Summary of theoretical and practical implications of the proposed forecasting framework
| Dimension | Key insights | Relevance |
|---|---|---|
| Model integration | Combination of ARIMA, Prophet, XGBoost and LSTM | Enhances comparative understanding of linear vs nonlinear forecasting techniques |
| Explainability | Use of Integrated Gradients and PDPs for model interpretation | Bridges gap between performance and transparency, improving model trust and usability |
| Causality analysis | Granger causality confirms vessel arrivals as leading indicators | Supports inclusion of operational variables in predictive modeling |
| Feature sensitivity | High dependency on lagged container volumes and ship arrivals | Enables prioritization of critical data inputs in forecasting systems |
| Operational planning | Forecasts support berth scheduling, crane allocation, and labor planning | Reduces congestion and improves resource efficiency in Gulf ports |
| Policy alignment | Model outputs traceable to input drivers | Facilitates compliance with national planning strategies (e.g. Vision, 2030) |
| Resilience | Forecasting models robust under dynamic, uncertain trade conditions | Enhances ports' ability to anticipate and respond to disruptions |
| Academic value | Context-specific modeling in underexplored Gulf port settings | Contributes to localized forecasting theory in maritime logistics |
| Dimension | Key insights | Relevance |
|---|---|---|
| Model integration | Combination of ARIMA, Prophet, XGBoost and LSTM | Enhances comparative understanding of linear vs nonlinear forecasting techniques |
| Explainability | Use of Integrated Gradients and PDPs for model interpretation | Bridges gap between performance and transparency, improving model trust and usability |
| Causality analysis | Granger causality confirms vessel arrivals as leading indicators | Supports inclusion of operational variables in predictive modeling |
| Feature sensitivity | High dependency on lagged container volumes and ship arrivals | Enables prioritization of critical data inputs in forecasting systems |
| Operational planning | Forecasts support berth scheduling, crane allocation, and labor planning | Reduces congestion and improves resource efficiency in Gulf ports |
| Policy alignment | Model outputs traceable to input drivers | Facilitates compliance with national planning strategies (e.g. Vision, 2030) |
| Resilience | Forecasting models robust under dynamic, uncertain trade conditions | Enhances ports' ability to anticipate and respond to disruptions |
| Academic value | Context-specific modeling in underexplored Gulf port settings | Contributes to localized forecasting theory in maritime logistics |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.