Traffic flow prediction and intelligent scheduling system based on deep learning

By integrating multi-source traffic data in real time and utilizing deep learning, a traffic flow prediction and intelligent scheduling system has been developed, which solves the problem of traffic flow prediction deviation under emergencies and achieves efficient and reliable scheduling decisions and traffic management.

CN122135571APending Publication Date: 2026-06-02SHANXI SHIPO ZICHEN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANXI SHIPO ZICHEN TECH CO LTD
Filing Date
2026-03-23
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Deep learning-based traffic flow prediction and intelligent scheduling systems struggle to make rapid short-term predictions when faced with extreme events such as sudden traffic accidents or temporary traffic control measures. This results in significant discrepancies between the predicted results and the actual traffic conditions, impacting the timeliness and reliability of subsequent scheduling decisions.

Method used

A deep learning-based traffic flow prediction and intelligent scheduling system is adopted. Multi-source heterogeneous data is accessed in real time through the traffic flow sensing terminal, and data cleaning, fusion and anomaly detection are performed. Spatiotemporal graph convolutional networks with attention mechanism are used for prediction, and scheduling strategies are generated by combining multi-objective optimization algorithms. Real-time monitoring and feedback optimization are performed through the scheduling execution terminal and the evaluation terminal.

Benefits of technology

It improves the reliability of short-term forecasts and the timeliness of scheduling decisions under abnormal events, dynamically responds to traffic changes, enhances overall traffic efficiency and system robustness, adapts to traffic characteristics in different regions and time periods, and reduces the risk of decision-making errors caused by forecast bias.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122135571A_ABST
    Figure CN122135571A_ABST
Patent Text Reader

Abstract

This invention relates to the field of intelligent transportation technology and discloses a traffic flow prediction and intelligent scheduling system based on deep learning. The system includes a traffic flow sensing end, a prediction decision end, a scheduling execution end, and a scheduling evaluation end. When making predictions and scheduling based on real-time traffic data, by accessing multi-source traffic flow data in real time and simultaneously performing anomaly detection, it can sense sudden traffic accidents and temporary traffic control extreme events in the road network and identify abnormal changes in traffic conditions. At the same time, the traffic flow prediction model built into the system can evaluate the credibility of its own prediction results in real time. When it determines that the prediction uncertainty is too high, it automatically issues an early warning, enabling subsequent decisions to combine real-time status and risk warnings, improving the reliability of short-term predictions and the timeliness of scheduling decisions under abnormal events, and reducing the risk of decision-making errors caused by prediction deviations.
Need to check novelty before this filing date? Find Prior Art