EN
Back to the archive

The encyclopedia · Engineering & Operations · Operational decision · 2012–2016

Fritom cut delivery distance ~24% by planning two depots as one network

Dutch logistics firm Fritom treated two autonomous carriers' requests as one routing problem, cutting yearly travel distance about 24%.

Fritom

the move

Fritom, a Dutch logistics provider with roughly €150 million annual turnover, ran two domestic freight carriers as autonomous business units, each planning its own routes from its own depot. That structure left requests crossing each other's territories and trailers returning empty.

Working with the University of Groningen, Fritom redefined the problem as a generalized pickup-and-delivery model and tested two collaborative configurations over 220 real operating days. A variable geographical division, where delivery areas shift with demand and shuttle capacity varies daily, outperformed both the fixed-division design and no collaboration.

The best configuration cut total travel distance about 24% versus separate operation and 6.6% versus Fritom's existing ad-hoc collaboration, with the variable split alone saving 202,573 km more than the fixed one.

why it works

  • Two depots planning separately cannot see the cross-dock and backhaul savings in each other's routes.
  • A variable division adapts to daily request imbalances, unlike a permanent geographic boundary.
  • Defining a new problem class forced the team to model what planners actually did, not a textbook variant.
  • The simulation used real requests, so managers could trust the projected savings before changing operations.
the payoffPlan collaboratively with a variable regional splitclever

what transfers

Merge the planning before merging the companies; the savings come from sharing one model.

what came after

The researchers formulated managerial recommendations for Fritom and described the implementation process for the new collaborative structure. Fritom's managers expected additional efficiency gains as they continued to develop the collaboration beyond the study's configurations.

references

spotted an error? The archive wants to know.

same kind of clever