Vehicle Repair Scheduling Using DTC Probabilities Across Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle diagnostic systems provide limited information on vehicle issues and recommended actions, often indicating only a general problem without specific resolutions, failing to assist users effectively in addressing vehicle faults.
Innovation Solution
A system utilizing a model, such as a BERT model, to analyze diagnostic trouble codes (DTCs) and provide multiple potential actions with associated probabilities for resolving vehicle faults, considering historical data from various vehicle makes and models, and enabling automated scheduling and execution of repairs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional diagnostic systems (CEL, OBD-II) are used to indicate vehicle issues, then users can identify that a problem exists, but the systems fail to provide specific actionable information and detailed resolutions
Solution Approach 1:
The patent introduces an intermediary processing layer between the OBD-II diagnostic codes and the user. This layer includes natural language processing models that translate raw diagnostic codes into actionable recommendations, bridging the gap between technical data and user-friendly guidance without requiring users to directly interpret complex code systems
Solution Approach 2:
The system enables users to obtain comprehensive diagnostic information and repair recommendations through automated processing of vehicle data. The natural language generation component creates detailed repair guidance that users can follow independently, reducing the need for professional mechanical expertise while maintaining information completeness
2Measurement precision
If multiple potential repair actions are provided with probability estimates, then users can make informed decisions, but the computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary processing by pre-computing probability estimates and repair recommendations based on historical data and vehicle specifications. When a diagnostic issue is detected, the system quickly retrieves and presents pre-analyzed repair options with success probabilities, avoiding time-consuming real-time computations while maintaining diagnostic accuracy
Solution Approach 2:
The patent transforms diagnostic data into multiple actionable parameters with associated probabilities. By changing the representation from single-code diagnostics to multi-parameter recommendations with confidence levels, the system enables users to make informed decisions without requiring complex real-time analysis
3Productivity
If automated repair scheduling and execution are implemented, then maintenance efficiency improves, but the extent of automation increases system complexity
Solution Approach 1:
The system integrates multiple functions into a unified automated platform: diagnostic code retrieval, natural language processing, repair recommendation generation, service center identification, and appointment scheduling. This multi-functional approach improves maintenance efficiency while consolidating automation complexity into a single coordinated system rather than multiple separate systems
Data Source
AI summary
Systems, methods, and computer-readable media are provided for automatically predicting and scheduling vehicle repairs. Particularly, a model (such as a natural language processing model, for example) may be trained using historical data relating to prior vehicle DTCs and actions taken to resolve the DTCs. The trained model may then be used to determine multiple potential resolutions to current DTCs provided by a vehicle. The model may also output probabilities associated with the various resolutions. Finally, the model may be configured to consider historical data associated with different types of vehicles than the current vehicle providing the DTC code. This allows for a larger data set to be considered by the model than if the model only considered data associated with the same type of vehicle.