What is temperature mapping?

Temperature mapping, also called thermal mapping or a temperature distribution study, is a documented study that characterises how temperature is distributed across a storage or process space over time. Calibrated sensors are placed at defined locations, data is recorded for a defined period, and the results are analysed to find the warmest and coldest locations, to check that the whole space stays within its acceptance limits, and to decide where routine monitoring sensors should go.

Why it is done

A single monitoring probe only tells you the temperature at one point. Airflow, door openings, shelving, lighting and equipment all create gradients, so the product on a top shelf near a light fitting can experience a different environment from the product beside the evaporator. Mapping gives evidence that the space, as used, is capable of holding product within its label conditions — which is why guidance for the storage of temperature-sensitive pharmaceutical products asks for it.

What a study produces

• A protocol: purpose, acceptance limits, sensor locations, duration, loading condition, sampling interval.
• Sensor calibration evidence, so that readings are traceable.
• The raw data set and a data-quality assessment (gaps, duplicates, sensor faults).
• Statistics per sensor and overall: minimum, maximum, mean and spread.
• Identified hot and cold spots and any excursions, with duration and magnitude.
• A report with findings, limitations and recommended monitoring locations, reviewed and approved by a responsible person.

How the results are used

The hot and cold spots become the locations for permanent monitoring. Excursions drive corrective actions such as airflow changes, re-zoning of shelving, or exclusion of an area from storage. The study is typically repeated after significant changes and at a defined interval, and often under both empty and loaded conditions, and across seasons where ambient conditions matter.

ValiTrac's mapping engine implements the analysis part of this workflow — data quality, statistics, hot/cold spots, excursions and a draft report — behind an explicit human review gate. It does not replace the protocol, the calibration of the sensors or the reviewer's judgement. If you came looking for the analysis itself rather than an explanation of it, the temperature mapping workspace is where logger exports are uploaded.

What gets mapped

The same method covers a wide range of spaces, and what changes between them is the duration, the number of sensors and the acceptance limits rather than the principle. A pharmacy refrigerator and a 4,000 m2 ambient warehouse are both mapped; one takes three days and nine sensors, the other takes a fortnight and dozens.

• Pharmaceutical refrigerators and cold rooms held at 2-8 C, where the evaporator and the door are the two extremes.
• Freezers at -20 C and ultra-low freezers at -70 to -80 C, where opening the lid matters more than anything the controller does, and where the sensor and logger specifications are often the limiting factor.
• Ambient warehouses and distribution centres nominally at 15-25 C, which are the hardest case because the roof, the loading bay and the seasons all act on them.
• Incubators and stability chambers, where the set point is tight and the load itself changes the distribution.
• Transport vehicles and passive shippers, mapped as part of qualification rather than as a storage study: the space moves, and so does its worst case.

Temperature and humidity mapping

Where the product storage condition states a humidity as well as a temperature, both are mapped in the same study. Stability chambers run to ICH conditions that specify both, and warehouses holding moisture-sensitive or hygroscopic product often carry a humidity limit too. The method does not change: the same sensors at the same positions record a second quantity.

What does change is how much the second number is worth. Humidity sensors drift more than temperature sensors, their calibration is harder to make traceable, and their uncertainty is usually several times larger in relative terms. A humidity acceptance criterion that does not state the sensor uncertainty alongside it is not a criterion, because a reading of 62 % against a 60 % limit means nothing until you know whether the sensor is good to one point or to five.

Ovens and furnaces are a different discipline

Thermal surveys of heat-treatment furnaces and industrial ovens use the same idea, and people reasonably search for them with the same words: calibrated sensors at defined positions, a survey over a defined period, acceptance against a stated criterion. The governing documents are different ones. AMS2750 covers aerospace heat treatment and CQI-9 covers automotive, rather than EU GDP or GMP Annex 15.

Those standards set their own requirements for thermocouple types and usage limits, instrument calibration intervals, survey frequency and furnace classification, and the numbers are specific to the current revision. This article does not reproduce them. A tolerance or a class limit quoted from memory is exactly the kind of figure that ends up copied into somebody's survey report, and the only safe source for it is the standard itself, at the revision your customer's specification names.

What the study can and cannot tell you

A mapping study is a snapshot of a space under a stated set of conditions. That is a strong result and a bounded one, and reports get into trouble by implying more.

The fourth row is the one people slide past. Sensors in a mapping study measure the air at product height, not the product. For a fast-responding item that is a fair proxy; for a pallet of dense goods it is conservative, because the product's thermal mass smooths what the air does. If a claim about product temperature is needed, that is a different study with sensors in or against the product, and the protocol has to say so.

How mapping and monitoring fit together

The two are often confused because both produce temperature data. Mapping is an occasional, dense study that establishes where the extremes are. Monitoring is a continuous, sparse record at a few fixed points that shows the space has stayed in control since. Each one is nearly useless without the other: monitoring at positions nobody has justified measures a place that may not matter, and a mapping study never followed by monitoring describes a week that ended years ago.

The link between them is explicit in regulation. The monitoring sensors go where the mapping said they should, and the monitoring data at those positions is the evidence for the periodic review that decides whether the mapping still holds.

The lifecycle

• Protocol written and approved, with limits, sensor plan and analysis rules fixed in advance.
• Sensors calibrated, clocks synchronised, positions placed and photographed.
• Data collected over a duration that covers the operating cycle, then checked for quality before any statistic is computed.
• Analysis against the protocol: per-sensor statistics, hot and cold spots by the stated criterion, excursions listed with duration and magnitude.
• Report drafted, marked not approved, reviewed by a competent person who did not do the analysis, then signed.
• Actions implemented — monitoring probes moved, exclusion zones marked, layouts changed — and the qualified state maintained by periodic review until a change or the review says otherwise.