A Multi-Step Approach Framework for Freight Forecasting of River-Sea Direct Transport without Direct Historical Data

Abstract

The freight forecasting of river-sea direct transport (RSDT) is crucial for the policy making of river-sea transportation facilities and the decision-making of relevant port and shipping companies. This paper develops a multi-step approach framework for freight volume forecasting of RSDT in the case that direct historical data are not available. First, we collect publicly available shipping data, including ship traffic flow, speed limit of each navigation channel, free-flow running time, channel length, channel capacity, etc. The origin–destination (O–D) matrix estimation method is then used to obtain the matrix of historical freight volumes among all O–D pairs based on these data. Next, the future total freight volumes among these O–D pairs are forecasted by using the gray prediction model, and the sharing rate of RSDT is estimated by using the logit model. The freight volume of RSDT is thus determined. The effectiveness of the proposed approach is validated by forecasting the RSDT freight volume on a shipping route of China.

A New Container Throughput Forecasting Paradigm under COVID-19

“…In the field of container throughput forecasting, some traditional economic models have been frequently used, such as the autoregressive integrated moving average (ARIMA) model [7], the seasonal autoregressive integrated moving average (SARIMA) model [8,9], the exponential smoothing model [10], the error correction model (ECM) [11], the auto-regressive conditional heteroscedasticity (ARCH) model [12], the multiple regression model [13], the vector autoregressive (VAR) model [14], and the grey forecasting model [15,16]. However, these econometric models are unable to capture the nonlinear part of the original data.…”

“…We compare the existing studies with our study in Table 1. SARIMA-ANN √ √ × × Schulze et al [10] SARIMA and Holt-Winters × × × × Fung et al [11] Error-correction model × × × × Munim et al [12] ARIMA-ARCH √ × × × Veenstra et al [13] Vector autoregressive model × × × × Gao et al [14] Structural change VAR model × × × × Peng et al [15] Grey model × × × × Guo et al [16] Grey model × × × × Ding et al [17] BP Neural Network × √ × × He et al [18] GM (…”