Driver Safety Score Estimation via Transaction Data Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for determining a driver's safety score in online on-demand transportation services are inefficient, relying on traditional interviews and questionnaires without a robust technology or algorithm, making it difficult to assess driving safety in a timely and comprehensive manner.

Innovation Solution

A system and method that utilize historical transportation service transaction data to extract relevant features, apply an estimation model to calculate a safety score, and offer contracts or insurance discounts based on this score, providing a more efficient and accurate assessment of a driver's safety.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional sample interviews and questionnaires are used to determine driving safety, then the implementation is simple and low-cost, but the timeliness and coverage are limited making it difficult to determine driving safety efficiently

Engineering Contradiction:
Improvetimeliness of safety assessmentVSAvoidcomplexity of assessment system
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical system of manual interviews and questionnaires with an automated electronic data processing system. The system automatically collects driving behavior data from terminals, processes it through servers using machine learning models, and generates safety scores without human intervention, thereby dramatically improving timeliness while managing complexity through automation.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by automatically collecting driving data from the driver's terminal device during normal operations. The terminal device itself serves as the data collection instrument, continuously gathering driving behavior information without requiring separate assessment activities, thus improving efficiency while keeping the system relatively simple to implement.

Inventive Principle:
Principle #25Self-service

2Productivity

If traditional sample interviews and questionnaires are used to determine driving safety, then the implementation is simple, but the coverage is limited

Engineering Contradiction:
Improvecoverage of safety assessmentVSAvoidcomplexity of data collection system
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system implements continuous data collection during normal driving operations rather than periodic assessments. The terminal device continuously monitors driving behavior parameters such as acceleration, braking, and route adherence, ensuring comprehensive coverage of driving safety throughout the entire operational period without interrupting service.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

The system uses the existing terminal device for multiple purposes: it serves as both the service delivery platform and the safety assessment instrument. The same device that provides transportation services also collects driving behavior data, eliminating the need for separate assessment equipment and expanding coverage without proportionally increasing system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If machine learning estimation models are used to determine safety scores, then the accuracy and efficiency improve, but the system complexity increases

Engineering Contradiction:
Improveaccuracy of safety scoreVSAvoidcomplexity of model algorithm
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the safety assessment into distinct functional modules: data collection at the terminal, data transmission, server-side processing, model training, and safety score generation. This segmentation allows the complex machine learning model to be developed and maintained independently on the server while keeping the terminal device relatively simple, thereby improving accuracy without overwhelming system-wide complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11279368B2System and method for determining safety score of driver
Publication Date: 2022.03.22 BEIJING DIDI INFINITY TECH & DEV CO LTD
  • US11279368B2 patent drawing
  • US11279368B2 patent drawing
  • US11279368B2 patent drawing

AI summary

The present disclosure relates to a method and system for determining a safety score associated with driver. The method includes: obtaining, by at least one computer, historical transportation service transaction data associated with an identification of a target driver; extracting, by the at least one computer, at least one target feature based on the historical transportation service transaction data; obtaining, by the at least one computer, an estimation model for estimating a safety score that reflects a safety expectation of a driver during transportation services; determining, by the at least one computer, a safety score associated with the target driver based on the estimation model and the at least one target feature; and providing, by the at least one computer, an offer to enter a contract to the target driver based on the safety score.