Machine Learning in US Tornado Claims | News | Brit

Brit Deploys Machine Learning Algorithm to Expedite US Tornado Claims

Brit Ltd (“Brit”) is pleased to announce the expanded use of a proprietary machine learning algorithm to accelerate the identification of US Tornado property damage, to be used in tandem with Brit’s access to ultra-high-resolution imagery.

Used by the Brit Claims team and its Delegated Claims Adjusters in the wake of Hurricane Ida, Brit has subsequently deployed this innovation to expedite the identification of insured property damage in response to catastrophic tornadoes in 2021.

The machine learning algorithm, developed by Brit’s Data Science team, assesses ultra-high-resolution aerial images and data, pinpointing, color-coding, and displaying properties by damage classification in the aftermath of catastrophic events. The technology enables Brit’s Claims team to proactively identify, triage and assign response activity even before the claims are reported.

Brit has been working successfully with the Geospatial Insurance Consortium (GIC) since April 2019, a non-profit organisation that captures best in class post-event aerial imagery for first responders and insurance companies. Utilising the GIC images and Brit’s machine learning algorithm, Brit’s Claims team has a virtual claims adjusting platform that can expedite claims payments in locations that cannot be immediately serviced by local field adjusters in the initial days following catastrophe events.

Mike Barry, Head of Global Property Claims for Brit, commented, “Innovation is a central pillar to Brit’s Claims strategy, and this includes a number of virtual and digital claims solutions for our customers when they need us the most.  We continue to evolve and innovate the use of our best in class aerial imagery and the damage classification algorithm developed earlier this year, enabling quicker and more efficient identification and payment of claims.”


Goldeneye Video Cover

Using machine learning algorithms for a faster claims response

Watch this video