Dynamic Building Loss Factor (dBLF) is a machine learning model that assesses the expected loss of individual buildings based on factors that drive structural vulnerability. It provides insight into not just whether a community is threatened, but the specific types of structures and conditions that carry the highest risk.
Like the Building Loss Factor Index (BLF), dBLF is also per-building, but it is designed to be dynamic at the time of fire impact and to account for structure-to-structure fire spread during urban conflagrations. It analyzes each threatened/damaged building when the fire reaches it, including the surrounding environment/conditions relevant to urban fire spread.
The dBLF evaluates vulnerability using a combination of:
-
Structure characteristics and building age
-
Topography
-
Vegetation
-
Building density and proximity to roads
The dBLF is one of two primary enhancements in Technosylva's urban conflagration modeling capability, alongside Wildland-Urban Interface (WUI) fuel mapping. Together, they address a key limitation of traditional wildfire modeling, which has historically classified urban areas as non-burnable and therefore provided limited visibility into how fire behaves once it reaches populated communities.
For utilities and fire agencies, dBLF outputs help identify which structures within a threatened area are most at risk, enabling more targeted mitigation decisions including asset hardening, undergrounding of lines, vegetation management, and community engagement priorities.