Identifying the impact of the COVID-19 pandemic on driving behavior using naturalistic driving data and time series forecasting

Katrakazas, Christos; Michelaraki, Eva; Sekadakis, Marios; Ziakopoulos, Apostolos; Kontaxi, Armira; Yannis, George · 2021 · Crossref

DOI: 10.1016/j.jsr.2021.04.007

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the lack of quantitative assessment regarding how the COVID-19 pandemic and associated lockdown measures impacted driving behavior. While previous research noted general reductions in traffic volume and crashes, specific changes in driver actions, such as speeding and harsh braking, remained unclear. The authors aimed to fill this gap by comparing observed driving indicators during the lockdown in Greece against forecasts derived from pre-lockdown data, thereby isolating the pandemic's effect from normal temporal variations. The methodology utilized naturalistic driving data collected via a smartphone application between January 1, 2020, and May 9, 2020. This period encompassed normal operations, the spread of the virus, and the full lockdown. The study focused on three key indicators: average speed, speeding, and harsh braking per 100 km. First, Extreme Gradient Boosting (XGBoost) algorithms were employed to identify the most influential COVID-19-related variables affecting these driving metrics. Subsequently, Seasonal AutoRegressive Integrated Moving Average (SARIMA) models were developed to forecast the expected evolution of these indicators based on pre-lockdown trends. These forecasts served as a baseline to compare against actual observed data during the lockdown period. The results revealed significant deviations in driving behavior during the lockdown compared to forecasted norms. Average speeds increased by 2.27 km/h on average, indicating that drivers tended to drive faster on emptier roads. Additionally, the frequency of harsh braking events increased to almost 1.51 per 100 km on average. Despite these negative behavioral shifts, the study noted a positive outcome in overall road safety: road crashes in Greece decreased by 49% during the COVID-19 months compared to the non-COVID-19 period. This reduction is attributed to the drastic decline in overall traffic volume, which outweighed the risks associated with increased speeding and harsh braking. The significance of this work lies in its provision of empirical, quantitative evidence on the nuanced impact of pandemic restrictions on driving behavior. It demonstrates that while lockdowns reduce crash frequency due to lower traffic exposure, they may simultaneously induce riskier driving behaviors, such as speeding and aggressive braking. These findings are crucial for policymakers and researchers, suggesting that future traffic management strategies must account for behavioral adaptations to low-traffic environments, rather than assuming reduced volume automatically equates to safer driving practices.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• telematics crash prediction
• exposure measurement

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: observational prevalence
• Methodological Resource: dataset resource
• Theoretical Contribution: computational model