
Quick answer
- IQ proves it was installed as specified. OQ proves it works as specified when empty. PQ proves it holds conditions when loaded and in daily use.
- Each stage must be completed, and its deviations closed, before the next one starts.
- WHO expects at least 24 hours of empty-chamber data in OQ, and door-opening recovery within a defined time (for example 30 minutes for a 2–8 °C store) in PQ.
- Acceptance criteria come from your URS and are fixed before testing begins.
- Requalify on a justified schedule and after any significant change.
- Download: IQ/OQ/PQ protocol templates for chambers, cold rooms and warehouses
On this page
What IQ, OQ and PQ mean in pharma
IQ, OQ and PQ are the three documented stages that prove equipment or a facility is fit for its intended GMP use. The definitions below follow WHO's technical supplement on qualifying temperature-controlled storage.
| Stage | Full form | Question it answers | Condition during testing |
|---|---|---|---|
| IQ | Installation Qualification | Was it built and installed to the design specification? | Installed, not yet operated for GMP use |
| OQ | Operational Qualification | Does it operate to specification under controlled conditions? | Empty |
| PQ | Performance Qualification | Does it perform consistently in routine use? | Loaded, normal operations |
Qualification vs validation. Qualification applies to premises, utilities and equipment. Validation applies to processes, methods and computerised systems. Qualified equipment is a precondition for a validated process; qualification alone does not validate the process.
Where the requirement comes from
| Regulation or guideline | What it says on qualification |
|---|---|
| Revised Schedule M (G.S.R. 922(E), 28 Dec 2023) | Section 5 extends qualification to premises, utilities, equipment and process (DQ, IQ, OQ, PQ). Section 20 requires IQ and OQ of computerised system hardware and software. In force for large manufacturers from 28 June 2024. Schedule M guide. |
| EU GMP Annex 15 (in force 1 Oct 2015) | Lifecycle from URS through DQ, FAT/SAT, IQ, OQ, PQ. IQ and OQ may be combined. Adds a requalification chapter. A revision concept paper was published in 2026. |
| WHO TRS 961 Annex 9, Supplement 7 (TRS 992 Annex 5, 2015) | Every new temperature-controlled store must be qualified before release for routine storage of temperature-sensitive products. |
| US FDA 21 CFR 211 | Equipment must be of appropriate design and suitably located; qualification is the evidence inspectors review. |
Where URS and DQ fit: the full qualification sequence
IQ, OQ and PQ test against requirements written earlier. If those requirements are missing or vague, there is nothing to test against. Annex 15 sets the order below; WHO says each stage must be completed before the next begins.
- 1URS (User Requirements Specification): what the equipment must do, written by the user. For a stability chamber: setpoint 30 °C / 65% RH, tolerance ±2 °C / ±5% RH, alarms, data retention, Part 11 needs.
- 2DQ (Design Qualification): documented check that the proposed design meets the URS and GMP. Done before purchase order or construction.
- 3FAT (Factory Acceptance Test): optional; tests at the supplier before shipping. Annex 15 recommends it for novel or complex equipment.
- 4SAT (Site Acceptance Test): optional; checks after delivery. Justified FAT/SAT results can be reused in IQ/OQ if transport and installation did not affect them.
- 5IQ, then OQ, then PQ.
- 6Release for routine use, then periodic review and requalification.
The V-model pairing (each test stage verifies one specification level):
| Specification | Verified by | Example for a 2–8 °C cold room |
|---|---|---|
| URS | PQ | Holds 2–8 °C everywhere product is stored, loaded, in daily use |
| Functional specification | OQ | Controller holds setpoint empty; alarms fire at limits; recovers after power loss |
| Design specification | IQ | Refrigeration unit model, panel build, sensor count and wiring match drawings |
The link between them is the traceability matrix: every URS line maps to at least one test with a pre-set acceptance criterion. Auditors check it to confirm nothing in the URS went untested.
