Related Experiment Video
Updated: Jul 31, 2026

Evaluation of an Exclusive Spur Dike U-Turn Design with Radar-Collected Data and Simulation
Published on: February 1, 2020
Modeling accident occurrence at signalized tee intersections with special emphasis on excess zeros
1Department of Civil Engineering, National University of Singapore, Singapore. engp0436@nus.edu.sg
This study uses a zero-inflated negative binomial model to analyze accident data from Singapore intersections. It identifies key geometric and traffic factors influencing crash risk, distinguishing true safety from random chance.
Area of Science:
- Traffic Engineering and Safety
- Transportation Science
- Statistical Modeling in Road Safety
Background:
- Identifying hazardous intersection factors is crucial for effective remedial treatments.
- Distinguishing truly safe intersections from those with no accidents due to random chance is a common challenge in accident studies.
- Accident data analysis often requires specialized models to handle excess zero-accident records.
Purpose of the Study:
- To develop a statistical model for identifying geometric and traffic factors contributing to accident occurrence at signalized intersections.
- To address the issue of 'excess zeros' in accident data by employing a zero-inflated negative binomial model.
- To analyze accident data from 104 signalized tee intersections in Singapore over a 9-year period.
Main Methods:
- Utilized accident data from 104 signalized tee intersections in Singapore spanning 9 years.
- Employed the zero-inflated negative binomial (ZINB) model to account for excess zero-accident records.
- Analyzed the probability of accident outcomes based on various geometric and traffic variables.
Main Results:
- Factors increasing accident occurrence include uncontrolled left-turn slip roads, permissive right-turn phases, horizontal curves, short sight distances, numerous signal phases, and high total/left-turn approach volumes.
- Factors reducing accident occurrence include right-turn channelization, left-turn lane acceleration sections, median railings, and gradients over 5%.
- A discernible trend of decreasing accidents over the study period was observed.
Conclusions:
- The zero-inflated negative binomial model effectively distinguishes between inherently safe intersections and those with no accidents due to chance.
- Specific geometric designs and traffic conditions significantly influence accident risk at signalized tee intersections.
- Findings provide valuable insights for targeted safety improvements and urban traffic management strategies in Singapore.
Related Concept Videos
Routh-Hurwitz Criterion II
The first scenario occurs when a singular zero appears in the first column of the Routh table. This situation creates a division by zero issues. To resolve this, a small positive or negative number, denoted as epsilon (∈), is substituted for the zero. The stability analysis proceeds by assuming a sign for ∈. If ∈ is positive, any sign change in the first column of the Routh...
Introduction to Horizontal Curves
Introduction to Vertical Curves
Vertical Curve: Problem Solving
Elevation of Intermediate Points on Vertical Curves
Real Zeros of Polynomials

