Dynamic Rate Generation via Predictive Model for Rental Insurance

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Rental car insurance often uses a single static standard rate that is not tailored to individual users and lacks incentives, resulting in a lack of control and personalized pricing.

Innovation Solution

A predictive model system that uses transaction data, including mileage and rental information, to generate a dynamic rate based on user-specific parameters such as location, date, time, name, and age, adjusting the rate in real-time and accounting for historical data on accidents and rental history.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a single static standard rate is used for rental car insurance, then the pricing process is simple and fast, but the rate is not tailored to individual users and lacks personalization

Engineering Contradiction:

Improvepricing process simplicityVSAvoidrate personalization

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent transforms the static standard rate into a dynamic rate that automatically adjusts based on user-specific parameters. The system collects data including location, date, time, name, and age, then uses a predictive model to generate personalized rates in real-time, making the pricing both adaptive and operationally efficient

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes multiple parameters simultaneously to create personalized rates. It incorporates user-specific parameters (location, date, time, name, age) and historical data parameters (accident history, rental history) into the predictive model, allowing the rate to be customized for each individual while maintaining automated processing

Inventive Principle:

Principle #35Parameter changes

2Device complexity

If a static standard rate is used, then the system is simple to maintain, but users lack control and incentives related to their rate

Engineering Contradiction:

Improvesystem complexityVSAvoiduser control and incentive selection

Core Design Contradiction:

Device complexityVSEase of operation

Solution Approach 1:

The patent implements a feedback mechanism where user behavior and historical data influence future rates. The predictive model analyzes accident history, rental history, and other user-specific parameters to generate rates that reflect individual risk profiles, giving users control and incentives based on their driving behavior and history

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system enables users to effectively serve themselves by automatically generating personalized rates based on their own data. The predictive model uses user-specific parameters and historical information to create customized rates without requiring manual intervention, giving users control over their insurance pricing

Inventive Principle:

Principle #25Self-service

3Measurement precision

If transaction data including mileage and rental information is collected and processed, then the predictive model accuracy is improved, but the data processing complexity increases

Engineering Contradiction:

Improvepredictive model accuracyVSAvoiddata processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple data sources and parameters into a unified predictive model. It combines transaction data (mileage, rental information) with user-specific parameters (location, date, time, name, age) and historical data (accident history, rental history) into a single model that generates personalized rates, achieving high accuracy while managing complexity through integration

Inventive Principle:

Principle #5Merging (Combining)

4Adaptability or versatility

If a predictive model is trained and updated with user data, then the dynamic rate customization is improved, but the computational resources and time required increase

Engineering Contradiction:

Improverate customizationVSAvoidmodel training and updating time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by pre-collecting and organizing user data including transaction information, accident history, and rental history. The predictive model is trained in advance on this accumulated data, so when a user requests a rate, the system can quickly generate personalized rates without extensive real-time processing, reducing the time loss while maintaining high customization

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12141875B2Systems and methods for predictive model generation

Publication Date: 2024.11.12 CAPITAL ONE SERVICES LLC

AI summary

A predictive model system may include an application comprising instructions for execution on a device including one or more processors coupled to memory, and one or more servers in data communication with the application. The one or more servers may be configured to receive a first set of data which may be updated to include transaction data and at least one selected from the group of mileage rental information; create a predictive model based on the first set of data; transmit, to the application, one or more requests including a second set of data, the second set of data comprising at least one selected from the group of location information, date, time, name, and age of a user; update the predictive model based on the received second set of data received; generate a dynamic rate based on the updated predictive model; and transmit, to the application, the dynamic rate.