In advanced traffic management systems, an important function of traffic operation is determining whether the traffic is free flow or congested to make possible rapid and effective response by traffic control facilities. Also, it is desirable that the extent of congestion be quantified. From the observations of traffic flow, the time series analysis presents repetitive or regular behavior over time, which distinguishes it from classical statistics, which assumes independence over time. Frequency domain analysis, a branch of time-series analysis, helps to find implicit characteristics that cannot be obtained from time domain analysis. By taking advantage of tools in frequency domain analysis, this paper proposes a new criterion function that can detect the onset of congestion. It is found that the changing rate of the cross-correlation between density dynamics and flow rate determines traffic transferring from free flow phase to the congestion phase. The new method suggests that an un-returnable transition will occur only if the changing rate of the cross-correlation exceeds a threshold. Based on real traffic data, detection of congestion is conducted in which the new scheme performs well compared to previous studies.
Abstract