Difference between IQ, OQ and PQ
The three stages differ in what they test against, the condition of the equipment, and who usually signs. A 2–8 °C cold room is used as the example throughout.
| IQ | OQ | PQ | |
|---|---|---|---|
| Tests against | Design specification and drawings | Functional specification | URS (intended use) |
| Equipment state | Installed, not operating for GMP use | Running, empty | Running, loaded as in normal use |
| Typical checks | Model and serial numbers, utilities, wiring, calibration certificates, manuals, spare parts | Controller and setpoints, high/low alarms, empty mapping, power failure, SOPs and training | Loaded mapping, door-opening recovery, seasonal performance |
| Typical duration | 1–3 days of inspection | At least 24 h of empty data, plus alarm and power tests | Several days of loaded data; often repeated in summer and winter |
| Cold room pass example | Refrigeration unit matches specified model and capacity | All loggers stay within 2–8 °C for 24 h empty | All loggers return to 2–8 °C within 30 min of door closing |
| Typical executors | Engineering, vendor, validation | Validation, engineering | Validation, QA, operations |
| Output | IQ report; deviations closed before OQ | OQ report; deviations closed before PQ | PQ report; equipment released for routine use |
IQ and OQ can be combined into a single IOQ protocol where justified (Annex 15). WHO notes that OQ and PQ for storage areas are often run as one sequence after IQ.
Installation Qualification (IQ): what to verify
IQ confirms that what was installed matches what was specified, item by item. Every check is recorded in three columns: specified, as found, and pass or fail. Fill in the "specified" column before the inspection starts, so the inspector is checking against the design and not against what happens to be there.
Critical components to list first
- Refrigerators, freezers, small chambers: thermostat or controller, its control sensor, and the independent monitoring device and its sensor.
- Cold rooms, freezer rooms, walk-in chambers: controller, sensors, cooling unit, condenser, evaporator, and the room enclosure itself (panels, doors, seals). Where components are duplicated, check each one.
IQ checklist
| Area | What to check | Example of a fail |
|---|---|---|
| Identity | Manufacturer, model, serial number, internal asset ID | Unit installed is a lower-capacity model than the one specified |
| Electrical supply | Voltage (e.g. 415 V, 3-phase, 50 Hz), circuit amperage, breaker label, earthing, isolating switch, emergency power | Circuit rated 20 A where 30 A was specified |
| Environment | Cleanliness, fumes, vibration, ambient temperature and RH at handover | Refrigeration unit vibrating on its mount |
| Calibration | Current certificate for every critical controller and sensor, traceable to national standards; tag on each | Sensor certificate expired or not traceable |
| Software and firmware | Versions recorded, configuration backed up | Firmware version differs from the one in the design documents |
| Documents | Drawings, manuals, wiring diagrams, material certificates | As-built drawings missing |
| Spare parts | List supplied parts and check them against specification | Spare sensor is a different type from the installed one |
| Auxiliary equipment | Generator, UPS, building alarm, network; record the interfaces | Generator changeover not documented |
| Maintenance readiness | Asset in equipment inventory, labelled, maintenance logbook in place | No preventive maintenance entry created |
Two practices auditors look for
- A calibration check straight after installation, before OQ starts, so transport or installation has not shifted any sensor.
- Every deviation on a separate deviation form, cross-referenced from the IQ sheet, with a corrective action and an owner. WHO's rule: do not begin OQ until relevant IQ deviations are corrected.
All sheets are signed by the inspector and checked by a reviewer who did not perform the inspection.
