Traffic Data Prediction Using Reference Period Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurate real-time traffic data prediction is challenging due to the complex and time-varying nature of road traffic systems, influenced by natural and human factors, making it difficult to maintain prediction accuracy and operational ease.
Innovation Solution
A method that selects reference traffic data from historical periods based on data similarity and time intervals to predict traffic data in target periods, using a combination of video sensing devices and database analysis to determine correlation coefficients and vector distances for accurate forecasting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex models are used to account for natural and human factors, then prediction accuracy improves, but system complexity increases
Solution Approach 1:
The patent segments the traffic prediction problem by dividing historical periods into multiple time intervals and selecting specific reference periods based on similarity metrics. Instead of using a single complex model to capture all factors, the system breaks down the prediction into manageable segments (historical period division, similarity calculation, reference period selection) that can be processed independently and combined for the final prediction.
Solution Approach 2:
The patent introduces similarity calculation as an intermediary mechanism between historical traffic data and prediction results. By calculating similarity between current and historical periods based on traffic flow, occupancy, and speed, the system selects appropriate reference periods without requiring a complex model to directly interpret all natural and human factors. This intermediary approach simplifies the overall system while maintaining prediction accuracy.
2Reliability
If more historical data is analyzed to improve prediction accuracy, then prediction reliability improves, but computational time increases
Solution Approach 1:
The patent performs preliminary actions by pre-dividing historical periods into multiple time intervals and pre-calculating key traffic parameters (traffic flow, occupancy, speed) for each interval. This preprocessing allows the system to quickly retrieve and compare relevant historical data during prediction without performing complex computations on entire historical datasets, thus improving reliability while reducing computational time.
Solution Approach 2:
Instead of analyzing all historical data, the patent applies partial action by selecting only the most relevant reference periods based on similarity calculations. The system calculates similarity metrics and selects a limited number of reference periods that best match the current traffic conditions, avoiding the need to process excessive historical data while still achieving reliable predictions.
3Measurement precision
If real-time data processing is performed to maintain accuracy, then prediction accuracy improves, but operational complexity increases
Solution Approach 1:
The patent segments real-time data processing into distinct modules: data acquisition from multiple sources, similarity calculation between current and historical periods, reference period selection, and prediction generation. Each module handles a specific aspect of the processing, making the overall system easier to operate and maintain while preserving real-time accuracy through systematic data handling.
Solution Approach 2:
The patent uses copying by retrieving and comparing historical traffic data patterns rather than recreating complex models in real-time. The system copies relevant historical periods that match current conditions and uses them as references for prediction, simplifying real-time operations while maintaining accuracy through pattern matching.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method for predicting traffic data includes: obtaining traffic data in a plurality of periods; determining a reference period from the at least one historical period based on a time interval between each of the at least one historical period and the target period; determining a data similarity between the target period and the reference period based on traffic data at a first time point in the target period and traffic data at a time point corresponding to the first time point in the reference period; obtaining reference traffic data at a time point in the reference period corresponding to a second time point to be predicted in the target period; and predicting traffic data at the second time point in the target period based on the data similarity between the target period and the reference period, and the reference traffic data in the reference period.