2D Feature Recognition for Vehicle BOM Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vehicle maintenance is hindered by incomplete or unavailable bills of materials (BOMs), especially for small batch production, older, or modified vehicles, leading to increased maintenance cycle times and downtime due to the difficulty in identifying and acquiring necessary parts.
Innovation Solution
The use of 2D feature recognition technology to generate imagery of vehicle configurations, compare them to a survey library, and create a revised BOM or 3D model, allowing for accurate part identification and efficient maintenance planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional BOM-based part identification is used, then part identification can be performed when BOM is complete, but maintenance time increases and productivity decreases when BOM is incomplete or unavailable
Solution Approach 1:
The patent replaces manual BOM-based part identification with an automated image recognition system. The system captures images of vehicle parts, processes them through AI/ML algorithms to identify parts automatically, and generates BOM data without human intervention. This substitution of mechanical/manual processes with automated systems resolves the contradiction by maintaining high identification accuracy while dramatically reducing maintenance cycle time.
Solution Approach 2:
The patent creates digital copies (images) of physical vehicle parts and uses these copies for identification purposes. By capturing images of parts and comparing them against a database of known parts, the system eliminates the need for physical inspection and manual BOM consultation, thereby reducing maintenance time while preserving identification accuracy.
2Loss of information
If manual inspection and BOM consultation are performed to identify parts, then part identification can be accurate when information is available, but vehicle downtime increases due to the time required for identification and part acquisition
Solution Approach 1:
The patent performs preliminary actions by capturing images of vehicle parts and pre-processing them through the image recognition system before maintenance begins. The system pre-generates BOM data and identifies all necessary parts in advance, so when maintenance starts, parts are already identified and can be ordered immediately, eliminating delays associated with on-site manual inspection and information gathering.
Solution Approach 2:
The system enables the vehicle data to serve itself by automatically extracting configuration information from images and generating BOM data without requiring external manuals or expert knowledge. The image recognition system acts as a self-service mechanism that autonomously captures, processes, and interprets vehicle information, thereby reducing both information loss and time loss.
3Manufacturing precision
If comprehensive vehicle survey is performed to identify all parts for modification, then retrofitting accuracy improves, but service time increases
Solution Approach 1:
The patent replaces comprehensive manual vehicle survey and inspection with automated image capture and recognition systems. The system quickly captures images of the vehicle and its components, automatically identifies parts, their locations, and specifications through AI/ML processing. This substitution maintains high retrofitting accuracy by precisely identifying all necessary components while dramatically reducing the time required for the survey process.
Solution Approach 2:
The system performs preliminary identification and classification of all vehicle parts through image recognition before the actual retrofitting work begins. By pre-identifying installation locations, power sources, and restrictive structures from captured images, the system prepares all necessary information in advance, ensuring accurate retrofitting execution while minimizing the duration of vehicle service interruption.
Data Source
AI summary
Methods and systems for populating fleet maintenance data using 2D feature recognition are disclosed. In one embodiment, a method of determining a configuration of a vehicle includes surveying the vehicle using an imaging device to generate 2D imagery of a configuration of the vehicle. The generated 2D imagery of the configuration may be compared to a survey library of 2D images to identify a part in the configuration. Existing data from legacy systems may be extracted for the part. The part may be added to a bill of materials and used to create a 3D model of the vehicle.