Cold storage temperature monitoring on a live, colour-coded floor plan

Cold-chain operators live with a paradox: the freezers are more reliable than ever, and the compliance paperwork is heavier than ever. The gap between “the equipment is fine” and “the auditor is satisfied” is where a spatial view earns its keep. A floor plan where every probe shows its reading lets the shift lead answer “is anything drifting?” at a glance — and every reading written to a table is kept in that row’s history for when QA has to prove it.

People say “temperature heatmap”, but what a cold-room team needs is usually simpler and more honest: each probe or room coloured by its own reading. Spatial Map does not interpolate between probes — a colour on the map always belongs to a real measurement. That is done with ordinary data links plus style rules:

Heatmap of activity is a density map: every record adds weight where it sits, regardless of its value. It is the right tool for questions like “where do excursions keep happening?” — for example, a table where each excursion or door-open event is a row with a position, shown with Change history (weighted by events) over the last 7 or 30 days.

Display modes in the New data link dialog

Every reading a probe emits is one point in space. If the probes are all clumped near the door because that is where the cabling reached, the map will be confidently wrong. Two rules cover most sites: one probe per bay or case at product height, and one near every airflow inlet and outlet.

• Bay corners or case ends at product height — the coldest and warmest spots.
• Airflow inlet and outlet — short-cycling shows here before it reaches product.
• The dock threshold — the first place warm air bleeds in.
• Any spot the last audit flagged — put a probe there permanently.

Our Retail Mart Store demo has a “Cold-chain temperature” layer fed by a sensor table with seven probes: two on the Dairy chillers (set point 3 °C) and five on the Frozen Food freezers (set point −18 °C). Each record carries temperature_c, setpoint_c, humidity_pct, battery_pct, door_open_events_today, status and last_reading_at, plus the probe’s position and floor.

The link is a Live location marker with a thermometer icon, labelled with the reading (for example 3.3 °C), placed on the right floor from the record’s floor field. At the moment of our capture one freezer island read −11.8 °C against a −18 °C set point — exactly the case a threshold rule should paint red. Hover a marker and the tooltip shows sensor name, location, reading, set point, humidity, door openings and battery.

Connect fields for a Live location marker

Most cold-chain sensor stacks already publish somewhere: a vendor cloud with a REST endpoint, a BMS that can export or call out over HTTP, or a gateway you control. Spatial Map has no MQTT, Modbus or OPC UA connector — it takes data over HTTP. Pick the shortest path.

1. 1
   Create a sensor table

   Tables → Create new table, kind Sensor. Add the fields your probes send — reading, set point, unit type, battery — plus a location name or longitude and latitude. The sensor ID makes a good row key.

2. 2
   Get readings in

   Let your gateway write rows with an API key from the table’s API & access tab (60 requests a minute per key; batch up to 1,000 rows per bulk call). Or, if the vendor cloud has a readings endpoint, use the table’s Pull from a REST API so our servers fetch it on a schedule.

3. 3
   Bind readings to the map

   In the editor: Data → Manage data links → New link. Pick the table and a display mode — Live location marker for probes, Color a shape or Value badge on shape for rooms — and match each record by coordinates or by name.

4. 4
   Add the colour bands

   In the editor’s side panel, Style rules → Add rule for drift and for excursion (next section), then tick both on the link. Add the reading, set point and battery to the Map tooltip.

Creating a table and picking its kind

It is tempting to configure a rule per probe. Don’t. Two bands cover nearly every real event: drift (the reading creeping above target) and excursion (a hard breach of the permitted range). A style rule can combine conditions with All match (AND), so one pair of rules per unit type is enough:

Rules run top-down and the first match wins, so put excursions above drift. Everything that matches no rule keeps the layer’s default colour — blue for “in range”. Use your own food-safety limits; the values above are examples.

Auditors want a continuous record of readings, a record of what happened when a limit was crossed, and a plan for what happens next. Every write to a table row is kept in that row’s history: open a probe in the viewer and its History lists each change with its time. The Data screen lets you filter records by source and time and Export CSV, and a report on the sensor table exports to PDF or Excel for the audit file.

If the map shows a warm case and the thermometer next to the probe is cold, work outward from the sensor: the probe first (battery, calibration date), then the bridge (last successful run on the Integrations Health tab, or the table’s last write), then the map binding (did someone rename the area the record matches?). Very often the answer is a tired probe battery — the demo’s records carry battery_pct for exactly that reason, and a “battery low” rule makes it visible.

For a store-floor version with shelf stock and RFID next to the cold chain, read retail shelf monitoring on a live store map. Try it on your own site with a 7-day free trial — no card needed, up to five facilities; Pro is $60 a month or $600 a year for up to four (pricing).