In a world where supply chains are complex, dynamic and prone to disruptions, providing accurate Estimated Time of Arrival (ETA) predictions is challenging but critical. With potentially hundreds of factors influencing the outcome, a simple and static model can no longer suffice in the world of today. Some shippers and Logistics Service Providers are tempted to experiment with Machine Learning-based ETA prediction methods, but hiring data science talents and building an end-to-end machine learning pipeline can be cost-prohibitive.

Oracle Transportation Management (OTM) Cloud can help. With the recent 21A update, OTM now has the ability to provide machine learning-based shipment transit time predictions. Leveraging the rich shipment histories in OTM, this capability identifies shipments truly at-risk, and provides customers with accurate ETA predictions.

Through rigorous research using real-world shipment data, Oracle's machine learning capability has demonstrated a reduction of prediction errors by upwards of 65%. Furthermore, Oracle's data pipeline is pre-built, the model building process is automated, and the results are actionable.