Using AI to assess minor damage in motor accidents Qorus-NTT DATA Innovation in Insurance Awards 2025- Nominated
BrazilCategory
Product & Service InnovationKeyword
AI & Generative AI, Car & Mobility insurance, Claims managementBusiness Line
Accident InsuranceDistribution Channel
Online / Direct
Innovation presentation
This pilot project has successfully implemented Artificial Intelligence (AI) in the assessment of insured vehicle claims at MAPFRE Brazil. As this is a complex business process, the focus was on applying the AI solution in cases of minor damage, where it has managed to streamline the vehicle damage assessment process. It has also improved the efficiency of the claims adjusters, who limited supervised the assessment completely via AI, thereby reducing the assessment and claim handling times, as well as improving the customer experience.
The flow begins in a web app that facilitates guided capture for customers to quickly send photos of their damaged vehicle from their smartphone, simplifying this process. Next, the images are analyzed to classify the severity of the damage, determining whether it is minor, medium, serious or a total loss. In cases of minor damage, the analysis is carried out using AI to identify the existence of damage, the type of repair the vehicle requires, the spare parts needed and the costs of repair or replacement, delivering a complete and automatic appraisal that is overseen and complemented by the expert assessor where necessary.
The tool combines the use of deep neural networks applied to image processing, a connection with third-party systems that provide vehicle data and integration with MAPFRE's accident and appraisal systems. Benefits:
• Application of cutting-edge technologies that bring precision and speed to the claims assessment process, automating the damage assessment process. Thanks to the automatic AI assessment, the initial costs of claims with partial losses are known early on. • Achieving greater efficiency in the use of enterprise resources, freeing up claims adjusters to focus their time on more complex claims. • Radical transformation of the customer experience, making users’ lives easier by offering a personalized, immediate response that is easily accessible on their smartphone, without having to travel anywhere, with clear, easy-to-follow guidance and instructions. • Aligning the insurance world to a new digital context that requires immediacy in transactions. • Creating new digital methods of offering services to our clients, ensuring transparency in the process at all times and reinforcing people's trust in MAPFRE throughout the entire user experience.
Results: • Customer satisfaction: 4.6 out of 5 stars in the ratings for the webapp for sending photos of the accident, and +71.05 points obtained in the Net Promoter Score (NPS®). • The SLA for the time used by the AI to perform the damage assessment was 20 seconds. • Automation rate: For parts within the AI detection range, 50% of minor damage claims resulted in full automation of the appraisal using AI, and the other 50% had partial automation. • Reduction in the claim cycle time: the average life of minor damage claims assessed via AI went from 32 to 26 days on average, saving 6 days. It is important to note that MAPFRE Brazil only closes out the claim process fully once the vehicle repair has been completed.
Interested in learning more?
Qorus has a library of almost 8,000 innovation case studies across critical areas like customer experience, sustainability, marketing & distribution and more that can be used to inform your decision-making.