Industrial change is difficult to observe in real time: official statistics and administrative registers often capture closures, conversions, and land-use shifts with substantial lags and at coarse spatial resolution. This contribution proposes an OSM-based Industrial Transition Index (OSM-ITI) that leverages the full edit history of OpenStreetMap (OSM) to detect and time-stamp micro-level transformations of industrial spaces, treating OSM as a socio-technical sensor produced by distributed contributors.
We focus on industrial and production-related features (e.g., landuse=industrial, building=industrial/warehouse) and define “transition events” as changes in status or function recorded in OSM (e.g., industrial → brownfield/abandoned, industrial → retail/commercial, warehouse → residential/office). Using open tools (ohsome/OSHDB, PostGIS, QGIS), we reconstruct transition time series for Italian NUTS3 provinces and selected urban areas, and compute: (i) transition frequency, (ii) median time-to-conversion, and (iii) persistence/lock-in scores consistent with path-dependence interpretations.
We then relate OSM-ITI dynamics to fully open contextual indicators (Eurostat/ISTAT, where available at territorial level) on sectoral employment structure, business demography proxies, and land-use/soil consumption datasets, to assess whether areas with faster OSM-recorded transitions are also those exhibiting measurable structural change. The goal is not to claim OSM as “ground truth”, but to evaluate when OSM history provides earlier, policy-relevant signals than traditional sources.
The contribution is twofold. For applied economics, OSM-ITI provides a new, reproducible indicator to study creative destruction, lock-in, and place-based industrial transition under strict data constraints. For the GFOSS/OSM community, it demonstrates how open geospatial infrastructures can generate interpretable socio-economic signals and releases a fully reproducible workflow (queries, code, documentation) that can be adapted across sectors and countries.