Roof Condition

The Roof Condition pack provides detailed information about the state of repair of a roof - both short term damage due to catastrophe, and longer term degradation.This AI Pack can be purchased as an add-on to the Roof Characteristics pack. It adds AI Layers and AI Parcel outputs related to the condition of the roof. It operates in a similar way to the Roof Shape, in that overall area (percentage and absolute) is calculated as an attribute of the roof, as well as the actual polygon features being available for the relevant sections.

Examples of roof condition several months post-catastrophe. Area and confidence of each damage type can indicate the severity.


Roof Condition includes a group of five AI Layers:

The layers are defined in the AI Layer Glossary page. Each of these layers has been intersected with the "Roof" feature class, such that each roof has attributes relating to roof condition pre-calculated for each layer (area, ratio and confidence). Each layer is also present as a vector layer in its own right of polygons representing each piece of damaged roof.

Characteristics and Recommended Use

There are two main scenarios in which roof condition is useful. The slow degradation over time of roof quality, indicating risk and need of repair. This can be assessed through ponding (on flat roofs), rusting (on metal roofs) and discolouration (on tile and shingle roofs). A roof exhibiting structural damage for an extended period of time is likely to be uninhabitable, due to water leakage.

The second scenario is more immediate impact of catastrophes such as wind and storms. This is primarily indicated by the presence of structural damage. In the months after a catastrophic event, it is possible to track over successive surveys the rate at which structurally damaged roofs become temporarily repaired, and temporarily repaired roofs are renewed with roofs in good condition.

Version History 

For a precise change log, refer to the AI Generations  pages. This is a convenient summary to describe changes to this AI Pack over time:

Gen 1-3

  • Not available.

Gen 4

gen4-building_storm-1.0  - Roof Condition introduced in raster model, vectorised features, and attributes of roof.

gen4-lightning_bolt-1.0  - Added a “low confidence roof condition” attribute for each of the five roof condition layers. This is in addition to the existing, unchanged standard roof condition vectors and attributes. The low confidence roof condition will describe a larger area, and a lower confidence than the standard one. Typically roofs degrade gradually; this makes it easier to identify new roofs without any weathering (for example, for tile and shingle staining), but has a much higher rate of false positives.