Dive Brief:
- OOCL's partnership with Microsoft Research Asia will seek to implement deep and reinforcement learning technology to optimize shipping operations, Stephen Ng, the company's spokesman, told Supply Chain Dive in an e-mail.
- The AI tools will be deployed to help the company understand shipping patterns and associated variables, like vessel speed and weather data. In turn, the company will be able to "make more accurate decisions for optimized container routing (or rerouting) plans."
- "We are operating in a very complex shipping network with multiple variables affecting ship operations," Ng said. "Disruptions to shipping schedules because of bad weather, for example, may incur substantial operational costs to the company."
Dive Insight:
OOCL's responses explain both how the company can attribute a $10 million cost-saving figure to AI and provide a relatable use case for the emerging technology.
The technology builds on existing programs at OOCL, dedicated to optimizing its operations. Since 2012, OOCL has monitored its vessel movements, tying weather data to location, terminal activities and speeds in order to track and provide estimated arrival times at ports.
The control-tower initiative is depicted in the images within this post.
In previous coverage, Supply Chain Dive noted this automated collection and analytics could be considered AI applications. However, the recent news takes OOCL deeper: using machine learning, the shipping line will more actively adjust vessel routes, based on predictive data.
It's a transition from analytics to a prescriptive use case, building on data that the company already collects. The cost savings, in this case, come from avoiding disruptions in the first place, along with all the associated costs (think additional fuel, delay fees, loss of productivity).
"It is widely known that shipping network operations involve multiple parties and variables that can change at any moment on the fly," Ng said. "The enhanced visibility through predictive capabilities and efficiencies we gain from a more smooth and robust operational network can greatly benefit our customers’ time sensitive supply chains."
