model Performance Across Different Stages

  • Compares the performance of different machine learning models across pandemic stages
  • Random Forest consistently outperformed other models
  • Model accuracy improved during the pandemic periods

Critical Time Periods for Severe Accidents

  • Shows how peak accident times shifted during the pandemic
  • Evening rush hour accidents decreased during Covid-19
  • Late night/early morning accidents increased significantly during the pandemic

Contribution and Implications

  • The study reveals that human perception factors like air pressure, humidity, and temperature became more critical during the pandemic than traditional road safety factors
  • Findings demonstrate significant shifts in accident patterns and timing during Covid-19, requiring adaptive traffic management strategies
  • Results provide evidence-based guidance for improving transportation system resilience during future public health crises

Data Sources

  • Accident stages visualization based on data from Table 3 in the article
  • Severity distribution chart created using data from Table 3 and Figure 1
  • Factor importance visualization derived from Figures 5, 9, 13, and 17
  • Model performance comparison based on Table 4
  • Time period analysis constructed from textual data in Results section and Figures 7, 10, and 16