Fazley RabbiThe problem: you find out too late
"The fridge stopped overnight and we only noticed in the morning." For a pharmacy, clinic, cold room or food business, that sentence can mean stock thrown away and, for medicines and vaccines, a risk to patients. Manual checks a few times a day help, but they cannot catch what happens at night, on holidays or during a long power cut.
A temperature monitoring system watches each unit continuously and alerts the right people when something goes wrong. It is an extra layer of protection. It does not replace a good fridge or cold room, regular servicing by qualified technicians, your written storage procedures, or the rules that apply to medicines and vaccines in your field.
How such a system works
In plain terms there are four parts:
- Sensors inside each unit measure temperature, and sometimes humidity or whether the door is open. In medicine and vaccine fridges, a probe is sometimes placed in a small bottle of liquid so it follows the temperature of the stored products rather than every brief swing in air temperature when the door opens.
- A small device reads the sensors and stores the readings.
- Connectivity sends the readings to a dashboard, usually over Wi-Fi or a mobile SIM.
- Alerts reach phones by SMS, phone call, app notification or WhatsApp, depending on how it is set up.
You also get a temperature history for each unit, which helps with record keeping. Ask your regulator or supervising authority which records you are required to keep and in what form; do not assume a dashboard meets every requirement.
Planning for power cuts and network gaps
In Bangladesh, a monitoring system that stops when the power goes is not much use. Plan for:
- A battery in the monitoring device, so it keeps measuring and sending during an outage.
- A power-failure alert, so you know the moment mains power drops, before the temperature starts to rise.
- Readings stored on the device and uploaded when the network returns, so the record has no gaps.
- An offline alert. If the device stops reporting, that silence should itself trigger a warning.
- A second connection path where the signal is weak, such as Wi-Fi with a mobile SIM as backup.
Decisions you need to make
- Which units to monitor, and who is responsible for each.
- The allowed range for each unit. Take it from the product manufacturer's instructions and the rules you follow, not from guesswork or from an article like this one.
- Alert delay. A short delay prevents false alarms every time the door opens, but it should not be so long that a real problem waits.
- Who gets alerted, and in what order. If the first person does not respond, the alert should move on to the next.
- What to do when an alert arrives. A written step-by-step: check the unit, move stock to a backup fridge, call the technician, record what happened.
Common mistakes
- Placing the sensor right next to the cooling element or the door, where readings are not representative.
- Sending every alert to one person who may be asleep or travelling.
- Never checking sensors against a reference thermometer. Sensors drift; have them checked on a regular schedule.
- Testing only on installation day. Repeat the tests: unplug the device, leave the door open, switch off the router.
Your part and the expert's part
You decide the units, the ranges, the responsible people and the response procedure. An IoT developer helps choose reliable sensors, build or configure the device, and set up backup connectivity, dashboards and alert escalation. Refrigeration repairs and calibration should stay with qualified technicians. This is the kind of system I build under IoT device development.
Start small
Pick the one fridge or cold room where a failure would hurt most. Monitor it for a month with alerts going to two people, and test the power-cut and offline alerts on purpose. Once it works reliably, extend it to the other units. A small IoT pilot is the right size for this. If you want to talk it through, get in touch.