Operational Qualification (OQ): the tests and what passes
OQ proves the equipment works as specified while it is empty. WHO's rule: do not start OQ until the IQ deviations are corrected. Run the tests below in this order, using the same logger layout throughout so results can be compared.
| # | OQ test | How it is done | Typical acceptance criterion |
|---|---|---|---|
| 1 | Calibration status | Check every critical controller and sensor | Current certificate, traceable to national standards, tagged with ID, cal date and due date, entered in the calibration programme |
| 2 | SOPs and training | Check operation, alarm response and maintenance SOPs exist | SOPs approved and matched to the installed equipment; training record for each SOP |
| 3 | Control panel and setpoints | Exercise controls, displays and setpoint changes | All functions work per manufacturer specification |
| 4 | Alarm tests | Trigger at least one high and one low temperature alarm, plus door-open and power alarms if fitted | Alarm activates on every configured channel (sounder, SMS, email, call); settings recorded |
| 5 | Empty-chamber distribution study | Loggers at defined points, empty, at setpoint | All points within limits for the whole run; WHO minimum is 24 hours |
| 6 | Door-opening recovery (empty) | Open the door for the longest routine duration | Returns within limits inside the time set in the URS |
| 7 | Power failure | Cut power until out of range, or for a fixed period (e.g. 2 hours), then restore | Holdover time and recovery time recorded; pass/fail only if the URS sets a limit |
Notes inspectors raise
- Alarm tests must reach a person. An alarm that sounds on the panel but never sends the SMS fails, because no one acts on it at 2 a.m. Test every escalation contact listed in the alarm SOP.
- Power failure results are design data. WHO sets no fixed pass mark for large stores. The holdover time tells you how long you have before product is at risk and how big your generator or UPS must be. Write it into the excursion SOP.
- One empty study is enough for OQ. WHO recommends seasonal studies for full characterisation, but only one empty run is required at OQ stage.
- Logger specification matters. WHO lists the minimum for qualification loggers: a range covering the extremes (for example −30 °C to +60 °C), a programmable interval of 1–15 minutes, a 3-point traceable calibration certificate with error no worse than ±0.5 °C, and Part 11-capable software.
How many loggers, where to place them and how to analyse the data is covered in our temperature mapping guide. This chapter only sets out what OQ must prove.
Performance Qualification (PQ): proving it in routine use
PQ proves the equipment holds its conditions loaded, with doors opening and people working around it. WHO's rule: start PQ only after OQ deviations are corrected, and release the equipment only after PQ deviations are corrected.
What PQ covers
| PQ test | How it is done | Typical acceptance criterion |
|---|---|---|
| Calibration re-check | Confirm controllers and sensors are still in calibration | All current; no drift beyond tolerance since OQ |
| Loaded distribution study | Same logger layout as OQ, with product or a representative load at normal or maximum loading | Every point within the specified range for the full run |
| Door-opening recovery (loaded) | Open the door(s) for the longest duration seen in routine work; with several doors, use the real sequence | All loggers back within range in the URS time; WHO's example is 2–8 °C within 30 minutes of closing |
| Each operating condition | For multi-setpoint stability chambers, repeat at every condition in use (e.g. 25 °C/60% RH, 30 °C/65% RH, 30 °C/75% RH, 40 °C/75% RH) | Within ±2 °C and ±5% RH of each setpoint |
| Seasonal performance (areas affected by ambient) | Repeat the loaded study in the hottest and coldest periods | Within range in both. In most of India that means April–June and December–January; check RH in the monsoon too |
Three PQ decisions to fix in the protocol, not during the run
- 1Load pattern. State the load as a percentage of capacity and how it is arranged. Inspectors ask whether PQ was run at worst-case (maximum) load.
- 2Door-opening duration. Base it on observed behaviour, such as your longest receiving or pick event, and record where that number came from.
- 3Study duration. State the hours or days and the logging interval before starting. A study cannot be cut short because the data look good.
After PQ, the qualified state has to be maintained. Monitoring sensors drift between qualifications; see our blog on why monitoring drifts after validation.
Writing qualification acceptance criteria (with typical values)
Acceptance criteria must be numeric, written into the protocol and approved before testing starts. Changing them after you see the data invalidates the qualification and is a common inspection finding. Derive each limit from three inputs: the product's labelled storage condition, the URS, and the equipment's specified capability.
