Purpose -To accurately forecast logistics freight volume plays a vital part in rational planning formulation for a country. The purpose of this paper is to contribute to developing a novel combination forecasting model to predict China's logistics freight volume, in which an improved PSO-BP neural network is proposed to determine the combination weights. Design/methodology/approach -Since BP neural network has the ability of learning, storing, and recalling information that given by individual forecasting models, it is effective in determining the combination weights of combination forecasting model. First, an improved PSO based on simulated annealing method and space-time adjustment strategy (SAPSO) is proposed to solve out the connection weights of BP neural network, which overcomes the problems of local optimum traps, low precision and poor convergence during BP neural network training process. Then, a novel combination forecast model based on SAPSO-BP neural network is established. Findings -Simulation tests prove that the proposed SAPSO has better convergence performance and more stability. At the same time, combination forecasting models based on three types of BP neural networks are developed, which rank as SAPSO-BP, PSO-BP and BP in accordance with mean absolute percentage error (MAPE) and convergent speed. Also the proposed combination model based on SAPSO-BP shows its superiority, compared with some other combination weight assignment methods. Originality/value -SAPSO-BP neural network is an original contribution to the combination weight assignment methods of combination forecasting model, which has better convergence performance and more stability.

Logistics freight volume forecasting plays a crucial role in the formulation of a rational planning strategy for a country. To improve forecasting accuracy, a novel multilevel decompose–ensemble method based on trend and wavelet decomposition combined with a linear regression model (LR) and an auto regression model (AR) is proposed in this study. First, the original data are resolved into trend and non-trend subseries by trend transform. Then, non-trend subseries are further broken down into one approximation subseries and several detailed subseries by wavelet transform. With respect to their different dynamically changing features and influencing factors, trend subseries are forecast by LR and non-trend subseries are, respectively, forecast by AR. The final prediction results are the summation of these subseries predictions. Forecasting results prove that the complex forecasting problem has been decomposed into some simple problems based on this multilevel decompose–ensemble method, which can improve prediction accuracy, when compared with individual models, the traditional decompose–ensemble method and a combination model. Consequently, the proposed method is a feasible forecasting approach for freight volume. According to this multilevel decompose–ensemble forecasting method combined with LR and AR, China's logistics freight volume in 2017 will increase to 69 010·16 million metric tonnes, and the average annual growth for the coming 5 years is 10·9%.