.png)
How to Design a Multi-Site Fridge and Freezer Monitoring System
A fridge alarm is only useful if it tells the right person, at the right time, about the right cabinet. That sounds simple until the system has to cover a prep kitchen, two bars, a retail display area, several food trucks and an off-site storage facility.
Multi-site temperature monitoring fails when it is designed like a gadget rollout instead of a food safety system. The goal is not to scatter sensors everywhere. The goal is to know, with confidence, whether chilled and frozen stock is being held within your required limits, and to prove it later.
This guide explains how to work out the number of sensors and gateways you need, where probes should go, how to plan wireless coverage, and how to set up alarms and reporting across multiple locations.
Start with a complete cold chain map
Before choosing sensor quantities, list every place where chilled or frozen goods sit for more than a short handling period. Do this site by site, not from memory. A floor plan helps, but a walk-through is better because teams often add underbench fridges, display cabinets or freezers that never make it onto asset registers.
Group each asset by what it does:
Area | Typical assets to include | Monitoring note |
Kitchens | Upright fridges, underbench fridges, prep fridges, freezers, coolrooms | High door opening and heat from cooking equipment can affect readings |
Bars | Bottle fridges, keg coolers, ice cream freezers, garnish fridges | Often crowded with stainless steel and glass door units |
Retail areas | Open displays, grab-and-go cabinets, deli cases, ice cream freezers | Product temperature can vary from front to back |
Food trucks | Upright or bench fridges, chest freezers, prep wells | Power, movement and mobile coverage matter |
Storage facilities | Walk-in coolrooms, freezer rooms, pallet storage, dispatch fridges | Larger spaces may need more than one probe |
For each item, record:
Asset type and size
Location and site name
Temperature requirement
Door opening frequency
Whether it has separate compartments
Power source
Existing internet or mobile coverage nearby
Who responds to alarms at that site
For chilled potentially hazardous food in Australia, the common food safety benchmark is 5°C or below. For frozen storage, many businesses use an operational target around -18°C, though the key legal principle is that frozen food remains frozen. Use the limits set by your food safety program, product requirements and internal policy.
Calculate the number of sensors by temperature zone
The simplest rule is this: monitor each separate temperature-controlled volume.
That usually means one sensor per fridge, freezer, display case or coolroom zone. Do not count sensors by room. A kitchen with one coolroom, three upright fridges, two underbench fridges and one freezer is not “one kitchen sensor”. It is at least seven monitored zones.
Use this baseline:
One sensor for each single-compartment fridge
One sensor for each single-compartment freezer
One sensor for each underbench or prep fridge
One sensor for each retail display cabinet
One sensor for each bar fridge or keg cooler that stores perishable stock
One sensor for each food truck fridge or freezer
One sensor for each walk-in coolroom or walk-in freezer as a minimum
Then add sensors where one reading cannot represent the whole space.
Add extra probes for large or uneven spaces
A large coolroom, freezer room or long retail case can have warm and cold spots. Air near the evaporator may be colder. Areas near doors may warm during service or delivery. Product near the front of an open display can sit in a different air stream to product at the back.
Add extra sensors when:
The room is large enough that stock sits far from the cooling unit
Doors open often during service or receiving
Product is stored on high racks and low pallets
There are known warm spots
The unit has multiple doors or long display runs
Different product types need different limits
For a small walk-in coolroom, one well-placed probe may be enough. For a large storage coolroom, two or more probes are often better: one near the warmest likely product area, and one that confirms general room performance. In a freezer room with racking, place probes where they reflect product conditions, not just the coldest air outlet.
Treat separate compartments as separate assets
Many cabinets look like one unit but hold separate zones. A dual-temperature unit, a fridge-freezer combo, or a prep bench with independently controlled wells may need more than one sensor.
Count each controlled compartment separately if:
It has a separate thermostat
It has separate airflow
It can fail independently
It stores a different risk category of product
Staff use one section more heavily than another
This matters in bars and retail areas, where compact units often mix storage, service and display functions in one footprint.
Factor in redundancy for critical stock
Some stock is more expensive or more sensitive. A bulk freezer in a warehouse, a seafood coolroom, a vaccine fridge, or a central production coolroom might justify a second sensor.
Redundancy is not about clutter. It gives confidence when a reading looks wrong. If one probe reports a rise and a second confirms it, the alarm is easier to trust. If only one probe changes, staff can check for probe damage, door issues or local airflow effects.
For critical assets, a useful design is:
One primary probe in the normal monitoring position
One secondary probe in the likely warmest product area
Independent alarm rules if needed
Place probes where they represent product risk
Sensor count is only half the design. Poor placement creates false confidence or false alarms.
A probe should measure conditions that reflect the stock, not the cooling system. Avoid placing probes where they will read artificially cold, warm or unstable.
