Urban short-term traffic flow data is highly nonlinear and with time-varying characteristics, especially changing abruptly before and after a period of congestion. In this paper, a Markov switching (MS) model is proposed to model the short-term traffic flow forecasting. The model is suitable to describe a nonlinear time series of urban traffic flow. The proposed model classifies traffic flow conditions into two states corresponding to two regimes of the MS model respectively. The traffic flow model changes under each state, switching at a certain probability controlled by a first-order Markov chain. Comparisons with real data for traffic flow validate the method with accurate short-term forecasts.
Urban Short-Term Traffic Flow Forecasting Using the Markov Switching Model
First International Symposium on Transportation and Development Innovative Best Practices ; 2008 ; Beijing, China
2008-04-04
Conference paper
Electronic Resource
English
Urban Short-Term Traffic Flow Forecasting Using the Markov Switching Model
British Library Conference Proceedings | 2008
|Reliable Short-Term Traffic Flow Forecasting for Urban Roads
Transportation Research Record | 2013
|