Look at excursion data across any network and the distribution is not even. The trunk legs are clean. The warehouses are clean. Almost everything that goes wrong goes wrong in the last few hours, at the delivery door.
The arithmetic
A frozen line-haul from hub to hub has two door events: one at loading, one at unloading. A pharmacy route has forty. Each opening is a heat load, and a van refrigeration unit sized for a sealed box cannot pull back forty times across a working day while also driving between stops.
Stop asking the van to do it
The answer is not a bigger unit. It is to move the control from the vehicle to the package. Product travels in shippers that were conditioned overnight, with phase-change packs matched to the band, and the van becomes a mild, stable environment rather than the primary control. Open the van forty times and the product never notices.
What this requires
- A conditioning room and the discipline to use a full overnight cycle.
- Qualified profiles for each route length, not a single generic shipper.
- A logger inside each shipper, read out at the drop.
- Spare conditioned shippers on the van for re-packs after a failed delivery.
The failed delivery problem
Nobody plans for it and it happens on every route. If a drop fails at hour six, that shipper has now been opened and is going back to the depot on a clock nobody is watching. The rule is simple: a failed delivery gets re-packed into a fresh conditioned shipper at the next opportunity, and the original one goes back for re-conditioning. It costs a shipper. It saves a consignment.
Where to spend first
If you have a limited budget for cold chain improvement and you are spending it on the trunk leg, you are almost certainly spending it in the wrong place.
