<oai_dc:dc xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd"><dc:creator>HeiGIT</dc:creator><dc:description>This dataset provides AI-derived road surface type, road width, road surface change (unpaved -&gt; paved), and a Humanitarian Passability Index (HPI) using PlanetScope satellite imagery (2020 &amp; 2024). The analysis focuses on the global arterial road network, defined as OSM highways classified as motorway, trunk, primary, and secondary, including their associated _link classes. These road types form the backbone of long-distance, high-capacity transportation and national connectivity. **This dataset supports humanitarian logistics, climate resilience, transportation planning, SDG monitoring, infrastructure investment, and large-scale accessibility assessments.** For more details on methods, refer to the following paper: TBA ### Relationship to the Previous Mapillary-Based Dataset HeiGIT previously released a global dataset providing AI-derived surface attributes from Mapillary street-level imagery. - The Mapillary dataset provides detailed, ground-level information but is limited by uneven imagery availability and regional coverage gaps. - The new PlanetScope-based dataset provides near-complete and globally consistent coverage along arterial roads, enabling continental and national-scale planning. ### Accuracy &amp; Data Quality The PlanetScope surface model achieved 89.2% accuracy, significantly outperforming OSM surface tags (64.7%). This improvement reflects rapid infrastructure development and inconsistencies in OSM mapping of surface conditions in many regions. ### Summary of Road Statistics Approximately 0.0011 million km of arterial roads are mapped in OSM in this region. - Paved roads: 0.0011 million km (97.8388%) - Unpaved roads: 0.0 million km (0.0%) - Missing OSM surface tags: 0.0001 million km (6.8536%) - PlanetScope predictions cover: 0.0011 million km (1427.5625%) of missing data ### AI-Derived Attributes Included - **DL Surface Prediction 2020** - Paved / Unpaved / Unknown (PlanetScope model) - **DL Surface Prediction 2024** - Paved / Unpaved / Unknown - **Road Surface Change (2020-2024)** - transitions from unpaved -&gt; paved - **Road Width Class** - Categorized into 3 classes - **Humanitarian Passability Index (HPI)** - Combines surface type and road width ### Road Width Class Summary - **Class 1 (&lt;3.5 m)** - Light vehicles only; high blockage risk - **Class 2 (3.5-5.5 m)** - Single-lane heavy truck; alternating flow - **Class 3 (&gt;5.5 m)** - Two-lane heavy truck; unimpeded flow ### Humanitarian Passability Index (HPI) - **HPI Alphanumeric code** (e.g., P3, U1) - **HPI Subtype** (e.g., PAV_DUAL, UNP_LIGHT) - **HPI Score 1-6** (1 = highest passability, 6 = lowest) ### HPI Class Summary - **Paved, Two-Lane (&gt; 5.5 m)** - Primary Corridor: All-weather, high-capacity supply route. Codes: P3 / PAV_DUAL / Score: 1 - **Paved, Single-Lane (3.5-5.5 m)** - Reliable Chokepoint: Slow, but passable in most weather conditions. Codes: P2 / PAV_SINGLE / Score: 2 - **Paved, Light Vehicle (&lt; 3.5 m)** - Limited Access Route: Usable by small vehicles; impassable for heavy trucks. Codes: P1 / PAV_LIGHT / Score: 4 - **Unpaved, Two-Lane (&gt; 5.5 m)** - Vulnerable Corridor: High capacity but at risk of rapid degradation in adverse weather. Codes: U3 / UNP_DUAL / Score: 3 - **Unpaved, Single-Lane (3.5-5.5 m)** - High-Risk Chokepoint: Dry season only; highly vulnerable to weather. Codes: U2 / UNP_SINGLE / Score: 5 - **Unpaved, Light Vehicle (&lt; 3.5 m)** - High-Risk Track: Dry season only, 4×4 required. High risk of impassability. Codes: U1 / UNP_LIGHT / Score: 6 ### Included OSM Attributes - Road Segment ID - highway classification - surface tag - OSM road length (meters) ### Geographic Attributes - country_iso_a2 - continent This dataset integrates OSM attributes with deep-learning predictions generated from PlanetScope imagery (2020 and 2024) using HeiGIT's large-scale geospatial AI pipelines. Explore more HeiGIT datasets on HDX: https://data.humdata.org/organization/heidelberg-institute-for-geoinformation-technology More information: https://heigit.org/ We welcome feedback and use-cases. Contact us at: communications@heigit.org</dc:description><dc:identifier>brunei-darussalam-planet-road-surface-data</dc:identifier><dc:publisher>Humanitarian Data Exchange</dc:publisher><dc:rights>Public</dc:rights><dc:title>Brunei Darussalam: Planet Road Surface, Width, and Passability Data</dc:title><dc:type>Datasets</dc:type><dc:coverage>Brunei Darussalam</dc:coverage><dc:date>2020-06-01 to 2024-12-31</dc:date></oai_dc:dc>