ETTC Ranges for Freeway Conflict Types from 12 Chinese Interchanges

A dataset of Extended Time-to-Collision (ETTC) ranges for longitudinal and lateral vehicle conflicts, derived from 12 interchange diverging areas on two multilane freeways in China. The data was created by Feng Tang using image recognition to extract 48 vehicle motion parameters and surrogate safety measures, with conflict labels at 30-second intervals. It was last updated on April 29, 2026, and is shared under a CC-BY-4.0 license.

Use Cases

• Training machine learning models for crash risk prediction based on surrogate safety measures like Modified Time-to-Collision.
• Analyzing spatiotemporal patterns of severe lateral and longitudinal conflicts near exit ramps.
• Interpreting key risk factor contributions for crash types using SHAP analysis frameworks.
• Designing dangerous driving warning systems for freeway diverging areas.

Strengths

• Includes 48 vehicle motion parameters and surrogate safety measures extracted via image recognition.
• Provides conflict labels at a refined 30-second interval for temporal analysis.
• Covers 12 specific interchange diverging areas from two multilane freeways, offering a defined spatial scope.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 5.5 KB, indicating limited scope or a highly aggregated summary.

Provenance

Source

figshare, author Feng Tang.

Collection Method

Data was generated by analyzing video from 12 interchange areas using image recognition technology to extract motion parameters and surrogate safety measures.

Time Range

The temporal resolution of the conflict data is 30-second intervals, but the overall collection period is not specified.

Freshness

Last updated 2026-04-29 17:54:59; freshness should be verified.

Geography

Data is from two multilane freeways in China.