Brit develops ML algorithm for catastrophe claims

Brit has launched a proprietary machine learning algorithm designed to accelerate the identification of post-catastrophe property damage using ultra-high-resolution imagery.

The proof of concept is being used by the insurer’s claims team and its delegated claims adjusters in the wake of Hurricane Ida.

Sheel Sawhney, group head of claims and operations, said: “A claim is the single most important interaction that an end client will have with their insurer and this will often be at a time of significant difficultly. We are therefore continually focused on improving the service we offer and how quickly we can provide resolution for our customers. Innovation and technology are critical to the equation. This use of machine learning techniques and the best available imagery is further evidence of how our award-winning claims team is finding new ways to increase the speed and accuracy of claims payments.”

To develop the new tool, Brit’s Data Science team created and overlaid a machine learning algorithm to access the ultra-high-resolution ariel images and data such that it pinpoints, color-codes, and displays properties by damage classification within days after a catastrophe event. This enables the team to proactively identify, triage and assign response activity before claims are reported.

The insurer has been working with the Geospatial Insurance Consortium since April 2019, a non-profit organisation that captures best in class post-event ariel imagery for first responders and insurance companies.

    Share Story:


Cyber risk in the transportation industry
The connected nature of the transport and logistics industries makes them an attractive target for hackers, with potentially disruptive and costly consequences. Between June 2020 and June 2021, the transportation industry saw an 186% increase in weekly ransomware attacks. At the same time, regulations and cyber security standards are lacking – creating weak postures across the board. This podcast explores the key risks. Published April 2022.

Political risk: A fresh perspective
CIR’s editor, Deborah Ritchie speaks with head of PCS at Verisk, Tom Johansmeyer about the confluence of political, nat cat and pandemic risks in a world that is becoming an increasingly risky place in which to do business. Published February 2022.