Search

Research on the Optimisation Method of Intersection Signal Phase Duration Based on Dynamic Traffic Demand

Author(s)

Tianci Wang

Affiliation(s)

Telecommunication Engineering and Management, International College, Beijing University of Posts and Telecommunications, Beijing, China

Abstract

With the acceleration of urbanisation and the growth of motor vehicle ownership, the problem of traffic congestion at intersections is becoming more and more prominent, and the traditional traffic signal timing scheme is difficult to cope with the dynamic fluctuation of traffic flow and other complex situations. This paper focuses on the integration of intelligent transportation system and artificial intelligence technology, and aims to solve the traffic signal timing optimisation problem by establishing a real-time evaluation and prediction model. The research includes constructing a dynamic timing optimisation model based on real-time traffic flow, developing intelligent regulation algorithms adapted to different time periods and unexpected conditions, and realising multi-objective optimisation such as reducing vehicle delays and safeguarding pedestrian safety; meanwhile, embedding learning techniques, integrating multi-source sensor data to construct a prediction model, balancing the traffic demand of multiple ends, and ensuring the system's adaptability to the complex environment. It is proposed to use multi-source heterogeneous data fusion technology, hybrid neural network prediction model and other methods to solve the problem of insufficient adaptability of the traditional scheme in dynamic traffic fluctuations and other scenarios, and the results of the research can support the upgrading of the intelligent transport system, with the ability of low-latency decision-making in complex environments.

Keywords

Dynamic Traffic Demand; Intersection Signals; Phase Duration Optimisation; Multi-Source Data Fusion; Hybrid Neural Network; Multi-Objective Optimisation Algorithm; Intelligent Transport System

References

[1] Dong Yibing, Zeng Hui, Li Jianke, et al. DPRT-YOLO: A real-time target detector for complex driving environments in intelligent connected vehicles [J/OL]. Computer Engineering and Applications. 2025-03-17 [2] P. Jiang, F.B. Chai, Z.C. Xie. Research on adaptive control of traffic signals based on PLC [J]. Highway Traffic Technology, 2025, 41 (3): 213-222. [3] Chunmei Zhou. Traffic signal light control method in smart city construction based on improved reinforcement learning [J]. Journal of Guiyang College (Natural Science Edition), 2025, 20 (1): 47-52. [4] Zhang Yan, Ren Anhu, Chen Yang. Design of intelligent traffic control system based on machine vision [J]. Information Technology, 2025, (2): 55-60. [5] Wang Z., Shi C., Zhang K.. YOLOv8-based track inspection recognition technology for turnout junction environment [J]. Science, Technology and Engineering, 2025, 25 (13): 5476-5483. [6] Sun W.J., Li Q.Y., Zhang J., et al. A review of deep neural network-based incremental learning research [J]. Data Analysis and Knowledge Discovery, 2025, 9 (1): 1-30.