Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 20, 2026Future TransportationOpen Access

Seasonal Patterns and Future Projections of ADAS and ADS Crashes: A Time-Series Forecasting Study

View Full Paper
Ask AI
Bookmark
Share

Authors

JBJoydeep BanikMMMd Emon MiahAHA K M Akther Hossain

Discussion

Loading...

Member takes

Overview

Time-series forecasting reveals crash trends in ADAS and ADS, emphasizing the need for timely safety measures.

Key Points

  • The aim is to analyze crash trends and predict future crash counts of ADAS and ADS vehicles using time-series forecasting models.
  • Developed seasonal autoregressive integrated moving average and Facebook Prophet models.
  • Trained on 30 months of crash data from July 2021 to December 2023 and validated on 6 months (January–June 2024).
  • Utilized validation metrics such as root mean square error, mean absolute percentage error, and Theil’s U1 statistic.
  • Facebook Prophet outperformed SARIMA, achieving an RMSE of 2.71 for ADAS and 2.24 for ADS; MAPE was 6.9% for ADAS and 8.85% for ADS.
  • ADAS crashes exhibited a bimodal pattern, peaking in January and May–June, with troughs in February–March and August–September.
  • ADS displays a trimodal pattern with peaks in April–May, August, and October.

Cite This Study

Banik et al. (2026) studied this question.

synapsesocial.com/papers/6a0d4f34f03e14405aa9a744https://doi.org/10.3390/futuretransp6030105
View Full Paper
Ask AI
Bookmark
Share