OpenStreetMap relies on a wide variety of contributions in terms of mapped themes, covered areas, and completeness levels. This flexibility is key to its richness and success. However, it also generates a high degree of data heterogeneity, which is often cited by its detractors or those seeking to minimize its value.
Some services, such as dashboard.ohsome.org and github.com/hotosm/osm-analytics, have attempted to address these concerns about heterogeneity, but they seem to have been discontinued.
This presentation introduces a new approach that analyzes not territories as a whole or a particular theme, but rather the various components that comprise the backbone of homogeneous OpenStreetMap (OSM) data. These components include named places that are linked together and have a known extent. They are also provided with a road grid, essential points of interest, and main buildings. Finally, there are complementary points of interest.

To analyze cities and towns, this approach evaluates what defines their urban areas, such as land use zones and street networks, according to various indicators, such as age of update, geometric accuracy, and coverage. It also provides a quality or completeness index ranging from 1 to 10 with six colors ranging from red to dark green.
Based on these urban areas, the approach also provides a completeness index for the points of interest located there. This index is categorized by type (health, education, other amenities, shops, offices) and surface of the urban area.

These indices highlight significant contrasts between regions and between small towns in the same region. The visual approach to analysis is complemented by the ability to enhance data for objects that can be edited using imagery : a click on the map provides an opening link to JOSM. For objects that require completion in the field, we will present a workflow that uses these analysis layers and OSM applications for Android.

Technically, this approach does not use a dedicated backend served by a separate web service on the frontend. Instead, analyses are automatically produced and shared as thematic maps by country via a spatial data infrastructure (SDI) based on geOrchestra. Each produced analysis takes the form of freely accessible OGC layers (WMS/WFS) and can be used by any OGC client or downloaded in various GIS formats.