In our upcoming release, expected in January 2021, we provide the ability for our customers to leverage Machine Learning (ML) in the context of their transportation business processes. Users can use ML-based predicted transit times and shipment ETAs to identify shipments truly-at-risk and provide customers with accurate ETA predictions. Based on proprietary AutoML technology that automates much of the data science pipeline and workflow (data cleansing, data enhancement, test/train splits, algorithm and feature selection, etc.), the solution enables the business user to leverage ML within the context of the shipment lifecycle. Users can create models specific to their business scenarios (mode, geography, business unit, etc.), identify influential factors and fine tune each model for accuracy and performance. Predicted ETAs are actionable and can be used to enable better decision-making and to anticipate and alleviate exceptions. The model learns continuously and can adapt to changing conditions in the network. Near-term roadmap here includes refining the predicted ETAs with in-transit events, incorporating environmental factors into the prediction, and providing order-level transportation lead-time predictions to enable better order promising.
Brian Ramos began working with Oracle Transportation Management at DHL Supply Chain in 2006. Beginning in Operations, he has supported OTM solutions in Customer Service, Transport Planning, Systems Analysis, Functional Consulting, Solution Architecture and Product Management roles. He holds a Master of Engineering in Logistics degree from the Massachusetts Institute of Technology.