Multivariate short-term traffic flow forecasting using time-series analysis

Authors:
Bidisha Ghosh;Biswajit Basu;Margaret O'Mahony
Affiliations:
Department of Civil, Structural, and Environmental Engineering, Trinity College, Dublin, Ireland;Department of Civil, Structural, and Environmental Engineering, Trinity College, Dublin, Ireland;Department of Civil, Structural, and Environmental Engineering, Trinity College, Dublin, Ireland
Venue:
IEEE Transactions on Intelligent Transportation Systems
Year:
2009

Citing 2
Cited 4

Bayesian forecasting and dynamic models (2nd ed.)

Bayesian forecasting and dynamic models (2nd ed.)
Space-time modeling of traffic flow

Computers & Geosciences

Traffic signal optimization in "La Almozara" district in Saragossa under congestion conditions, using genetic algorithms, traffic microsimulation, and cluster computing

IEEE Transactions on Intelligent Transportation Systems
Collective traffic forecasting

ECML PKDD'10 Proceedings of the 2010 European conference on Machine learning and knowledge discovery in databases: Part II
Road traffic prediction using context-aware random forest based on volatility nature of traffic flows

ACIIDS'13 Proceedings of the 5th Asian conference on Intelligent Information and Database Systems - Volume Part I
GCBN: a hybrid spatio-temporal causal model for traffic analysis and prediction

WAIM'13 Proceedings of the 14th international conference on Web-Age Information Management

Quantified Score

Hi-index	0.00

Visualization

Abstract

Existing time-series models that are used for shortterm traffic condition forecasting are mostly univariate in nature. Generally, the extension of existing univariate time-series models to a multivariate regime involves huge computational complexities. A different class of time-series models called structural time-series model (STM) (in its multivariate form) has been introduced in this paper to develop a parsimonious and computationally simple multivariate short-term traffic condition forecasting algorithm. The different components of a time-series data set such as trend, seasonal, cyclical, and calendar variations can separately be modeled in STM methodology. A case study at the Dublin, Ireland, city center with serious traffic congestion is performed to illustrate the forecasting strategy. The results indicate that the proposed forecasting algorithm is an effective approach in predicting realtime traffic flow at multiple junctions within an urban transport network.