The cold chain is a booking problem

The set point is written into the booking and into the contract, and it is the temperature the machine maintains for the air it delivers. For block frozen product the commercial standard is well below freezing with a stated tolerance, and the tolerance matters as much as the number.

Two mistakes recur. Loading product that is not already at temperature, because a reefer maintains temperature and does not pull it down from warm, so warm product arrives warm and everything downstream is a claim. And setting a temperature appropriate for chilled cargo on a frozen booking, which is a clerical error that ruins a container.

The machine logs its own performance and the data can be downloaded at discharge. On top of that an independent recorder placed inside the load gives the temperature of the cargo rather than of the delivery air.

Both outputs belong in the document file, handed over with the bill of lading and the certificates, not kept in a drawer by whoever happened to download them. A claim without a temperature record is an argument. A claim with one is arithmetic.

Where the value justifies it, the independent recorder should be placed by someone who is not the seller, and its position in the load should be photographed.

• At the port of loading, between the arrival of the packed container and the vessel: is it plugged, and who confirms it.
• At transhipment, which on many routes into West Africa and Southern Africa means a day or more in a hub port. This is where the chain is most often broken and least often checked.
• At the discharge port, between landing and customs release. Clearance delays and plug availability are separate problems that arrive together.
• On the inland leg, if there is one, and whether the vehicle is genset equipped or the container travels unpowered.

We ask these four questions at booking rather than at arrival, and on a programme we ask who at the terminal confirms the plug status and how that confirmation reaches the file.

Block frozen product in cartons behaves differently from individually frozen product in bags: the block is thermally slow and forgiving, the loose product is fast and unforgiving. Carton strength matters because a crushed carton in the middle of a stack restricts air flow, and restricted air flow is the mechanism behind most uneven-temperature claims.

Loading practice follows from that. The load has to be stowed so that the delivery air can travel: product off the floor channels where the design requires it, nothing blocking the return air path at the top, and the load not stacked above the red line painted inside the box for exactly this reason.

Less volume than a dry box of the same length, because of the insulation and the machinery, and on a typical frozen food loading plan about 25 tonnes net. That figure is the planning number for a programme: five containers a month is roughly 125 tonnes, ten is roughly 250.