Good probe positions
Use these principles for most fridges and freezers:
Place the probe in the middle third of the cabinet where practical
Keep it close to stored product, without burying it in food
Mount it away from fans, evaporator outlets and defrost heaters
Keep it away from doors, gaskets and direct outside air
Secure it so staff cannot knock it into the wrong place
Allow airflow around the probe
In a fridge, a small bottle of food-safe glycol or another approved thermal buffer can help the sensor behave more like food than air. This reduces nuisance alarms from short door openings. If you use a buffer, make sure the alarm delay and reporting method still support your food safety process.
In a freezer, use a probe rated for the expected temperature and moisture conditions. Cables, seals and mounting clips need to survive frost, cleaning and door movement.
Poor probe positions
Avoid these common mistakes:
Hanging the probe near the door
Taping it to a metal wall
Sitting it directly under the cold air outlet
Dropping it loose into a drawer
Placing it behind stock where staff cannot inspect it
Mounting it where cleaning knocks it loose
Running cables through crushed door seals
A bad location can hide a real problem. A probe beside the evaporator may show safe temperatures while product near the door warms above limit. A probe near the door may trigger constant alarms even when product remains acceptable.
Special cases for displays and prep units
Retail displays and prep fridges need more care because the cold zone may not be even.
For open retail cases, monitor the area most likely to warm, often near the front or top, while still following the manufacturer’s load line. For long displays, use more than one probe if the case has uneven airflow or separate sections.
For prep fridges, place the probe where it reflects the chilled storage space, not directly beside uncovered pans during service unless that is the risk you intend to monitor. If the top rail and base cabinet behave differently, treat them as separate zones.
For food trucks, mount probes so they stay fixed during travel. Loose probes can shift behind stock or against freezer plates, giving unreliable readings.
Work out how many gateways each site needs
Sensors collect readings. Gateways move those readings to the cloud or central platform. The number of gateways depends on wireless range, building materials, interference, power and internet access.
A single gateway can often receive from several sensors, but only if the radio path is reliable. Commercial kitchens and storage sites are harsh wireless environments. Stainless steel, coolroom panels, concrete walls, basements, refrigeration motors and dense stock can all reduce signal.
Start with this process:
Group sensors by physical area
Place kitchens, bars, retail zones, receiving coolrooms and storage rooms into logical clusters.
Check the gateway’s supported sensor capacity
Each gateway has a maximum number of devices it can handle. Stay below that limit, especially on high-risk sites.
Test signal from real probe locations
Do not test from the middle of an empty room. Test with the sensor inside the fridge or freezer, door closed, stock in place if possible.
Allow margin
A signal that barely works during setup may fail during a busy service, after a delivery, or when the room layout changes.
Add gateways where the building blocks signal
A second gateway is cheaper than missed readings and alarm doubt.
Typical gateway planning by area
For a small kitchen with several nearby fridges, one gateway may cover the area if it has clear signal and stable internet.
For a venue with a kitchen, multiple bars and a retail fridge area, plan at least one gateway per separated area if walls, distance or coldroom panels reduce signal. Bars often need their own gateway because bottle fridges sit under counters, behind stainless steel and near busy equipment.
For storage facilities, gateways should sit outside coolrooms where practical, with sensors transmitting out through the door, wall path or approved antenna arrangement. Do not place consumer-grade gateways inside freezer rooms unless the hardware is rated for that environment.
For food trucks, use a mobile-ready gateway with cellular data or a system that can store readings when offline and upload later. The gateway also needs reliable power. If the truck disconnects from mains power overnight, the monitoring design must account for battery life and alarm behaviour.
Design alarms around real response times
Alarm settings should catch unsafe conditions without training staff to ignore alerts. If every door opening creates an urgent alarm, people will stop trusting the system.
Set alarms using three layers: threshold, delay and escalation.
Set clear thresholds
Use thresholds that match the asset’s purpose and your food safety program. A chilled storage fridge may alarm above 5°C after an allowed delay. A freezer may alarm if it rises above your chosen frozen storage limit.
Some teams also use warning levels. For example, a fridge might send a warning before it reaches the critical limit. That gives staff time to check a door, move stock or call maintenance before the situation becomes serious.
Use delays carefully
A short spike during loading is not the same as a failed compressor. Alarm delays help separate normal use from real risk.
Use shorter delays for:
Critical stock
Small fridges that warm quickly
Assets with a history of failure
Overnight storage where no one is opening doors
Use longer but still controlled delays for:
Busy service fridges
Delivery coolrooms
Retail cases during restocking
Keep delays documented. If a regulator, auditor or manager reviews the data, the reason for each alarm rule should make sense.
Escalate to the right people
A centralised system should not send every alert to every manager. Build an escalation path.
