Fiji Road Surface, Width, and Passability for 800 km of Arterial Roads

A collection of AI-derived road surface types, width classes, and passability scores for approximately 800 km of arterial roads in Fiji, developed by HeiGIT. It utilizes PlanetScope satellite imagery from 2020 and 2024 to supplement OpenStreetMap data with deep-learning predictions that fill gaps in existing surface tags.

Use Cases

• Logistics planning using the Humanitarian Passability Index (HPI) to identify high-risk blockages
• Infrastructure monitoring by tracking transitions in the Road Surface Change column
• Route optimization for heavy vehicles using Road Width Class 3 segments

Strengths

• 89.2% surface prediction accuracy
• Multi-temporal change detection between 2020 and 2024
• Fills surface tag gaps for 30% of Fiji's arterial road network

Limitations

• Geographic scope limited to arterial road classes only
• Width data is categorical (3 classes) rather than continuous measurements
• AI-derived predictions may contain noise compared to ground-level surveys

Provenance

Source

HeiGIT (Heidelberg Institute for Geoinformation Technology)

Collection Method

Deep learning analysis of PlanetScope satellite imagery integrated with OpenStreetMap data

Time Range

2020 - 2024

Freshness

Last updated March 2026; temporal coverage spans 2020 to 2024.

Geography

Fiji