Kernel density estimation and K-means clustering to profile road accident hotspots

Area of Science:

• Urban planning
• Transportation safety
• Spatial analysis

Background:

• Effective road safety strategies require accurate identification of accident hotspots.
• Understanding spatial patterns of road accidents is crucial for targeted interventions.

Purpose of the Study:

• To develop and apply a methodology for identifying and classifying road accident hotspots in London, UK.
• To utilize Geographical Information Systems (GIS) and clustering techniques for spatial analysis of road accidents.

Main Methods:

• Kernel Density Estimation (KDE) was employed to map the spatial distribution of injury-related road accidents.
• K-means clustering was used to classify hotspots based on environmental data and accident patterns.
• Road accident data from London (1999-2003) and appended environmental data were utilized.

Main Results:

• A methodology combining GIS and KDE was developed to identify basic spatial units of accident hotspots.
• Five distinct groups and 15 clusters of road accident hotspots were identified using K-means clustering.
• The classification provides a robust framework for understanding and addressing accident-prone areas.

Conclusions:

• The developed methodology offers a systematic approach to classifying road accident hotspots.
• The identified clusters can inform targeted road safety campaigns and interventions.
• This spatial analysis contributes to evidence-based urban planning for enhanced road safety.