Most AI conversations in logistics over the past few years have focused on visibility: better dashboards, forecasts and alerts.

That work mattered, and it’s far from finished.

But it has also created a new frustration for the people running chemical supply chains: teams increasingly know exactly what’s going wrong, in real time, yet still have to do the manual work to fix it.

A flagged delay still requires someone to open three systems, check a slot booking and write the customer update by hand.

That gap – between insight and action – is where the next phase of AI adoption is heading. Not another layer of reporting, but AI that can take on the routine parts of a response and act in the time it takes a person to read the alert.

Return to Scenario 02 from earlier in this paper. A bulk load and a tanker load are due at the same dock on the same day, each with different equipment, safety and documentation requirements. A delay to either triggers a cascade of rebooking, communication and paperwork.

This is precisely the kind of situation where the problem isn’t a lack of information. It’s the number of manual steps needed to act on it. The team already knows there’s a problem. What they don’t have is the time to work through those steps three or four times a shift, every shift, across every site.

This is also why the sector’s low AI adoption rate – still around 25% across logistics broadly – represents a significant opportunity. The complexity that makes chemical logistics harder to plan also creates more scope for AI to reduce manual coordination, moving beyond simply surfacing problems to helping teams resolve them