Case Study
Predicting and Preventing Large Loss with Aioi Nissay Dowa Europe
Take preventative action to significantly reduce costs whilst potentially preventing severe injuries and even saving lives.
of riskiest drivers identified that contribute 50% of large losses.
estimated savings each year in claims cost by preventing potential large losses.
of claims costs attributable to 5% riskiest drivers.
Problem:
The identification of large losses within the global insurance market is a formidable challenge, primarily due to the industry's inherent complexities. These complexities stem from the diversity of insured risks, the dynamic nature of risks, and the infrequent yet severe nature of large losses. Additionally, data quality issues, fragmented information, and the global reach of insurers further compound the problem.
To address this challenge effectively, insurers need innovative solutions and advanced analytics tools to enhance early identification methods. Doing so is crucial not only to safeguard individual insurers from substantial financial impacts but also to ensure the stability and resilience of the entire global insurance market.
Solution:
Mind Foundry has built a unique solution to predict and prevent large losses (of over £50,000) based on Aioi Nissay Dowa Europe's (AND-E) specific requirements that could not be met by any off-the-shelf solutions.
We combined a number of different data sources, including driving behaviour data, geospatial data, and environmental data, into a single Machine Learning solution. The solution looks at the likelihood of an accident occurring on specific routes taken.
Machine Learning models are used to analyse continuous behavioural and historical driving data, looking specifically at customers driving patterns, like speeding and braking, as well as road familiarity, because the data shows that drivers are more likely in certain circumstances to have an accident on roads they are more familiar with. This is then paired with road risk data, such as how many accidents there are on the M4 each year. The models are then aggregated to provide an all-inclusive risk score.
As new patterns emerge, using a Continuous Metalearning capability, the solution automatically and continuously integrates new data back into the model to improve performance. New risks can be effectively governed and analysed by human operators, while enabling the model to learn new types of large loss, meaning no manual retraining is needed to continuously adapt, optimise operations, and learn with AI.
Results:
The solution empowers AND-E to target the right customers with preventative safety campaigns. If even one large loss is prevented, this could mean saving lives.
Together we have:
-
Identified the top 5% of riskiest drivers that contribute 50% of large losses.
-
Saved an estimated £2m+ each year in claims cost by preventing potential large losses.
-
Enhanced the driver feedback process to help high risk drivers.
AND-E and Mind Foundry will together continue to explore how we can embed data driven insights into our pricing solutions to improve the overall experience and value for our customers.
"What was critical in the success of the large loss solution, was that AND-E had the domain expertise and previous knowledge about large loss accidents, proven and enhanced by the Machine Learning models created by Mind Foundry. As human insight helped inform what the models learned, the model’s insights helped the insurance experts learn more about the world and how to understand and better manage the risk in their portfolio."
Greg Cole
UK Claims Director
Contact us for more information.
4 min read
AI Insights from Insurance Leaders in 2024
by Mind Foundry
8 min read
AI vs Fraud in the UK and Japan
by Nick Sherman
Stay connected
News, announcements, and blogs about AI in high-stakes applications.