ARC Advisory Group engaged in an informative discussion with Derek Gittoes, VP Supply Chain Management Product Strategy at Oracle, as part of ARC’s Digital Supply Chain Forum. Derek recently authored an article on Logistics Viewpoints describing recent advancements in logistics predictability through the application of machine learning. And we thought this to be a great opportunity to get further details on this hot topic from a practitioner on the front line of logistics application development.
We asked Derek to provide greater detail on a few key points. Namely, how does machine learning help with predicting shipping transit times? Secondly, why should shippers and logistics service providers consider using machine learning in their transportation management systems, and why now? Third, what are Oracle’s machine learning capabilities in transportation management? Finally, what other notable trends are you seeing in the logistics space? Below are some key points made in the discussion. Watch the attached video for the full interview.
Machine Learning and Accurate Transit Time Prediction
Prediction of shipping transit times sounds simple but is actually extremely complex. Transit times are driven by a large number of variables, such as origin, destination, mode of transportation, day of week, and numerous environmental factors. Due to the complexity, most organizations revert to a more simplistic, static model. Machine learning is so powerful because it can ingest large amounts of shipment data and automatically determine which attributes have an impact in a given situation.
Why Machine Learning is Going Mainstream
Machine learning technology has been around for a long time but there have been substantial obstacles to mainstream adoption. These traditional obstacles included the development, training, and interpretation of the model; access to all of the data sources; robust computational power to process it all; and more. Oracle has removed these obstacles, allowing the machine learning technology to be readily adopted by Oracle Transportation Management Cloud users.
Oracle Transportation Management Machine Learning Capabilities
The most recent update to Oracle Transportation Management Cloud introduced a new machine learning based feature that is used to predict shipment transit times that uses the customer’s historical shipment data to train the model and update it over time. Oracle’s research has demonstrated that the machine learning based predictions can reduce prediction errors by upwards of 65 percent. Furthermore, Oracle’s data pipeline is pre-built, the models are pre-defined, and the ultimate output is the machine learning based prediction and confidence interval available to the transportation planner.
Other Notable Logistics Trends
Real-time visibility is an essential piece of information to customers. IoT technology is making visibility data more robust and readily available. Finally, tactical planning is in constant need, as organizations look to assess impacts to their logistics network, whether planned changes or disruptions.
Derek H. Gittoes is vice president for supply chain management product strategy and leads the team responsible for Oracle’s portfolio of cloud products for logistics and order management. Derek has 27 years of experience in supply chain software. Prior to joining Oracle, he was vice president of product solutions at G-Log, where he was responsible for product marketing. Derek was also a founder and managing director of Transport Dynamics, a company that specialized in the development of optimization software for transportation companies.