A Period-Specific Combined Traffic Flow Prediction Based on Travel Speed Clustering
Bin Feng,Jianmin Xu,Yongjie Lin,Penghao Li
DOI: https://doi.org/10.1109/access.2020.2992657
IF: 3.9
2020-01-01
IEEE Access
Abstract:Short-term traffic flow forecasting has always been an interesting research at the fields of Intelligent Transportation Systems. This paper presents a time-based combined traffic flow prediction model based on field data collected by loop detectors at signalized intersections, which are used to signal optimization, route choice, traffic monitoring, etc. Firstly, the traffic flow and corresponding travel speed by hour is processed for error elimination and correlation analysis. Secondly, time of day is divided into three groups (peak, flat-peak and low-peak period) in terms of hourly travel speed clustering such as to separately develop prediction formula for each period with avoiding the overfitting of a single 24-hour model. And then, a combined prediction model based on time partition is proposed for 24-hour traffic flow forecasting, which adopts grey theory model for flat-peak and low-peak periods and back-propagation artificial neural network for peak hours, respectively. Finally, in tests that used field data from Xingzhong Rd, Zhongshan, China, the developed combined method based on speed clustering shows promise in reducing mean absolute error, mean absolute percentage error and mean squared error. Further exploration with excessive experiments for comparison analysis exhibits that the period-specific combined model conducts a more accurate and reliable prediction than the individual model and existing combined ones with the same structure for 24-hour.
computer science, information systems,telecommunications,engineering, electrical & electronic