Typical starting values by equipment type (confirm against your own URS and product labels)
| Equipment | Typical range | Stability or uniformity criterion | Recovery criterion |
|---|---|---|---|
| Stability chamber, long-term (India, climatic zone IVb) | 30 °C / 75% RH | Every point within ±2 °C and ±5% RH (ICH tolerance) | Back within tolerance in the URS time after a door opening |
| Stability chamber, accelerated | 40 °C / 75% RH | Every point within ±2 °C and ±5% RH | As above |
| Stability chamber, intermediate or zone II long-term | 30 °C / 65% RH or 25 °C / 60% RH | Every point within ±2 °C and ±5% RH | As above |
| Pharmaceutical refrigerator, cold room | 2–8 °C (ICH refrigerated long-term: 5 °C ± 3 °C) | Every point within 2–8 °C; no reading at or below 0 °C | WHO example: within range 30 minutes after door closes |
| Freezer | −25 °C to −10 °C (USP freezer range) | Every point within range | Defined in URS |
| Ultra-low freezer | −80 °C or −70 °C setpoint, per product | Every point at or below the product limit | Defined in URS; record holdover time |
| Controlled-ambient warehouse | Per label, e.g. 15–25 °C or "store below 30 °C" | Every point within the label range, in hot and cold seasons | Dock-door recovery defined in URS |
Rules for writing a criterion that holds up
- Write a number, not a word. "Within 2.0–8.0 °C at all 15 points for 24 hours" is testable. "Temperature uniform" is not.
- Use the same precision as your loggers. With ±0.5 °C loggers, a criterion of ±0.2 °C cannot be proved.
- Say what happens at the edge. State whether 8.0 °C is a pass, and whether a single reading out of range fails the study or triggers a deviation.
- Don't copy criteria between rooms. A protocol reused from another facility without re-deriving limits from its own URS is a common and serious failure.
IQ OQ PQ protocol and report: the documents auditors check
An inspector reviewing a qualification asks for one file per equipment item. If any document below is missing, unsigned or out of sequence, expect a finding.
Every IQ, OQ and PQ protocol needs seven sections (WHO structure):
- 1Approval page and change history (author, reviewer, approver, version log)
- 2Acronyms and glossary
- 3Description and rationale: what is being qualified and why
- 4Scope and objectives
- 5Key parameters: setpoints, ranges, alarm limits
- 6Test procedures, each with title, target, step-by-step method, acceptance criterion and a blank data sheet
- 7Report template
The qualification file
| Document | When it is created | What auditors check |
|---|---|---|
| URS | Before purchase | Specific, testable requirements; approved by QA |
| DQ report | Before purchase or build | Each URS line traced to the design |
| Validation master plan or qualification plan | Before IQ | Scope, sequence, responsibilities |
| IQ, OQ, PQ protocols | Approved before each stage | Approval date earlier than execution date; criteria pre-set |
| Executed data sheets | During testing | Specified, as found, pass/fail; signatures and dates; raw logger files attached |
| Deviation reports | Whenever a result fails | Impact assessment, corrective action, closure before the next stage |
| Calibration certificates | Before OQ | Every logger and every control or monitoring sensor, traceable, in date |
| Traceability matrix | Updated through the project | Every URS line covered by a test |
| IQ, OQ, PQ reports | After each stage | Results summary, deviations, conclusion; reviewed by someone who did not execute |
| Release for routine use | After PQ | QA sign-off with date |
Requalification: triggers and how often
No regulation sets a single requalification interval. Annex 15 requires equipment to be evaluated at an appropriate frequency to confirm it stays in a state of control. Where a fixed period is used, it must be justified with defined criteria. WHO adds that qualification is repeated whenever the installation or the way it is used changes significantly.
