Applying Artificial Intelligence to Short-Term Traffic Flow Forecasting

Abstract:

Short-term traffic flow forecasting has a high requirement for the responding time and accuracy of the forecasting method because the result is directly used for instant traffic inducing. Based on the introduction of the fuzzy neural network model for short-term traffic flow forecasting together with its detailed procedures, this paper adopt the particle swarm optimization algorithm to train the fuzzy neural network. Its global searching and optimization algorithm helps to overcome the shortcomings of the traditional fuzzy neural network, such as its low efficiency and “local optimum”. A case study is also given for the PSO algorithm to train the fuzzy neural network for traffic flow forecasting. The result shows that the average square error is 0.932 when the PSO algorithm is put to use for the network training, which is 3.926 when the PSO is not used. Thus result is more accurate and it requires less time for the training procedures. It proves this method is feasible and efficient.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 5214-5217
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.5214
• Online since: January 2012
• Authors: Hai Huang
• Keywords: Artificial Intelligence (AI), Fuzzy Neural Network (FNN), Particle Swarm Optimization Algorithm (PSO), Traffic Flow Forecasting

References

[1] Zhaosheng Yang. Theory and Models for Urban Traffic Flow Inducing System, Beijing: People's Traffic Press, 2003. (in Chinese).

[2] Stephanedes YJ, Michalopoulos P G, Plum R A. Improved Estimation of Traffic Flow for Real-time Control. Washington: Transportation Research Record 795, TRB: 28-29.

[3] Fayyad U M. Automating the analysis and cataloging Knowledge Discovery and Data Mining. Menlo Park, CA: of sky surverys. Advances in MIT Press. (1996).

[4] Jack M. Zurada. Introduction to Artificial Neural System. West Publishing Company, 1993. Wang Ding-wei, Wang Jun-wei, Wang Hong-feng, Zhang Rui-you, Guo Zhe. Intelligent Optimization Methods [M]. Beijing: High Education Press of China. 2007. (in Chinese).

[5] YOSHIDA H, KAWATA K, FUKUIAMA Y, et al. A Particle Swarm Optimization for Reactive Power and Voltage Control Considering voltage Security Assessment [J]. IEEE. Trans. on Power Systems, 2000, 15(4). pp: 1232-1239.

DOI: 10.1109/59.898095

[6] COELLO C A. A Comprehensive Survey of Evolutionary-based Multi-objective Optimization Techniques [J]. Knowledge and Information Systems, 1999, 1(3). pp: 269-308.

DOI: 10.1007/bf03325101

[7] Wang Ding-wei, Wang Jun-wei, Wang Hong-feng, Zhang Rui-you, Guo Zhe. Intelligent Optimization Methods [M]. Beijing: High Education Press of China. 2007. (in Chinese).

[8] PARSOPOULOS K E, VRAHATIS M N. Particle Swarm Optimizer in Noisy and Continuously Changing Environments [J]. Artificial Intelligence and Soft Computing, 2001(1). pp: 289-294.