Telematic Visit Attribution Using Ranked Candidate Polygons
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately attribute visits to physical locations using telematic automotive driving data due to differences in metadata tagging and location pinging behavior compared to mobile device data, leading to challenges in determining visitation rates and demographics.
Innovation Solution
A method that transforms raw telematic automotive driving data into place visitation data by defining candidate polygons, identifying relationships among them, assigning ranks based on distance and metadata, and aggregating visitation events to estimate visitation rates, using a geographic information system and computer processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If mobile device location data is used for visit attribution, then visitation information can be obtained, but the data lacks metadata about device status and pings occur inside building footprints rather than parking areas
Solution Approach 1:
The patent segments the location data processing by creating separate handling paths for mobile device data and telematic automotive data. It divides the candidate polygons into different types (building footprints, parking lots, access roads) and applies different ranking logic for each data source, thereby resolving the contradiction by treating each data type with appropriate segmentation strategies
Solution Approach 2:
The patent introduces an intermediary ranking system that acts as a mediator between raw location data and visit attribution results. The ranking algorithm serves as an intermediary layer that transforms telematic data with parking lot pings into accurate visit attributions by considering multiple factors including distance to building, time of day, and day of week patterns
2Loss of information
If telematic automotive driving data is used for visit attribution, then metadata about device status is available, but location pings occur at parking lots rather than inside buildings
Solution Approach 1:
The patent adds another dimension to the location matching problem by considering temporal dimensions (time of day, day of week) and spatial relationships (distance to building, hierarchy of polygons) rather than relying solely on direct geometric containment. This multi-dimensional approach resolves the contradiction by compensating for the parking lot ping location with additional contextual dimensions
3Productivity
If candidate polygons are ranked based on distance alone, then simple processing is achieved, but relationships among clusters of polygons are not considered
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing the hierarchical relationships among candidate polygons and their clusters before the actual visit attribution process. This pre-processing step creates a structured framework that enables efficient querying and ranking during runtime, resolving the contradiction by performing complex relationship analysis in advance rather than during real-time processing
Data Source
AI summary
Systems and methods for estimating visitation rates at a physical location in which predetermined relationships are identified among associated cluster of candidate polygons associated with each of a plurality of physical locations. Telematic data received from a vehicle is used to identify a location of the vehicle during a visitation event. Ranks are assigned to the candidate polygons based on both the predetermined relationships and a distance between the vehicle and the candidate polygons. The ranks can be aggregated with information from further visitation events to estimate visitation rates for the plurality of physical locations, and the estimated visitation rates can be stored in a memory.