Researchers Submit Patent Application, “System And Method For Wildfire Risk Assessment, Mitigation And Monitoring For Building Structures”, for Approval (USPTO 20230023808): Fortress Wildfire Insurance Group

2023 FEB 13 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- From Washington, D.C., NewsRx journalists report that a patent application by the inventors O’Dell, Michael (Alameda, CA, US); Wall, John (Frankford, DE, US), filed on July 12, 2022, was made available online on January 26, 2023.

The patent’s assignee is Fortress Wildfire Insurance Group (Rocklin, California, United States).

News editors obtained the following quote from the background information supplied by the inventors: “Wildfires are an increasingly important factor for homeowners to consider in various parts of the world, including in North America. In high-risk areas wildfires present a real danger of total loss of a building structure, such as a residential home.

“Despite the increasing chance of loss, and the ongoing chance of total loss, approaches to assessing the likelihood of that loss, figuring out how to mitigate that chance, and monitoring the risk over time are not particularly useful due to their lack of precision, lack of actionable data, and unrealistic approaches to implementing and scaling to match the size of the problem. For example, some approaches are simply generic guidelines (like directing the general removal of flammable material without further reasoning or direction) or require an expert on-site to do an evaluation of how to protect a property (which is then highly dependent on the expert and lacks any quantitative analysis of ignition risk to a property based on surrounding environment, fuels, and ignition characteristics of a structure).

“There is accordingly a need in the art for an improved system and method for wildfire risk assessment, mitigation, and monitoring.”

As a supplement to the background information on this patent application, NewsRx correspondents also obtained the inventors’ summary information for this patent application: “In one aspect, a method includes receiving at least one image of a property, where the property includes at least one primary structure. The method also includes identifying structural features of the at least one primary structure by determining an inventory of exterior features of each at least one primary structure on the property, where the features are located along a perimeter of the at least one primary structure and include dimensions and material composition. The method also identifies structural features by utilizing a machine learning feature detection algorithm on each of the at least one image to characterize primary structural features used to model fire susceptibility. The method further includes determining at least one non-primary structure fuel source on the property and surrounding the property by utilizing a machine learning fuel load algorithm for detecting major vegetation including at least one of a tree and a shrub, detecting secondary structures including at least one of a shed and a fence, detecting neighboring rooftops within a distance of the primary structure that would contribute to advancement of a wildfire, and detecting at least one of a footprint of the primary structure, roof characteristics of the primary structure, and local topological features, where the local topological features include at least one of a slope, a road, a hydrant, and an arroyo. The method also includes determining non-primary structure fuel sources on the property and surrounding the property by determining a feature-specific attribute for each of the structural features of the at least one primary structure and for features of each of the non-primary structure fuel sources. The method finally includes determining non-primary structure fuel sources on the property and surrounding the property by calculating a fuel load in terms of thermal energy generation potential utilizing the feature-specific attributes.

“In another aspect, a computing apparatus is disclosed comprising a processor and a memory. The memory stores instructions that, when executed by the processor, configure the apparatus to perform the method described above.”

The claims supplied by the inventors are:

“1. A method comprising: receiving at least one image of a property, wherein the property comprises at least one primary structure; identifying structural features of the at least one primary structure by: determining an inventory of exterior features of each at least one primary structure on the property, wherein the exterior features are located along a perimeter of the at least one primary structure and include dimensions and material composition; utilizing a machine learning feature detection algorithm on each of the at least one image to characterize primary structural features necessary to model fire susceptibility; determining at least one non-primary structure fuel source on the property and surrounding the property by: utilizing a machine learning fuel load algorithm for: detecting major vegetation including at least one of a tree and a shrub; detecting secondary structures including at least one of a shed and a fence; detecting neighboring rooftops within a distance of the primary structure that would contribute to advancement of a wildfire; and detecting at least one of a footprint of the primary structure, roof characteristics of the primary structure, and local topological features, wherein the local topological features include at least one of a slope, a road, a hydrant, and an arroyo; determining a feature-specific attribute for each of the structural features of the at least one primary structure and for features of each of the at least one non-primary structure fuel sources; and calculating a fuel load in terms of thermal energy generation potential utilizing the feature-specific attributes.

“2. The method of claim 1, further comprising: overlaying a multi-dimensional grid on the at least one image of the property, wherein the grid comprises a plurality of tiles and divides the property into analysis points, each analysis point represented by one tile, and wherein a centroid of the primary structure is centered on the grid; encoding each tile within the grid with associated fuel element details and structural element details; and encoding tiles within the grid with associated topographical data.

