Unilever : What happens when machine learning meets ocean shipping?

Getting our products into the hands of the 2.5 billion consumers in 190 countries who use them daily requires complex logistics. And an integral part of that is ocean shipping.

At any given time, we have around 12,000 containers on 1,500 ships criss-crossing the seas with raw materials, packaging and finished goods. In fact, by the time you finish reading this article, another Unilever container will have started its journey.

Shipping is one of the oldest industries in the world, and getting a container across waters involves numerous parties from customs agents and port authorities to inland transporters and freight forwarders. Across all these, there's a multitude of information systems that don't talk to each other, which results in lack of end-to-end visibility.

Until now, that is.

In collaboration with several partners, we have developed the Virtual Ocean Control Tower: a first-of-its-kind digital innovation that has transformed how we manage our sea freight logistics.

This unmanned system seamlessly connects all parties in the chain through a real-time electronic data interchange (EDI) information flow.

That means we have at our fingertips highly accurate, continually updated data on the status of each shipment at every stage of its journey, from its point of departure to when it reaches the distributor, store or factory.

Crucially, the system uses machine learning and predictive analytics, so every potential issue along the way - from port congestion to demurrage and detention charges, temperature deviation and ETA changes - is proactively identified and an alert automatically sent to everyone who needs to know.