Travel Time Data Adjustment Using Reference Link Zoning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing travel time prediction techniques face low accuracy due to insufficient probe data on certain roads, often resulting from low traffic volume or poor communication, leading to inadequate statistics and reduced prediction reliability.
Innovation Solution
A travel time data processing apparatus and method that utilizes zoning and road level information to select relevant link data from other roads with similar characteristics, enabling accurate travel time calculation even with limited probe information, by referencing zoning information and road level attributes to enhance data processing and prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If probe data is collected only from the target road, then data processing is simple, but prediction accuracy is low when probe data is insufficient
Solution Approach 1:
The patent segments the data processing by first identifying roads with insufficient probe data, then selectively applying different processing methods: using the target road's own data when sufficient, and using reference road data when insufficient. This segmentation resolves the contradiction by applying simplicity where possible and accuracy where necessary.
Solution Approach 2:
The patent introduces reference roads as intermediaries to bridge the gap when target road data is insufficient. These reference roads serve as mediators providing substitute probe data, enabling accurate predictions without requiring complex data collection infrastructure on the target road itself.
2Reliability
If reference road data is used to supplement insufficient probe data, then prediction accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent changes the parameter of data sufficiency by dynamically selecting between using target road data and reference road data based on whether the probe data count meets a predetermined threshold. This parameter-based selection resolves the contradiction by automatically adjusting data sources to maintain reliability while minimizing processing complexity.
Solution Approach 2:
The patent performs preliminary identification of roads with insufficient probe data before actual travel time calculation. By pre-categorizing roads based on data sufficiency, the system prepares the appropriate data sources in advance, reducing real-time processing complexity while ensuring prediction reliability.
3Measurement precision
If strict selection criteria are applied for reference road selection, then data relevance is high, but processing time increases
Solution Approach 1:
The patent applies partial action by using zoning information and road level as key selection criteria for reference roads, rather than evaluating all possible attributes. This partial selection approach maintains sufficient data relevance while significantly reducing processing time compared to comprehensive evaluation of all road characteristics.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A travel time data processor extracts a link that is present in the neighborhood of a target link, as a reference link. The travel time data processor selects an identical time zone of the reference link that is identical with the time zone of the target link in which the number of sample data is less than a reference number, and searches for another time zone having similar statistical data to that of the selected time zone. The travel time data processor adds the number of sample data in a time zone of the target link corresponding to the searched another time zone of the reference link to the number of sample data in the time zone of the target link that is equal to or less than the reference number, and generates statistical data from the summed-up sample data.