Visitor Volume Estimation Using Multi-Source Data Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining visitor volume, such as human surveys and traffic data recorders, fail to provide accurate real-time data on the number of travelers by various modes of transport and their origins and destinations, and do not effectively combine data from different sources to estimate visitor volume accurately.
Innovation Solution
A system that utilizes a panel of vehicles with known information, associating them with vehicles not in the panel through GPS data and traffic sensors, to calculate visitor volume by matching timestamps and determining behavioral characteristics, and combines data from various sources including satellite imagery, mobile devices, and traffic monitors to provide real-time estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traffic data recorders and human surveys are used to determine visitor volume, then data collection is possible, but real-time accuracy and completeness of visitor volume estimates deteriorate
Solution Approach 1:
The patent combines multiple data sources including traffic data recorder information, mobile device GPS data, satellite imagery, and weather data into a unified analytical system. This integration allows the system to cross-validate and supplement data from individual sources, achieving both real-time processing capability and high estimation accuracy by leveraging the strengths of each data source while compensating for their individual limitations.
2Measurement precision
If data from multiple sources is combined to estimate visitor volume, then estimation accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent segments the complex multi-source data processing task into distinct functional modules: a data collection module that gathers information from various sources, a data normalization module that standardizes different data formats and quality levels, an analysis module that processes the normalized data, and an output module that generates visitor volume estimates. This segmentation allows each module to handle specific processing requirements independently, reducing overall system complexity while maintaining high estimation accuracy.
Solution Approach 2:
The patent introduces data normalization as an intermediary process between raw data collection and final analysis. This normalization layer standardizes data from diverse sources (traffic recorders, mobile devices, satellite imagery) into a common format and quality standard, making subsequent processing simpler and more efficient. The intermediary normalization step prevents complexity from propagating through the entire system by resolving data incompatibilities early in the processing chain.
3Productivity
If real-time data processing is implemented, then timely visitor volume insights are provided, but data accuracy and reliability may deteriorate
Solution Approach 1:
The patent implements preliminary data validation and quality assessment routines that operate in real-time as data is collected from various sources. These preliminary actions include checking data completeness, verifying data format compliance, and assessing data quality metrics before the data enters the main processing pipeline. By performing these validation actions upfront, the system ensures that only reliable data proceeds to analysis, maintaining both real-time processing speed and data accuracy.
Data Source
AI summary
The present invention is a system and method for improving the accuracy with which human behavioral insights regarding visitor volume to places can be predicted. The process aligns at least two different data sets and normalizes the information by removing statistical bias. In a particular implementation motor vehicle data is collected from a known subset of road-borne vehicles and compared to data collected from unknown vehicles. The data is then increased or discounted to remove biases present in the known subset. The system analyzes the normalized data and returns a report including predictions to a human user.