Related Experiment Video
Updated: Dec 3, 2025

Evaluating the Effect of Roadside Parking on a Dual-Direction Urban Street
Published on: January 20, 2023
Investigating exposure measures and functional forms in urban and suburban intersection safety performance functions
Kai Wang1, Shanshan Zhao1, Eric Jackson1
1Connecticut Transportation Safety Research Center, Connecticut Transportation Institute, University of Connecticut, 270 Middle Turnpike, Unit 5202, Storrs, CT 06269-5202, USA.
This study introduces the Generalized Negative Binomial-P (GNB-P) model to improve crash prediction at intersections. It identifies optimal exposure measures and functional forms for Safety Performance Functions (SPFs), enhancing traffic safety analysis.
Area of Science:
- Traffic Engineering and Safety
- Transportation Research
- Statistical Modeling in Transportation
Background:
- Accurate prediction of intersection crashes is vital for effective safety interventions.
- Traditional Safety Performance Functions (SPFs) often struggle with complex relationships between exposure and crash types.
- The choice of exposure measure and functional form significantly impacts SPF reliability.
Purpose of the Study:
- To propose and evaluate the Generalized Negative Binomial-P (GNB-P) model for intersection crash prediction.
- To identify the most reliable exposure measures (major road AADT, minor road AADT, total AADT) for different crash types.
- To determine the most appropriate functional forms (Power, Hoerl 1, Hoerl 2) for Negative Binomial (NB) models.
Main Methods:
- Development and application of the GNB-P model to analyze crash data for stop-controlled and signalized intersections.
- Estimation of three SPF functional forms using various exposure measures for different crash types.
- Modeling over-dispersion using exposure measures to account for crash data variability.
Main Results:
- The mean-variance structure of NB models and over-dispersion vary significantly by crash data and intersection type.
- Minor road AADT positively correlates with over-dispersion for Same-Direction Crashes (SDC), Intersecting-Direction Crashes (IDC), and Single-Vehicle Crashes (SVC).
- Specific functional forms and exposure measures demonstrated superior performance for different crash types (e.g., Power function for Opposite-Direction Crashes (ODC), Hoerl function 2 for SVC, Hoerl function 1 for SDC and IDC).
Conclusions:
- The GNB-P model provides a more robust approach to modeling intersection crash counts and their relationship with exposure measures.
- Estimating over-dispersion using exposure measures leads to more reliable SPF predictions.
- The study offers guidance on selecting optimal exposure measures and functional forms for intersection SPFs based on crash type.
Related Concept Videos
Mechanistic Models: Compartment Models in Individual and Population Analysis
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
Parametric Survival Analysis: Weibull and Exponential Methods
Weibull Distribution
The Weibull distribution is a flexible model used in parametric survival analysis. It can handle both increasing and decreasing hazard rates, depending on its shape parameter...
Strategies for Assessing and Addressing Confounding
Confounding can be addressed at both the design phase of a study and through analytical methods after data...
Hazard Rate
Analysis of Population Pharmacokinetic Data

