Metastable critical point estimation is enjoying much attention with increasingly serious traffic problems in cities. However, related research on metastable critical point estimation seldom employs travel data due to the lack of data acquisition and the neglect of catastrophe characteristic of traffic flow. This problem can be solved with the massive application of Electronic Registration Identification of the motor vehicle (ERI), which is an emerging technology to identify a unique vehicle based on Radio Frequency Identification (RFID). This paper presented an ERI data based framework for identification of critical phase transition interval of traffic state and estimation of metastable critical point considering the catastrophe characteristics of traffic flow. We utilized real data into the presented algorithms. Finally, the metastable critical traffic flow characteristic parameter with the various observation period was also analyzed quantitatively based on LSTM model, which can provide an important basis for traffic congestion early warning and control.

CICTP;Metastable critical point;ERI data;Catastrophe characteristic of traffic flow;Data gap;GMM;LSTM