Why measure at all
A refrigerated truck or container has its own thermostat, but the display on the unit only shows what the unit senses, usually in the air it is circulating. It does not show what is happening inside the cargo, at a pallet in the corner, or during the minutes a door stood open at a dock. Temperature monitoring fills that gap by placing sensors in the load and recording what they experience.
Monitoring serves three purposes. It detects problems while there is still time to act, it documents conditions for the receiver and regulators, and it provides evidence if the cargo is later disputed or claimed against. The broader design of temperature-controlled supply chains is covered in the cold chain overview, and the equipment itself is in refrigerated transport.
Types of devices
Data loggers. A small device with a sensor, a clock and memory. It records the temperature at set intervals and stores the readings. Nobody sees the data until the logger is retrieved and read, often at the destination. Some are single-use and some are reusable and calibrated. Many print or display a simple pass/fail summary, and others connect to a computer or phone to download a graph. Loggers are inexpensive per shipment, but they tell you about a problem only after the fact.
Real-time trackers. These add a communication link, usually cellular and sometimes satellite, and send readings and location during the trip. Because they report as they go, they enable alerts and intervention. They cost more and depend on coverage. Developments in 5G connectivity and the Internet of Things are making such devices smaller and cheaper, though coverage gaps at sea, in remote areas or inside metal containers remain practical limits.
Indicators. Simple chemical or color-changing labels show whether a threshold was crossed, without any record of when or for how long. They are cheap and useful as a backup or a quick visual check, but they carry far less information than a logger.
Built-in unit sensors. Reefer controllers log their own supply and return air and set point. These records are useful, but because they measure air at the unit rather than product temperature, they are often complemented by independent sensors.
What the sensor actually measures
There are two main choices.
- Ambient or air sensors measure the air around the product. They respond quickly to changes, such as an open door.
- Probe or product-simulating sensors measure the product itself or something that mimics its thermal mass, such as a sensor in a small glycol-filled vial. Product temperature changes more slowly than air, so a brief spike in air temperature may not mean the product warmed much.
Which to use depends on what needs protecting and on any applicable guidance. Documenting the choice makes the readings easier to interpret.
Placement
Temperature is rarely uniform in a load. Spots near the door, the top layer, the sides and the middle of a pallet may differ, and the warmest and coldest points are often not where one would guess. Common practice is to place sensors in locations considered at highest risk, such as near doors or away from airflow, and sometimes to use more than one. For high-value cargo, planners may run a test shipment to find where the extremes occur. The sensor should not be buried in a way that makes it read unnaturally, nor placed against a cooling outlet where it will read colder than the cargo.
Sampling interval
The recording interval is a trade-off. Reading every minute catches short spikes, but fills memory and uses battery. Reading every hour could miss a brief event. Intervals of several minutes are commonly chosen for transit, but the right figure depends on the product’s sensitivity and the length of the trip. Longer trips need either a longer interval or a device with more capacity and battery.
Setting limits and alerts
Each product has an allowed range, and the alert thresholds are usually based on it. Some points to consider:
- High and low limits. Many products can be harmed both by warmth and by freezing.
- Time delays. A reading that exceeds the limit for a few seconds when a door opens may not matter, while a sustained rise does. Alerts that trigger on every blip create alarm fatigue, and people begin to ignore them. Rules that depend on duration help.
- Warning and critical levels. A warning threshold slightly inside the true limit gives time to react before the product is actually at risk.
- Who gets the alert. An alarm is only useful if someone receives it and can act, such as by calling the driver, redirecting to a cold store or arranging a replacement unit. Contact lists and procedures should be agreed before the trip.
Calibration and trust
A sensor is only as reliable as its calibration. Devices are often checked against a reference, and certificates are kept. A logger with an expired calibration can undermine the entire record. Clocks also matter, since timestamps link readings to events such as loading and handoff.
After the trip: reading the data
On arrival, the receiver reviews the readings. If they stay within range, the shipment is generally accepted. If they show an excursion, the next steps depend on the product and agreements. For food, the receiver may inspect, test, accept at a discount, or reject. For medicines, the product is usually quarantined until a qualified decision is made, as described in pharmaceutical logistics. In either case, a clear record helps decide who bears the loss, and it supports freight claims. Records also feed continuous improvement: patterns of excursions at a certain dock or carrier point to where the process needs fixing.
Large operators sometimes combine this data with analytics and cloud-based systems to spot weak lanes and recurring problems. Better data can improve supply chain visibility, but the improvement comes from acting on the data, not just collecting it.
Limits of monitoring
Monitoring does not cool anything, and it does not guarantee quality. A logger showing good readings does not prove the product was handled hygienically or kept in sound packaging. Readings can also be misleading if a sensor is misplaced, if it was inside a cold pocket while the rest of the load was warm, or if it was moved. Monitoring is best viewed as one layer alongside good equipment, good loading and trained people.
Takeaway
Choose the device for the risk: loggers for records, real-time trackers when intervention is possible, indicators as inexpensive backups. Place sensors thoughtfully, set realistic limits, agree on who responds, and keep the calibration and records in order.