Table 4 Forecasting result of each model of Shanghai Port.
From: A hybrid container throughput forecasting approach using bi-directional hinterland data of port
Model | MAE | RMSE | MPE | MAPE | SMAPE |
|---|---|---|---|---|---|
Naive | 170.0 | 208.15 | 3.80 | 3.80 | 3.90 |
ARIMA | 136.21 | 165.44 | -3.08 | 3.08 | 3.01 |
ARIMAX | 52.17 | 646.01 | -10.92 | 12.31 | 11.40 |
u-LSTM | 728.44 | 761.01 | 16.65 | 16.65 | 18.27 |
m-LSTM | 255.03 | 290.53 | 5.76 | 5.76 | 5.98 |
u-CNN | 136.76 | 166.42 | 2.73 | 3.14 | 3.21 |
u-GRU | 327.91 | 348.81 | 7.51 | 7.51 | 7.84 |
m-GRU | 158.29 | 194.85 | 3.57 | 3.57 | 3.66 |
Transformer | 589.43 | 620.04 | 13.45 | 13.45 | 14.51 |
Proposed Grey-CNN | 93.73 | 116.71 | 0.89 | 2.11 | 2.14 |
