In short: Air cargo handling is the most data-dependent operation on a UK airfield and usually the least connected. The cargo shed sits outside the terminal network, runs three shifts against customs and security deadlines, and loses handheld sessions at exactly the point where the record has to be perfect.
Key Takeaways
- The shed is the deadline, not the aircraft — Cut-off times for acceptance, screening and build-up are set hours before departure, so a handheld that drops out during build-up costs a flight, not a minute.
- Compliance runs on timestamps — ACC3 and regulated agent obligations, customs declarations and the Border Target Operating Model all rest on data captured at a point in the shed by someone holding a scanner.
- Cargo buildings defeat Wi-Fi by design — Roller doors open to the apron, aluminium ULDs stacked three high, elevating transfer vehicles moving through steel racking and multiple handler tenants running competing WLANs in one building.
In a nutshell
What does air cargo handling actually involve?
Air cargo handling is everything that happens to freight between a lorry backing onto a landside dock and a loaded unit load device going up the ramp into an aircraft hold. Acceptance and paperwork check. Weighing and dimensioning. Security screening, usually x-ray or explosive detection, sometimes screening dogs. Storage, which for the bigger operators means an automated racking system with elevating transfer vehicles rather than a floor stack. Build-up onto pallets and into containers, netting, weighing again, and the airside movement to the aircraft stand.
A large share of UK air cargo never sees a freighter at all. It moves in the belly holds of passenger aircraft, which is why Heathrow handles the highest value of UK air freight without being a dedicated freight airport, and why the cargo operation there is spread across the ageing Horseshoe area rather than a purpose-built modern facility. East Midlands Airport is the counterexample: the UK's busiest dedicated freight airport, home to DHL's UK hub and a major Royal Mail and UPS operation, working hardest between 11pm and 5am when the rest of the airfield is quiet.
The handlers doing the work are mostly not the airport. Menzies Aviation, Swissport, dnata and WFS operate sheds under lease, which is a detail that matters enormously for connectivity, because it means the building often has three or four tenants, each with their own IT function, each having solved the wireless problem separately, and none of them owning the fabric.
Why does the cargo shed break wireless networks?
Because the building is a metal box with holes in it, and the holes are the busy part.
A typical cargo shed is a high-bay warehouse with roller shutter doors down the landside face for lorries and down the airside face for tractor and dolly trains. During a build-up wave, most of those doors are open, so the RF environment is continuous with the apron. Inside, freight is consolidated into aluminium ULDs, and a stack of AKE containers or PMC pallets is a wall of metal that moves several times an hour. The larger sheds run automated storage systems where transfer vehicles travel steel racking under computer control, so the reflective geometry of the building changes minute by minute.
Add the screening hall, where x-ray machines and the associated shielding sit in the middle of the freight path. Add cold rooms for pharmaceutical and perishable freight, which are insulated metal chambers with their own coverage problem and their own temperature record. Add the fact that the airport's own network was designed around passengers and stops, in practice, at the terminal boundary, so the cargo estate is served by whatever the handler could get a fibre tail to plus a mesh of access points bolted to racking legs.
The observable symptom is always the same. Handhelds work in the office and the acceptance lane, degrade in the racking, and fail at the airside door. Staff adapt by writing on paper and keying it in later, which works, and which quietly destroys the timestamp accuracy that the whole compliance model rests on.
What breaks when the data lags the freight?
Three regimes, all of which assume the record is made where the event happens.
Security first. Cargo flying into the UK from outside the country requires a validated ACC3 or regulated agent chain, and cargo originating here moves under the DfT's regulated agent regime, which means every consignment carries a documented security status and a record of who screened it, how and when. Consignment security declarations are increasingly electronic. When screening data is captured on paper at the machine and entered at a terminal twenty minutes later, the audit trail still exists, but the operator has lost the ability to prove sequence, and sequence is the thing an inspector probes.
Customs second. Export and import declarations run through the Customs Declaration Service, with the inventory-linked systems at UK air cargo hubs matching declarations against the freight physically present. A mismatch between what the inventory system believes is in the shed and what is actually on the pallet is a hold, and holds at an air cargo shed are expensive because the aircraft does not wait. The Border Target Operating Model has added sanitary and phytosanitary checks at Border Control Posts for a wider set of goods, which puts more consignments into a physical inspection queue and raises the value of knowing precisely where each one is.
Third, the commercial layer. IATA's e-AWB has been the default for years, and Cargo iQ milestone reporting means forwarders and airlines expect status events at defined points: received from shipper, screened, ready for carriage, on hand at departure. A handler who reports those milestones late, or in a batch at the end of a shift, looks worse than one who reports them as they happen, even when the freight moved identically. Pharmaceutical freight under IATA CEIV Pharma raises the bar again, because the temperature record has to be continuous through the shed, the cold room and the ramp transfer.
What about tracking the ULDs themselves?
A ULD is an aircraft part with a serial number, and the industry loses track of a remarkable number of them. Unilode and Jettainer have both fitted large parts of their container fleets with Bluetooth tags, which solves the identification problem and creates a backhaul problem: a tag is only useful when something reads it and reports the read.
That is a clean case for site-wide connectivity, because ULD movements cross every boundary in the operation. A container is built in the shed, moved airside on a dolly, loaded, flown, unloaded at the other end, returned empty, and stored on a stack somewhere on the estate. Readers on the shed doors and the airside gates, connected continuously, turn that into a location history. The same infrastructure reads temperature tags on pharma shipments and tracks the tugs, dollies and forklifts, which are the assets handlers most often cannot find during a wave.
Why is the airside boundary the hard bit?
Because it is where the airport's network, the handler's network and the airline's systems all stop, and where the freight physically changes hands.
A tractor pulling a train of dollies from the shed to a stand crosses several hundred metres of apron. That is precisely the ground public mobile coverage handles badly, since apron areas sit under the shadow of terminal buildings and aircraft, and it is precisely where the loading supervisor needs the load instruction, the final weight and balance figures and the ability to report a change. Most operations resolve this with VHF radio and shouted numbers, which is why load sheets get amended in the aircraft door.
A single private 5G network across the cargo estate covers the shed, the truck docks, the apron road and the stands with one set of credentials, licensed spectrum under an Ofcom Shared Access licence, and quality of service the airport controls rather than shares with ten thousand phones in the terminal. It also gives the airport something politically useful: one network the handlers subscribe to, rather than four overlapping tenant WLANs interfering with each other inside a shared building.
The trade-off worth stating plainly is ownership. Handlers hold short leases and are reluctant to fund fixed infrastructure in a building they may leave. Airports own the fabric but do not run the cargo process. The deployments that work are the ones where the airport builds the network as estate infrastructure and sells connectivity to the handlers as a service, in the same way it sells power and floor space. Where each handler is left to solve it alone, nobody does.
Where should a UK airport start?
With the cargo estate's coverage map, which very few airports have, because the cargo area is usually outside the scope of the terminal network surveys.
Then take one measurable process. Truck turnaround at the landside dock is the easiest to instrument and the one hauliers complain about most. Milestone timeliness against Cargo iQ is the easiest to sell to the airlines. Either gives a number that moves within a quarter, on infrastructure that stays useful when the next project arrives.
Regional airports have the better version of this argument. Bristol, Liverpool, East Midlands and Prestwick all run cargo operations at a scale where a single private network covers the entire freight estate, and where the same radios also reach the GSE charging area and the maintenance apron. Building it once, for the whole airfield, costs less than the terminal Wi-Fi refresh most of them budgeted for last year.