How to Temperature-Map a Warehouse or Cold Room: A Step-by-Step Guide

To temperature-map a warehouse or cold room, you write a protocol that defines the space and acceptance criteria, distribute calibrated data loggers in a three-dimensional grid weighted toward expected worst-case points, run them continuously for a representative period, then analyse the data against your criteria and document the hot spots, cold spots and recommended monitoring locations in a report. Done properly, the whole exercise is a controlled study with a written protocol before it starts and a signed report after it finishes, not an ad-hoc scatter of sensors. The steps below walk through how a study is structured for a Singapore facility working to HSA GDP (GUIDE-MQA-013) or GMP expectations, the WHO TRS 961 Annex 9 protocol and its Supplement 8 on temperature mapping of storage areas, and, for chamber work, the performance-confirmation methods of IEC 60068-3-5 and IEC 60068-3-11.

Step 1: Define the space and write the protocol

Every study begins on paper. Before a single logger is placed, the protocol should record the internal dimensions and volume of the space, its intended storage conditions (for example 2 to 8 degrees Celsius for a pharmaceutical cold room, or 15 to 25 degrees Celsius for an ambient store), the acceptance criteria that define a pass or fail, the number of loggers and their planned positions, the study duration, and the tests to be performed. A written protocol matters because it fixes the acceptance criteria in advance. Deciding what counts as a pass after you have seen the data is exactly the kind of thing an auditor will challenge.

Step 2: Decide how many loggers and where they go

The number of loggers scales with the volume and complexity of the space. A small cold room might need nine to fifteen loggers, while a large warehouse can need dozens. The guiding principle is that the grid must be dense enough to characterise the whole volume and must deliberately over-sample the places most likely to fail. We cover the counting logic in detail in our sensor placement guide, but the essentials are:

• Build a three-dimensional grid: cover the floor plan and stack sensors at high, middle and low levels, because the vertical temperature gradient is often the largest source of variation.
• Put the corners and geometric extremes of the space in the grid.
• Add extra loggers at expected worst-case points: next to doors and loading docks, directly in and out of the evaporator air stream, against external or sun-exposed walls, near light fittings or heat-generating equipment, and in stagnant corners with poor airflow.
• Place at least one logger beside your existing permanent monitoring sensor, so you can compare the mapped reality against what your control system reports.

Step 3: Use calibrated loggers and record the traceability

A mapping study is only as trustworthy as its sensors. Every logger must carry a current calibration certificate, and the calibration should be traceable to national standards. At Unitest, mapping loggers are calibrated under our SAC-SINGLAS accreditation, LA-2023-0845-C, so the measurement chain is accredited and traceable even though the mapping study itself is a documented service rather than an accredited activity. Record each logger's serial number, calibration due date and its assigned grid position, so the report can tie every data trace back to a specific, calibrated instrument at a known location. If your loggers are due, our temperature calibration service and humidity calibration service can bring them back into traceability before the study.

Step 4: Run the study for a representative period

Set the loggers to record at a sensible interval, commonly every one to five minutes, and run the study long enough to capture representative behaviour. For a chamber or small cold room, a minimum of 24 to 72 hours of continuous logging is typical. For a warehouse, a full 7-day run is common so that the study spans a complete working week, including quiet nights, busy goods-in mornings and the weekend when doors stay shut and the HVAC may run differently. In Singapore, where the external heat and humidity load is high year-round, it is good practice to run studies that capture both the coolest overnight conditions and the hottest part of a working day.

Step 5: Perform the worst-case and open-door tests

A steady-state map tells you how the space behaves when nothing is going wrong. The stress tests tell you whether it can survive when something does. Two are almost always worth including:

• Open-door test. Hold the main door open for a defined period during a busy simulation, then measure how far the temperature drifts and how long the space takes to recover once the door is closed. This is directly relevant to any store with frequent goods movement.
• Power-loss or recovery test. Where it can be done safely, interrupt the cooling for a defined period and record how quickly the space loses control and then recovers. This tells you how much time you really have during a power event before product is at risk.

These tests turn a mapping study from a snapshot into a genuine risk assessment of the space.

Step 6: Analyse the data against the acceptance criteria

Once the run is complete, download every logger and analyse the full dataset against the protocol's acceptance criteria. The analysis should identify the warmest and coldest points by location and time, confirm whether the whole space stayed within range for the whole study, and quantify how the space responded to the open-door and recovery tests. Any excursion is investigated: was it a real problem, a door event, or a logger placed too aggressively in a worst-case spot that is not actually used for storage? The conclusion is a clear pass or fail against the pre-agreed criteria, not a subjective impression.

Step 7: Report, then place your permanent monitoring sensors

The study ends with an audit-ready, documented report: the protocol, the logger list and calibration traceability, the placement map, the data summary, the hot and cold spots, the stress-test results, the pass or fail against criteria, and the recommended locations for your permanent monitoring sensors. That last recommendation is one of the most valuable outputs of the whole exercise. Your ongoing monitoring probe should sit at the worst-case location the map identified, so that if that point stays in range, everywhere else does too. Storing product in the mapped hot spot should be avoided, or that zone should be down-rated to a warmer storage class.

What equipment a mapping study uses

A study is only as good as its instruments and its record-keeping. A typical warehouse or cold room study relies on a small kit of well-understood tools:

• Calibrated data loggers for temperature, and RH loggers where humidity is in scope, each with a current calibration certificate and a recorded serial number.
• A logging interval set fine enough to catch transient events (commonly one to five minutes) but not so fine that the dataset becomes unwieldy over a multi-day run.
• A placement drawing of the space showing exactly where each numbered logger sits, so the data can be tied back to a physical location.
• A written protocol and a report template prepared before the study, so the acceptance criteria and the reporting format are fixed in advance.

Because the whole exercise stands on the traceability of those loggers, they should be calibrated before deployment. Where the study covers humidity as well as temperature, both channels of each logger need traceable calibration, which is why studies in Singapore's humid climate often pair temperature and humidity work together.

How to prepare your facility for a study

A little preparation makes a study run cleanly and keeps the results representative. Before the mapping team arrives, it is worth doing the following:

• Decide whether to map empty or loaded. An empty space is easier to characterise, but a loaded space reflects real airflow around stock. Many facilities map both, or map loaded because that is the true operating condition.
• Confirm the normal operating pattern. The study should run during representative activity, so schedule it across a normal working week rather than a quiet shutdown period, unless the protocol specifically calls for both.
• Check the cooling system is in its normal state. Do not service or adjust the HVAC or refrigeration immediately before the study, or you will map a temporary condition rather than the real one.
• Brief the operations team. Staff should know the study is running so that door discipline and stock movement reflect normal practice, not an artificially careful week.

Preparing this way means the map reflects how the space genuinely behaves, which is the only version worth qualifying.

Common mistakes that fail an audit

• Writing the acceptance criteria after seeing the data instead of before the study.
• Using loggers with expired or missing calibration certificates, which breaks the traceability of the whole study.
• Mapping only the floor plan and ignoring the vertical gradient, so the ceiling hot spot is never found.
• Running the study for too short a period, or only overnight, so the busy-day peak is missed.
• Placing the permanent monitoring sensor in a comfortable average spot rather than the mapped worst case.

Get a study designed for your space

The steps are consistent, but the details (logger count, placement, duration and criteria) should be scoped to your specific warehouse, cold room or chamber. For the wider regulatory picture, see our guide to temperature mapping for HSA GDP compliance in Singapore. When you are ready to qualify a space, our temperature mapping service in Singapore and temperature and humidity mapping service cover warehouses, cold rooms and chambers, or you can request a scoped protocol and quote.