Fleet-wide route optimization at national scale
A produce-handling rule decided which shipments could share a truck. Encoding that rule into the routing algorithm itself — not checking it by hand afterward — is what made the system usable at fleet scale.
How it broke down
Problem
A Fortune 100 grocery retailer needed its entire delivery fleet routed against supply from multiple depots — at a scale generic routing tools weren't built to handle.
Constraint
Produce doesn't behave like the rest of the inventory. Shipments that looked combinable on paper couldn't share a route once produce-handling rules applied, and that rule had to be a hard constraint the optimizer respected — not a manual check layered on afterward.
Approach
Supply depots were folded directly into the routing model, so the optimizer weighed depot assignment and route combination together, with the produce-handling rule encoded as a hard constraint from the start — the same constraint-first principle behind every engagement.
Outcome
The same depot-aware, constraint-first pattern was delivered again for a nationally distributed beverage brand, without re-deriving the approach from scratch.
Have a routing problem like this one?
Your domain expertise ⊕ our engineering team. A 30-minute call is enough to tell whether this is a fit — bring the problem, no prep needed.
Book a callFounded by scientists turned AI leaders — national-lab research backgrounds, then senior data science and AI roles shipping production systems across logistics, security, and adtech.