Geocoding Vehicle Telematics via Clustering and Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing geocoding methods for vehicle telematics data are often inaccurate due to poor quality pre-existing data and inaccuracies in delivery timestamps, leading to incorrect geographic coordinates for refueling locations and unregistered deliveries, which can result in inefficient routing and potential fraud detection issues.
Innovation Solution
A method and system that correlate vehicle telematics data with vehicle metadata to determine accurate geographic coordinates by clustering coordinates based on vehicle stationary periods and using probabilities to match location names with clusters, employing a stable-marriage matching algorithm to resolve conflicts, and transmitting markers to devices for improved routing and fraud detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pre-existing geographic data is used for geocoding, then the process is simple and fast, but the accuracy of geographic coordinates is poor
Solution Approach 1:
The system performs preliminary clustering of geographic coordinates based on vehicle stationary periods before matching with location names. This advance organization of data structures enables more accurate geocoding results by grouping coordinates that represent the same physical location, thereby resolving the contradiction between accuracy and process complexity.
Solution Approach 2:
The patent introduces an intermediary clustering mechanism that acts as a bridge between raw geographic coordinates and location names. By creating intermediate coordinate clusters based on temporal and spatial proximity during vehicle stationary periods, the system improves coordinate accuracy without requiring direct complex matching between all possible data points.
2Measurement precision
If delivery timestamps are used for geocoding, then the process is straightforward, but inaccuracies in timestamps lead to incorrect geographic coordinates
Solution Approach 1:
The system dynamically adjusts the matching criteria for geographic coordinates based on the reliability of available timestamp data. When timestamps are unreliable, the system expands the temporal window for coordinate clustering, allowing for more flexible matching that accommodates timestamp inaccuracies while maintaining coordinate accuracy.
Solution Approach 2:
The patent changes the parameter of temporal proximity threshold dynamically based on data quality assessment. By adjusting the time window parameter for coordinate clustering according to the observed reliability of timestamp data, the system can maintain accurate geocoding even when timestamps are imperfect, resolving the contradiction between coordinate accuracy and timestamp reliability.
3Productivity
If multiple location names are matched to geographic coordinates, then more locations can be identified, but conflicts arise requiring complex resolution
Solution Approach 1:
The patent segments the set of location names into distinct clusters based on their association with specific geographic coordinate groups. By dividing the matching problem into smaller, location-specific segments rather than attempting to match all location names simultaneously, the system identifies more locations while reducing the complexity of conflict resolution through structured organization.
Data Source
AI summary
In an embodiment, a system can determine geocoded data from a database of geographic coordinates and metadata. The system correlates metadata, such as invoice data, to geolocation data, such as GPS or cellular data to determine geocoded data. The system further identifies one or more geographic coordinates for one or more location names, which may not have a corresponding metadata entry, by generating clusters of geographic coordinates. The clusters are then matched to one or location names using a matching algorithm. Accordingly, improved geocoded data may be determined.