A practical sequence might be:
Site staff receive the first alert
The duty manager receives the next alert if no one acknowledges it
Area management or maintenance receives escalated alerts
After-hours alerts go to the person who can physically act
Every alarm should show:
Site name
Asset name
Current temperature
Alarm limit
Time in alarm
Last reading time
Suggested action or response note
Also monitor system health. A fridge that stops reporting is a risk, even if the last reading was safe. Add alarms for low battery, sensor offline, gateway offline and power loss where supported.
Build a naming system before installing anything
Central reporting becomes messy fast if assets are named casually. “Freezer 1” means little when the business has 12 sites.
Use names that make sense without local knowledge. A clear format could be:
`Site - Area - Asset Type - Asset Number`
For example:
`Sydney CBD - Main Kitchen - Upright Fridge - 01`
`Brisbane West - Bar 2 - Bottle Fridge - 03`
`Melbourne North - Truck 4 - Chest Freezer - 01`
`Perth Storage - Freezer Room - Probe A`
Keep the label on the physical asset the same as the name in the platform. If staff receive an alert for `Bar 2 - Bottle Fridge - 03`, they should be able to find it in seconds.
Record these details for each sensor:
Field | Why it matters |
Sensor ID | Confirms the installed device matches the platform |
Asset name | Makes alarms and reports understandable |
Probe location | Helps interpret readings |
Gateway connection | Shows how data leaves the site |
Alarm limits | Links readings to required action |
Responsible team | Makes escalation clear |
Calibration or check date | Supports quality records |
Use centralised reporting to manage exceptions
The value of multi-site monitoring is not just live alerts. It is the ability to compare sites, spot repeat issues and produce records without chasing paper logs.
A good central dashboard should show:
Current status for every monitored asset
Sites with active alarms
Sensors that have stopped reporting
Temperature history by asset
Alarm acknowledgement and corrective action notes
Exportable reports for audits
User permissions by role and site
Managers should be able to filter by site, asset type or alarm status. Maintenance teams should be able to see repeat alarms that suggest door seal failure, icing, poor loading practice or equipment nearing failure.
For compliance records, focus on exception reporting. A full graph is useful, but the key questions are usually simple:
Did the asset stay within the required range?
If not, when did it go out of range?
Who responded?
What action did they take?
Was stock assessed or moved?
When did the asset return to range?
Plan for power, cleaning and maintenance
A monitoring system has to survive daily operations. That means thinking about practical details during design.
For sensors, check battery access, cold temperature rating, water resistance and cleaning exposure. A sensor under a prep bench may face splashes, knocks and chemical cleaning. A freezer probe may face ice build-up and brittle cable stress.
For gateways, confirm power points, internet connection, cellular signal if needed, and safe mounting locations. Avoid placing gateways where staff unplug them for appliances or chargers.
Set a routine check schedule:
Confirm probes are still in position
Check batteries and offline devices
Review alarm contacts after staff changes
Test escalation paths
Compare readings against a reference thermometer as required by your procedures
Update the asset register when equipment moves
Food trucks need extra checks because their setup changes often. Confirm the gateway powers on when the truck operates, stores data when coverage drops, and alerts the right person when parked after hours.
A practical example across several sites
Imagine a small group with three fixed venues, four food trucks and one storage facility.
The fixed venues each have:
Main kitchen coolroom
Kitchen freezer
Three underbench fridges
Two upright fridges
Two bar fridges
One retail grab-and-go cabinet
That is at least 10 sensors per venue, before adding any extra probes for large coolrooms or critical stock. If the coolroom is large or has a warm loading side, add a second probe.
Each food truck has:
One prep fridge
One underbench fridge
One chest freezer
That is three sensors per truck. Each truck may need its own mobile gateway unless the monitoring system uses another tested mobile data method.
The storage facility has:
One large coolroom
One freezer room
One dispatch fridge
That might start at five sensors: two in the coolroom, two in the freezer room and one in the dispatch fridge. More may be needed if the rooms are long, heavily racked or divided into different product zones.
Gateway counts would be tested on site, but a sensible starting design might include:
One or two gateways per fixed venue, depending on distance between kitchen and bar areas
One mobile gateway per food truck
Two or more gateways at the storage facility, depending on building layout and freezer room signal
The final design comes from testing. The asset list gives the sensor count. The signal survey confirms the gateway count.
What a good design looks like
A well-designed multi-site fridge and freezer monitoring system is easy to understand from the dashboard and easy to act on during a busy shift. Each fridge, freezer, cabinet, truck and storage room has a clear identity. Probes sit where they reflect product risk. Gateways have enough coverage margin to keep data flowing. Alarms reach the person who can fix the problem. Reports show what happened and what action followed.
Start with the cold chain map, not the hardware catalogue. Count every controlled temperature zone, add probes where one reading is not enough, test wireless coverage with doors closed, then build alarm rules around real response times. That approach gives the system a practical job: protect stock, support food safety and make multi-site reporting far less painful.
