Data-driven modelling of unloading hours using explainable gradient boosting models

Celal Çakıroğlu

doi:10.1016/j.aei.2026.104353

Data-driven modelling of unloading hours using explainable gradient boosting models

2026-01-23

0

PRE

AI

Mehmet Hakan Özdemir *

DOI：10.1016/j.aei.2026.104353

Original

Original request for help

Save

Abstract

En 中文

• Machine learning–based models (LightGBM and XGBoost) were developed to accurately predict unloading times in real logistics operations. • Both models achieved excellent predictive performance, with R2 values exceeding 0.99. • SHAP analysis revealed that the load of the leg has a greater impact on unloading time than gross truck weight and leg distance.

Data-driven modelling of unloading hours using explainable gradient boosting models

Abstract

Journal

Organization