Electric Vehicle Damage Modeling With Combustion-Engine Baselines

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems struggle to accurately predict and measure the damage potential and risk of electrically powered vehicles due to the lack of sufficient data and different driving characteristics compared to combustion engine vehicles, leading to increased uncertainty and inefficiency in insurance and risk assessment.

Innovation Solution

An automated system and method that utilizes a forward-looking modeling structure to combine measurable parameters from combustion engine vehicles with electric vehicle-specific factors, applying modification factors to predict damage potential and risk based on historical data and machine learning algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional damage prediction systems are used for electric vehicles, then the system structure remains simple, but the measurement precision and reliability of damage prediction deteriorate due to lack of sufficient EV-specific data

Engineering Contradiction:
Improvedamage prediction accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the damage prediction task into two distinct modules: a base model trained on combustion engine vehicle data and an EV-specific adjustment module that applies modification factors to adapt the base predictions to electric vehicle characteristics. This segmentation allows the system to leverage existing robust data while incorporating EV-specific adjustments, thereby improving measurement precision without requiring a complete redesign of the entire prediction system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces modification factors as an intermediary element that bridges the gap between combustion engine vehicle data and electric vehicle damage characteristics. These modification factors serve as adjustable parameters that translate base model predictions into EV-specific predictions, enabling accurate damage assessment for electric vehicles without needing extensive EV crash data from scratch.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If EV-specific damage data is collected and used, then the reliability of damage prediction improves, but the loss of time and resources for data collection and processing increases

Engineering Contradiction:
Improvedamage prediction reliabilityVSAvoiddata collection time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-calculating and storing modification factors for various electric vehicle types and configurations before actual damage prediction is needed. These modification factors are derived from available EV characteristics and crash test data in advance, so that when a real-world EV damage case needs assessment, the system can immediately apply the pre-prepared adjustment factors without needing to conduct new data collection or analysis at the time of prediction.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If modification factors are applied to adapt combustion engine vehicle data to electric vehicles, then the adaptability of the system improves, but the device complexity increases due to additional calculation layers

Engineering Contradiction:
Improvesystem adaptabilityVSAvoidcalculation complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system applies parameter changes by modifying key prediction parameters through multiplication or adjustment by modification factors. Instead of fundamentally changing the prediction algorithm, the system adjusts specific output parameters (such as expected damage costs, repair time, or component failure probability) by applying EV-specific modification factors to the base model predictions. This approach maintains the core calculation structure while adapting the results to electric vehicle characteristics.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4384930B1Automated method and system for determining an expected damage of an electrically powered vehicle
Publication Date: 2025.10.08 SWISS REINSURANCE CO LTD
  • EP4384930B1 patent drawingFigure 1
  • EP4384930B1 patent drawingFigure 2
  • EP4384930B1 patent drawingFigure 3

AI summary

Proposed is an automated system and method for measuring and/or allocating risk measures of loss for a replacement cost of an electric vehicle and/or parts of the electric vehicle. The system based on a forward looking model structure for electric vehicles (FLM-EV) returns rating modification factors for MTPL and MOD for electric vehicles. The input consists of either car make, model year or rating parameters like model year, maximum power etc. The output are modification factors that are applied on MOD or MTPL frequency, severity or expected loss for non-electric cars.