Cold chain monitoring in Europe: continuous records, real alerts, read-only by design

The European cold chain landscape

Pharmaceutical wholesalers, supermarket chains, meat and dairy processors, third-party logistics warehouses, central kitchens and restaurant groups all depend on refrigeration that works around the clock. Many sites run dozens of cold rooms, display cases and freezers, often from different manufacturers installed over the years.

Most of that equipment already has an electronic controller, and many stores have a central system manager. What is often missing is a layer that watches every unit with the same rules, escalates alarms when nobody reacts, and produces one consistent compliance report across sites and brands.

• Pharmacies and pharmaceutical wholesalers: 2–8 °C products and controlled room temperature storage.
• Food retail: display cases, frozen islands, back-of-store cold rooms.
• Meat, fish and dairy processing: chilling, storage and freezing rooms.
• Logistics: temperature-controlled warehouses and refrigerated transport.
• Food service: central kitchens, hotels and restaurant chains.

The regulatory framework

For medicinal products, the reference is the European Commission Guidelines of 5 November 2013 on Good Distribution Practice of medicinal products for human use (2013/C 343/01). They ask for suitable equipment and procedures to check storage conditions, an initial temperature mapping of storage areas under representative conditions, monitoring devices placed where the extremes of fluctuation occur, and repeated mapping after a risk assessment or significant changes to the facility or its temperature control equipment. Transport must also keep and document the required conditions.

For food, Regulation (EC) No 852/2004 on the hygiene of foodstuffs requires food business operators to put in place, implement and maintain permanent procedures based on HACCP principles. Temperature control of chilled and frozen food is one of the most common critical control points, and specific temperature values are set by product-specific rules and by each Member State.

Friolink produces the continuous records, alarms and reports these frameworks call for. Compliance, however, also depends on each company's own processes: written procedures, sensor calibration, mapping studies, staff training and documented corrective actions.

October 2026: when refrigeration became a cybersecurity issue

On the night of 11 to 12 October 2026, refrigeration systems in hundreds of French supermarkets were tampered with remotely. According to press reports (BleepingComputer, CNN and others), the affected sites used Danfoss AK-SM 8xxA system managers running older software versions, reachable from the public internet; French media put the number of affected stores between 250 and 350, and the Paris prosecutor's office opened an investigation. Security researchers counted around 2,750 Danfoss systems exposed online, about 1,590 of them in France. At the time of writing, exactly how the controllers were compromised had not been officially confirmed.

The lesson applies to every brand, not just one: a controller that can be changed from the internet is a single point of failure. Two practical takeaways for any operator:

• Do not expose controller or system manager web interfaces directly to the internet; use segmented networks and controlled remote access, and keep firmware up to date.
• Keep an independent monitoring layer that only reads data. If a setpoint is changed, maliciously or by mistake, the temperature rise is still detected and escalated.

How Friolink works in Europe

Friolink is a cold chain monitoring platform developed in Uruguay. We operate from Uruguay and serve other countries with equipment shipped to the customer and remote connection. The platform, alerts and reports are available in English, French, Spanish and Brazilian Portuguese.

We measure temperature, humidity and door status with our own sensors (WiFi, Zigbee or BLE), or with a bridge that reads your existing controllers in read-only mode: it never writes setpoints or parameters. In production today we read KMC Controls controllers over BACnet at a logistics operator in Uruguay with 30 cold rooms. Other brands common in Europe, such as Danfoss, Carel, Dixell/Copeland, Eliwell or Bitzer, are available as integrations on request over Modbus or BACnet, assessed model by model.

Alerts reach people through the app, email, Telegram and WhatsApp, with automatic escalation: first the shift supervisor, then the technician and, if nobody acknowledges, a monitoring centre via Contact ID/SoftGuard. Rules can combine conditions, for example door open for 5 minutes and temperature above 10 °C.

Every reading is validated on two servers. If the internet connection drops, data stays on site and is sent when the link comes back, so the history has no gaps.

Reports, TV dashboards and prediction

Compliance reports are generated per period and per unit: minimum, maximum and average temperature, percentage of time in range, minutes out of range, excursions, door openings and who acknowledged each alarm. They are available as PDF per unit and as a HACCP record.

Kiosk mode turns any TV into a live dashboard for the control room or the loading dock. Predictive analysis learns compressor cycles, recovery time after door openings and thermal drift, and compares each unit with similar equipment using anonymised data only, so you can act before a slow failure becomes a loss.

A short checklist you can use today

Want to see how this would look across your sites? Tell us how many units you have and which controllers they use. We will reply with a proposal based on a monthly subscription per number of units. Use the contact form or write to [email protected].

• List every temperature-controlled unit, its range and its alert threshold.
• Check where probes sit compared with your mapping study's warmest and coldest points.
• Set alarm delays that ignore normal defrost cycles but catch real drifts.
• Define who responds at night and at weekends, and who is next if they do not.
• Review which controllers or system managers are reachable from the internet.
• Test the full alarm path end to end at least once a month.

Frequently asked questions