Figure 7
A figure of four horizontal bar graphs shows mean absolute error for full and reduced models across four ports.The figure shows four bar graphs arranged in a vertical series. The top graph is titled “Sensitivity Analysis – Hamad Port”. The horizontal axis ranges from 0 to 80 in increments of 20 units. The markings on the vertical axis from top to bottom are: “Full Model”, “Containers”, and “Ship underscore Count”. The graph shows three horizontal bars. The bars extend to mean absolute error values as follows: Full Model: 56; Containers: 84; Ship underscore Count: 56. The second graph is titled “Sensitivity Analysis – Jeddah Port”. The horizontal axis ranges from 0 to 100 in increments of 20 units. The markings on the vertical axis from top to bottom are: “Full Model”, “Containers”, and “Ship underscore Count”. The graph shows three horizontal bars. The bars extend to mean absolute error values as follows: Full Model: 58; Containers: 100; Ship underscore Count: 58. The third graph is titled “Sensitivity Analysis – Salalah Port”. The horizontal axis ranges from 0 to 80 in increments of 20 units. The markings on the vertical axis from top to bottom are: “Full Model”, “Containers”, and “Ship underscore Count”. The graph shows three horizontal bars. The bars extend to mean absolute error values as follows: Full Model: 56; Containers: 92; Ship underscore Count: 56. The fourth graph is titled “Sensitivity Analysis – Shuwaikh Port”. The horizontal axis ranges from 0 to 80 in increments of 20 units. The markings on the vertical axis from top to bottom are: “Full Model”, “Containers”, and “Ship underscore Count”. The graph shows three horizontal bars. The bars extend to mean absolute error values as follows: Full Model: 58; Containers: 104; Ship underscore Count: 58. Note: All numerical data values are approximated.

Sensitivity analysis of feature removal on XGBoost forecast error (MAE)

or Create an Account

Close Modal
Close Modal