Skip to main content
ICTD 2025 Main banner
Final Program


Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back

ICTD 2025

Artificial Intelligence (AI) in Transportation

POSTER RECEPTION (View list of posters here)

Predicting Incident Clearance Time on Louisiana's Rural Interstates Using Ensemble Methods

Tuesday, June 10, 2025
4:45 PM – 6:00 PM MT
Abstract: Accurate prediction and analysis of factors impacting crash-related incident clearance time are the two main objectives of a Traffic Incident Management (TIM) system, as it can reduce the negative consequences resulting from traffic incidents. This study developed three machine learning models, namely XGBoost, CatBoost, and LightGBM, to predict crash-related incident clearance time in Louisiana rural interstates and utilized SHAP analysis to determine the influence of factors impacting it. Four ICT levels were defined based on 30-minute intervals: short (0-30), medium (31-60), intermediate (61-90), and long (greater than 90). The results suggest that XGBoost outperforms CatBoost and LightGBM in the collective model's predictive performance. It was found that different features significantly affect different ICT levels. Results indicate that crashes involving injuries, fatalities, heavy trucks, head-on collisions, roadway departure, and older drivers are the significant factors that influence ICT. The results of this study may be used to develop and implement strategies that lead to reduced incident duration and related challenges with long clearance times. Achieving this goal will require the accurate prediction and dissemination of ICT to both motorists, particularly those approaching the incident scene, and various agencies involved in incident management, such as the traffic management center (TMC), fire, police, and emergency medical services (EMS). This would allow stakeholders to make informed decisions regarding their travel.

Keywords: Rural interstate, incident clearance time, machine learning, SHAP analysis.

Learning Objectives:

Attendees can expect to learn the following from this session:
  • to list the significant factors influencing incident clearance times.
  • To identify the best tools for predicting ICT
  • to understand the application of XGBoost, CatBoost, and LightGBM models in predicting incident clearance times on rural interstates.
  • to apply these machine learning techniques to develop predictive models that assist in the management and planning of traffic incident responses.
  • to critically evaluate and compare the performance of XGBoost, CatBoost, and LightGBM models in different scenarios, understanding their strengths and limitations in the context of traffic incident management.