01337556

Component

https://doi.org/10.3141/2215-09

http://scholar.google.com....+Lu&publication_year=2011

Predictions of short-term traffic volume support proactive transportation management and traveler information services. To address the periodicity, nonlinearity, uncertainty, and complexity of short-term traffic forecasting, seasonal autoregressive integrated moving average (SARIMA) and support vector machine (SVM) models are often used separately to forecast the time series of traffic flow. SARIMA can discover intrinsic relations (correlations) in time series data, especially fit for modeling seasonal, stochastic time series. In contrast, SVM has a strong nonlinear mapping ability for input and output, appropriate for solving complex, nonlinear problems. The few combined schemes that have been presented previously tend to be complicated. A simple and effective novel hybrid methodology that combines the SARIMA and SVM models is proposed to take advantage of the individual strengths of the two models. Two key issues in building a hybrid model process are discussed: identification of SVM input dimensions via SARIMA and parameter optimization of the hybrid model with particle swarm optimization. Experimental results with real-world data sets indicated that the hybrid model was superior to the individual SARIMA or SVM model in terms of root mean square error and mean absolute percentage error for the prediction of the short-term traffic flow.

01352159

11-2557

English

pp 85-92

2011

Transportation Research Record: Journal of the Transportation Research Board

9780309167291

Print

Figures (1) ; Maps (1) ; References (49) ; Tables (3)

Freeways; Highway traffic control; Traffic flow; Traffic forecasting; Traffic volume; Traveler information and communication systems

Autoregressive integrated moving average models; Hybrid models; Short term forecasts; Support vector machines

Data and Information Technology; Highways; Operations and Traffic Management; Planning and Forecasting; I71: Traffic Theory

TRIS, TRB, ATRI

Feb 17 2011 6:12PM

• Analytic Hierarchy Process and Geographic Information Systems to Identify Optimal Transit Alignments
• Application of IntelliDrive Information System to Enhance High-Speed Intersection Safety
• Comparison of Earthwork Computation Methods
• Development of Web-Based Analysis System for Real-Time Decision Support on Arterial Networks
• DRIVE Net: E-Science Transportation Platform for Data Sharing, Visualization, Modeling, and Analysis
• Enterprisewide Information Management System for Transportation Right-of-Way Activities
• Innovative Stereo Vision-Based Approach to Generate Dense Depth Map of Transportation Infrastructure
• Multidimensional Data Mining of Traffic Anomalies on Large-Scale Road Networks
• Protecting Location Privacy: Policy Evaluation
• Traveler Information Tool with Integrated Real-Time Transit Information and Multimodal Trip Planning: Design and Implementation
• Uncertainty Issues in Integrating Geographic Information Systems and the Global Positioning System for Transportation