EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Collision Type Prediction Involving Elderly Drivers: A Machine Learning Framework With Explainability Using Kolmogorov–Arnold Networks

Philip F. Balyagati, Ntemi Methusela Masanja, Abdul S. Ngereza, Emmanuel Kidando, Angela Kitali

Journal of Advanced Transportation · 2026-01 · 2026 (1)

DOI: 10.1155/atr/1828684 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Advanced Transportation, online erschienen am 13.07.2026. Verfasst von Philip F. Balyagati und 4 weiteren Autor:innen. Die Arbeit nennt 72 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: U.S. Department of Transportation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Elderly drivers represent a growing and vulnerable population on U.S. roadways, exhibiting distinct collision patterns and heightened risk of severe injury. Accurate prediction and interpretation of collision types involving this demographic are critical for developing targeted safety interventions. This study evaluates the performance of four machine learning models—Kolmogorov–Arnold Networks (KANs), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Traditional Neural Networks (TNNs)—in classifying the manner of collision among elderly drivers using data from the Fatality Analysis Reporting System (FARS). KANs offer a novel, interpretable neural network architecture that can capture complex nonlinear relationships with explicit feature representations, addressing the interpretability challenge common in deep learning. Model performance is assessed through accuracy, balanced accuracy, macro and weighted F1‐scores, and confusion matrices, while SHapley Additive exPlanations (SHAP) provide insight into feature importance and interaction effects. Results indicate that KANs achieve competitive predictive performance comparable to RF, XGBoost, and TNNs, with the added advantage of enhanced interpretability. SHAP analysis reveals key predictors and feature interactions influencing the predicted collision type, such as the combined effect of speed limit and roadway surface condition shifting predictions toward angle and head‐on configurations and the interaction of lighting and weather influencing the model’s distinction across collision types. These findings demonstrate the potential of interpretable machine learning approaches to improve understanding of crash configurations among elderly drivers, enabling more informed policy and engineering interventions.

Förderung

  • U.S. Department of Transportation

Zitieren

APA

Balyagati, P. F., Masanja, N. M., Ngereza, A. S., Kidando, E., & Kitali, A. (2026). Collision Type Prediction Involving Elderly Drivers: A Machine Learning Framework With Explainability Using Kolmogorov–Arnold Networks. Journal of Advanced Transportation, 2026(1). https://doi.org/10.1155/atr/1828684

Vancouver

Balyagati PF, Masanja NM, Ngereza AS, Kidando E, Kitali A. Collision Type Prediction Involving Elderly Drivers: A Machine Learning Framework With Explainability Using Kolmogorov–Arnold Networks. Journal of Advanced Transportation. 2026;2026(1). doi:10.1155/atr/1828684

BibTeX
@article{Balyagati2026,
  author = {Balyagati, Philip F. and Masanja, Ntemi Methusela and Ngereza, Abdul S. and Kidando, Emmanuel and Kitali, Angela},
  title = {Collision Type Prediction Involving Elderly Drivers: A Machine Learning Framework With Explainability Using Kolmogorov–Arnold Networks},
  journal = {Journal of Advanced Transportation},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/atr/1828684}
}
Kostenlos starten →