Related Experiment Video
Updated: Oct 11, 2025
07:15
Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research
Published on: December 18, 2020
4.6K
A hybrid machine learning model for predicting Real-Time secondary crash likelihood.
1Department of Civil, Environmental & Construction Engineering, University of Central Florida, Orlando, FL 32816, USA.
Accident; Analysis and Prevention
|November 29, 2021
Summary
This study developed a hybrid machine learning model to predict secondary crashes in real-time. The model accurately forecasts secondary crash likelihood, enabling proactive traffic safety management and prevention.
Area of Science:
- Traffic Safety
- Machine Learning
- Transportation Engineering
Background:
- Secondary crashes pose significant risks within the impact zones of primary incidents, exacerbating traffic disruptions and safety concerns.
- Existing models often overlook the likelihood of secondary crash occurrence and lack real-time applicability due to feature availability.
- The need for real-time secondary crash prediction models that update frequently (e.g., every minute) is critical for timely interventions.
Purpose of the Study:
- To develop and evaluate a real-time machine learning model for predicting secondary crash likelihood.
- To address the limitations of previous studies by considering both the propensity for a primary crash to cause a secondary one and the independent likelihood of a secondary crash occurring.
- To incorporate real-time traffic flow features for enhanced prediction accuracy.
Main Methods:
- Developed two XGBoost models: one for predicting the likelihood of a primary crash leading to a secondary crash, and another for the likelihood of secondary crash occurrence.
- Proposed a hybrid model integrating the outputs of the two XGBoost models.
- Utilized real-time traffic flow data, including average traffic volume and occupancy, as key features.
Main Results:
- The hybrid model demonstrated significantly improved accuracy in predicting secondary crash likelihood compared to individual models.
- Key features for prediction included real-time traffic flow indicators such as average traffic volume and occupancy.
- The model's real-time prediction capability (5-10 minute forecast, minute-by-minute updates) was validated.
Conclusions:
- The developed hybrid model offers a robust solution for real-time secondary crash likelihood prediction.
- This model has strong potential for integration into proactive traffic safety management systems to prevent secondary crashes.
- Real-time traffic flow data is essential for accurate and timely secondary crash risk assessment.
Related Concept Videos
01:03Prediction Intervals
2.4K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
2.4K
01:18Steps in Outbreak Investigation
253
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
253