“3. The method of claim 2, further comprising: determining, when exposed to at least one fuel source, a thermal energy output and a probability of ignition failure for each of the structural features of the at least one primary structure and for each of the at least one non-primary structure fuel sources, utilizing a fire risk algorithm, the fire risk algorithm including: a plurality of threat vectors, comprising: a heat flux for radiant impact threat vector; a flame front contact for direct flame impingement threat vector; an ember mass accumulation and size population for firebrand accumulation threat vector; and an ember penetration probability computation threat vector, wherein the probability of ignition failure for each of the plurality of threat vectors comprises utilizing the structural features of the at least one primary structure, the features for each of the at least one non-primary structure fuel sources, spatial relationships between the at least one primary structure and the at least one non-primary structure fuel sources, and the feature-specific attributes for each of the structural features of the at least one primary structure and features for each of the at least one non-primary structure fuel sources; a direct evaluation routine to determine whether each structural feature of the at least one primary structure ignites under an influence of each of the plurality of threat vectors summed over all of a plurality of fuel sources with direct access to each structural feature of the at least one primary structure; a line of sight evaluation routine to determine the impact of each of the plurality of fuel sources within the line of sight of each tile including a portion of the primary structure; calculating an ignition failure determination for each structural feature at each tile including a portion of the primary structure, wherein the calculations include the impact of the plurality of fuel sources with direct access to the primary structure and the plurality of fuel sources within the line of sight of each tile including a portion of the primary structure; and an evaluation routine to determine whether each feature of the at least one non-primary structure fuel source ignites under the influence of each of the plurality of threat vectors.

“4. The method of claim 3, further comprising: updating the fire risk algorithm by cataloguing for each tile with an ignition failure, the following: a location of the tile; the at least one structural feature or each feature of the at least one non-primary structure fuel source that failed; each of the plurality of threat vectors that caused each structural feature and each feature of the at least one non-primary structure fuel source to fail; each individual fuel source contributing to each specific threat vector causing a feature failure; wind direction during the feature failure; and a failure surplus for each feature failure, wherein the failure surplus measures an extra heat flux the failed feature experienced over a non-failure or non-ignition state.

“5. The method of claim 3, further comprising: determining an influence of wind on each of the plurality of threat vectors including: calculating an impact of each direction of wind from at least north, south, east, and west compass headings on each of the plurality of threat vectors for each tile including a portion of the primary structure; and calculating the impact of a plurality of wind speeds on each of the plurality of threat vectors for each tile including a portion of the at least one primary structure.

“6. The method of claim 5, further comprising: determining an overall risk assessment for an entire property including: compiling a list of each ignited structural feature of the at least one primary structure and each ignited feature of the at least one non-primary structure fuel source; placing the list in a Failure Mode Effect Analysis (FMEA) framework; quantifying a relative risk of each item in the list in a Risk Priority Number (RPN) based on ignition impact by the tile and energy overage; calculating a cumulative risk score for each of the plurality of threat vectors by evaluating the ignition failures of the at least one primary structure and calculating the energy overage and failure mode by ignition failure; calculating a risk score for the entire property based on threat vector energy contributions to each ignition failure; generating a risk assessment report including risk scores for at least one of each ignited structural feature, each structural feature, each ignited feature of the at least one non-primary structure fuel source, each feature of the at least one non-primary structural fuel sources, and the risk score for the entire property.

“7. The method of claim 6, further comprising: associating a heat flux contribution from multiple fuel sources to at least one ignition failure point; identifying multiple failures in a same structural feature of the primary structure, wherein the multiple failures are caused by the heat flux contribution from the multiple fuel sources; augmenting the FMEA framework to reflect the multiple failures of the same structural feature of the primary structure due to the heat flux from multiple sources; and prioritizing risks based on each of the multiple failures in the same structural feature.

“8. The method of claim 7, further comprising: determining a remediation solution for the at least one ignition failure point, including: applying the prioritized risks to generate a remediation score for the FMEA framework, thereby providing the ability to address remediation solutions in a failure event; incorporating risk prioritization into the FMEA framework, thereby utilizing multiple failure contributions; and generating a remediation report comprising the at least one ignition failure point and including remediation solutions based on the remediation score, wherein the remediation solutions include at least one of hardening the primary structure to ignition and reducing fuel loads surrounding the primary structure.

“9. The method of claim 1, wherein the image is obtained through at least one of oblique satellite imagery, aerial imagery, ground imagery, real estate multiple listing service databases, and images from an application on a mobile device.

“10. The method of claim 1, wherein the machine learning feature detection algorithm characterizes structural features including at least one of a window, a door, a vent, and a soffit.”

There are additional claims. Please visit full patent to read further.

For additional information on this patent application, see: O’Dell, Michael; Wall, John. System And Method For Wildfire Risk Assessment, Mitigation And Monitoring For Building Structures. U.S. Patent Application Number 20230023808, filed July 12, 2022 and posted January 26, 2023. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(20230023808)&db=US-PGPUB&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)