Cold-chain risk concentrates at handovers, not during storage. Inbound dock, picking, loading, transfer and channel receiving all take product out of a controlled environment for a while. The work is shortening that exposure and recording it, not setting the chamber colder.
What happens after your goods reach us
- Receiving and location controlThree-zone storage with lot control and first-expiry-first-out
- Picking and mixed dispatchOrder picking, mixed loads, labelling and dispatch scheduling
- Channel spec compliancePallet height, label placement and remaining shelf life per channel
- Domestic deliveryTemperature-appropriate delivery to warehouses, stores or end customers
- Continuous temperature recordsStorage and transit logs, used directly as a required export document
The risk is not the chamber. It is the handover.
The chamber is the easiest part of the chain to control. It has fixed equipment, continuous monitoring, a named owner, and the product inside it is stationary. That is precisely why the chamber is rarely the source of a problem.
Problems happen when product moves. Waiting on the inbound dock, picking, loading, a second handling at a transfer hub, waiting on customs, dwell time at the channel receiving bay, holding in the back of a store: every one takes the product out of a controlled environment, and every one belongs to a different party.
That is the real difficulty in cold chain. It is not a technical problem, it is a boundary-of-responsibility problem. Goods sit on a dock for forty minutes and nobody has breached anything, because nobody’s contract covers those forty minutes.
So the work is shortening the exposure and recording it. Shortening it needs operational design: pre-cooled trailers, a temperature-controlled picking area, dock scheduling. Recording it is what makes the problem traceable and improvable. Neither is solved by setting the chamber colder.
Frozen: recrystallisation
Freezing works by turning water to ice so that microbial and enzymatic activity nearly stops. The catch is crystal size. Fast freezing forms small crystals that do limited damage; slow freezing forms large ones that puncture cell walls.
Fluctuation does the same thing as slow freezing. On each small rise the smallest crystals melt first, and on cooling that water does not reform as small crystals. It attaches to the crystals already large. After a few cycles the product holds a few very large crystals instead of many small ones.
The result only appears on thawing: weeping, soft texture, a mealy mouthfeel. By then nothing can be done, and it is impossible to see while frozen. The pack looks normal and the date mark is still valid.
So the specification is not below minus 18 degrees. It is below minus 18 with fluctuation controlled. The first constrains a set point; the second constrains the actual history, and the second is what decides quality.
Chilled: shelf life spent in advance
The shelf life on a chilled product assumes it stays at the declared temperature. Microbial growth accelerates as temperature rises, so every excursion spends part of that shelf life in advance.
The awkward part is that the spending is irreversible and invisible on the pack. The date mark still says what it always said, while the actual quality curve has moved forward. A consumer buys a poor product inside its date, and everyone involved complied with the label.
That is why channels set explicit rules on remaining shelf life at delivery: they cannot audit the chain, so they manage the risk with a percentage. For short-life items this is a hard constraint: total shelf life minus the required remainder leaves the days in which production, warehousing, delivery and shelf placement must all happen.
It is also why exported chilled goods need a continuous log. What a buyer wants is not the claim that you refrigerate but a record they can check. Without it a shipment can be held for inspection or refused.
Ambient goods have their own problem
Ambient storage is usually treated as the easy leg, but it carries two risks that get overlooked: humidity and temperature differentials.
Humidity affects packaging and powders. Cartons lose strength once they take up moisture and deform when stacked. Powder products in packaging with a weak moisture barrier cake, and caking has no corresponding disclaimer on a label.
Temperature differentials cause a subtler problem. Moving product from a cooler environment into a warmer one can condense moisture on the inside of the pack. For a dry product that moisture stays inside, and it becomes the starting point for later mould or caking.
So the specification for ambient storage is not "no temperature control". It is light, humidity and differentials controlled. Managing those three costs far less than a cold chain, but it is not free.
| Frozen storage | Below minus 18, managing the fluctuation band rather than the set point alone |
|---|---|
| Chilled storage | Set to each SKU declared condition, continuously monitored |
| Ambient storage | Light and humidity controlled, to avoid condensation inside the pack |
| Temperature records | Logged through storage and transit; a required export document |
| Lot control | First-expiry-first-out with two-way traceability |
| Channel receiving spec | Pallet height, label placement and remaining shelf life per channel |
| Dock operations | Trailer pre-cooling, scheduling, dwell control |
The receiving bay
The last leg of the chain is channel receiving, and the rules there are not ours. They belong to the channel. So it is not only a logistics question, it is a commercial condition.
A channel receiving specification typically covers pallet height, carton marking placement, barcode legibility, the proportion of shelf life remaining on arrival, and the receiving window. Fail any one and the whole load can be turned away, and the cost of a rejection includes the freight, the rescheduling and the shelf life that consignment has already spent.
The receiving window is the constraint most often underestimated. A narrow window means delivery has to arrive precisely. Arrive early and you wait, and if the wait is not in a controlled environment the chain breaks in the last mile.
So picking and dispatch scheduling work backwards from the channel rules rather than forwards from what is convenient in the warehouse. That is why the channel specification is part of how the warehousing operation is designed.
Putting storage, delivery and documentation under one point of contact does not make the physics go away. It makes the handover someone’s responsibility, which is the only reason it ever gets shortened, recorded or improved. Contracted separately, the handover belongs to nobody.