Vehicle Data Collection for Usage-Based Insurance Risk Events

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing UBI systems face challenges in accurately predicting driver risk due to large amounts of vehicle data that are difficult to transmit and store, noisy or biased data from ADAS signals, and the difficulty in generating new aggregation metrics that capture driving quality, leading to poor performance and customer resistance to continuous monitoring.

Innovation Solution

A method using a multi-arm bandit approach with delayed feedback to collect high-fidelity data points from vehicles, incorporating unstructured data sources like image data, and adjusting risk metrics based on actual claim data to improve risk prediction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If continuous monitoring and transmission of all vehicle data is implemented, then comprehensive risk assessment is improved, but data transmission and storage burden increases significantly

Engineering Contradiction:

Improverisk assessment accuracyVSAvoiddata volume

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent segments vehicle data into two categories: aggregated data points for continuous monitoring and high-fidelity data points for specific events. The system transmits aggregated data continuously and only transmits high-fidelity data when risk events are detected, reducing overall data transmission volume while maintaining comprehensive risk assessment capability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system implements partial monitoring by selectively collecting high-fidelity data only when risk events occur, rather than continuously transmitting all vehicle data. This approach uses a multi-arm bandit algorithm to determine when to switch between aggregated and high-fidelity data collection based on risk thresholds.

Inventive Principle:

Principle #16Partial or excessive action

2Productivity

If aggregated data is used for UBI calculation, then data transmission efficiency is improved, but measurement accuracy of driving quality deteriorates

Engineering Contradiction:

Improvedata processing efficiencyVSAvoiddriving quality measurement

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system dynamically adjusts the level of data collection based on risk conditions. When risk events are detected, the system transitions from aggregated data collection to high-fidelity data collection, ensuring accurate driving quality measurement when needed while maintaining processing efficiency during normal operations.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system uses feedback from the multi-arm bandit algorithm to determine when to switch between aggregated and high-fidelity data collection. The algorithm learns from delayed feedback (claim outcomes) to optimize the balance between data transmission efficiency and measurement accuracy over time.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If high-fidelity data is continuously collected, then risk prediction accuracy is improved, but energy consumption and data storage requirements increase

Engineering Contradiction:

Improverisk prediction accuracyVSAvoiddata collection energy

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system implements periodic switching between aggregated and high-fidelity data collection based on risk event detection. High-fidelity data collection is activated periodically only when risk events occur, rather than continuously, reducing energy consumption while maintaining risk prediction accuracy when needed.

Inventive Principle:

Principle #19Periodic action

4Adaptability or versatility

If risk metrics are updated frequently, then adaptability to changing driving patterns is improved, but system complexity increases

Engineering Contradiction:

Improverisk model adaptabilityVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system uses self-service through the multi-arm bandit algorithm that automatically updates risk metrics based on incoming data and delayed feedback from claims. The algorithm autonomously learns and adapts to changing driving patterns without requiring manual intervention, maintaining adaptability while managing system complexity.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20260057451A1Automatic data collection of high-fidelity vehicle data for usage based insurance

Publication Date: 2026.02.26 FORD GLOBAL TECH LLC

AI summary

Assessing driver risk for usage-based insurance (UBI) is provided. Risk metrics are sent, to a plurality of vehicles from a cloud server, defining indications of driving events that may occur on the vehicles that are indicative of an effect on UBI rates for the vehicles. The cloud server receives, from the vehicles, aggregated data points collected by the vehicles use in determining UBI rate and high-fidelity data points collected by the vehicles responsive to occurrence of the driving events. The high-fidelity data points are analyzed using a vehicle data service to determine updated risk metrics using a multi-arm bandit approach with delayed feedback based on a risk model and claim data indicative of actual claims made with respect to the plurality of vehicles. The updated risk metrics are provided to the plurality of vehicles to aid in detection of the driving events that affect the UBI rates.