Event-based triggers (requalify the affected stage)
| Change | Usual scope |
|---|---|
| Relocation of the equipment | Full IQ, OQ and PQ |
| Replacement of controller, compressor, evaporator or control sensor | IQ of the new part, then OQ; PQ if performance could be affected |
| Firmware or software change in the controller or monitoring system | Change control impact assessment; OQ of affected functions |
| New setpoint or operating range | OQ and PQ at the new condition |
| Change in load pattern, racking or layout | PQ |
| Recurring excursions or an alarm trend | Investigation, then targeted OQ or PQ |
| Major repair, panel or door replacement in a cold room | IQ of the repaired element, then OQ |
Time-based review (common industry practice, not a legal number)
- Stability chambers: OQ and PQ are usually repeated every year.
- Cold rooms and warehouses: often every 1–3 years, set by risk and justified in the qualification plan.
- Continuously monitored, stable equipment: a documented periodic review of alarm history, deviations, calibration and maintenance can justify a longer interval than a fixed calendar repeat.
For a worked example of setting the interval in a regulated lab, see How Often Should You Requalify a CRO Freezer?
Qualifying the monitoring system and the loggers you test with
Two things are often missed: the continuous monitoring system on the equipment needs its own qualification, and the loggers used in OQ and PQ need to be fit for the job.
The monitoring system is a computerised system. Revised Schedule M (Section 20) requires IQ and OQ of computerised system hardware and software. For a temperature and humidity monitoring system that typically means:
| Stage | What to verify |
|---|---|
| IQ | Sensor and gateway models, firmware versions, installed locations matching the drawing, calibration certificate for each sensor, network and power connections, user accounts created |
| OQ | High and low alarms reach every contact; time sync; audit trail records setpoint and limit changes; role-based access; data backup and restore; report output matches raw data |
| PQ | Runs in routine use for a defined period with real users responding to real alerts |
The electronic records rules (audit trails, e-signatures, access control) are covered on our 21 CFR Part 11 temperature monitoring page.
Loggers used for qualification studies. Check them against the WHO minimum before the study, not after:
- Range covering the expected extremes, for example −30 °C to +60 °C
- Programmable logging interval of 1–15 minutes, with memory for the full study
- 3-point calibration certificate traceable to national standards, error no worse than ±0.5 °C
- Data download to software that meets 21 CFR Part 11
Mindlabs Nova is built for qualification and validation studies: ±0.5 °C accuracy, temperature and humidity, a configurable logging interval, 20 months of recording and auto-generated PDF/CSV reports. It ships with a calibration certificate; with the PT-100 probe it logs down to −80 °C for ultra-low freezers. For continuous monitoring after release, Anchor sends real-time alerts.

Request a free Nova trial for your next qualification.
Eight qualification mistakes that become inspection findings
Most qualification findings are not about the equipment. They are about missing proof. The FDA has cited a stability chamber qualification in a warning letter because the studies did not always include specifications, acceptance criteria or raw data. In FDA observation counts presented in 2019, "calibration/inspection/checking not done" (21 CFR 211.68(a)) was cited 60 times and "equipment design, size and location" (211.63) 53 times.
| # | Mistake | What the inspector sees | Fix |
|---|---|---|---|
| 1 | Protocol approved after testing began | Approval date later than first data sheet | Approve every protocol before execution |
| 2 | Acceptance criteria missing, vague or changed later | "Uniform" instead of a number; revised limits after results | Numeric criteria in the approved protocol; changes only through change control before the test |
| 3 | Generic protocol copied from another site | Limits that do not match this room's URS | Re-derive limits from this equipment's URS |
| 4 | OQ started with IQ deviations open | Deviation closure dated after OQ start | Close or formally risk-assess deviations before the next stage |
| 5 | Expired or untraceable calibration | Logger certificate out of date on the study date | Check every certificate before the study starts and attach it to the report |
| 6 | Alarms tested only at the panel | No proof SMS or email reached anyone | Test every notification channel and contact |
| 7 | PQ run only empty, or at partial load | No loaded or worst-case data | Define the load pattern and run PQ loaded |
| 8 | No requalification after a change | Compressor replaced, no change control or retest | Link maintenance work orders to change control and requalification |