Correct and well timed site visitors circulate forecasting is a important activity of the clever transportation system (ITS). The anticipated outcomes supply the mandatory info to help the choices of directors and vacationers. To analyze development and periodic traits of site visitors circulate and develop a extra correct prediction, a novel technique combining periodic-trend decomposition (PTD) is proposed on this paper. This hybrid technique relies on the precept of “decomposition first and forecasting final”. The well-designed PTD strategy can decompose the unique site visitors circulate into three parts, together with development, periodicity, and the rest. The periodicity is a strict interval perform and predicted by biking, whereas the development and the rest are predicted by modelling. To show the common applicability of the hybrid technique, 4 prevalent fashions are individually mixed with PTD to ascertain hybrid fashions. Visitors quantity knowledge are collected from the Minnesota Division of Transportation (Mn/DOT) and used to conduct experiments. Empirical outcomes present that the imply absolute error (MAE), imply absolute share error (MAPE), and imply sq. error (MSE) of hybrid fashions are averagely decreased by 17%, 17%, and 29% greater than particular person fashions, respectively. As well as, the hybrid technique is strong for a multi-step prediction. These findings point out that the proposed technique combining PTD is promising for site visitors circulate forecasting.