A PIC/S and TGA‑rooted practitioner guide for QA teams. Learn the CML, SOPs, interval justification, uncertainty reporting and inspection checks to stay...
GMP requires that all measuring, weighing, recording and control equipment that affects product quality be controlled by a documented, risk based calibration programme with traceable standards and retained records. This means every instrument needs a defined calibration interval, justified by data rather than assumption, and a written procedure for handling results that fall out of tolerance. Inspectors treat calibration records, traceability and out of tolerance handling as priority checks during routine audits.
TL;DR:
- Calibration intervals must be based on actual risk, not just manufacturer recommendations, and often range from a few months to a year depending on instrument criticality.
- Maintaining a comprehensive calibration master list, with traceability and current due dates, is essential to meet GMP and inspector expectations.
- Calibration certificates must report measurement uncertainty and traceability, especially for electronic systems, which should be retained in auditable format.
- Out-of-tolerance or overdue instruments require immediate containment, investigation of impact, and documented corrective actions before reuse.
- Transitioning to digital calibration certificates and trend analysis tools improves data integrity and enables more confident, risk-based interval adjustments.
Calibration obligations under GMP are not a single rule but a consistent expectation repeated across major guidance documents. The PIC/S GMP guide part I, adopted by the TGA, requires that measuring, weighing, recording and control equipment be calibrated at defined intervals and that adequate records be kept. This requirement sits inside Chapter 3 on premises and equipment, and it links directly to the qualification activities covered in Annex 15.
Calibration cannot be treated as a stand-alone task. It needs to be integrated into the broader quality management system, with traceability back to the equipment qualification that happened when the instrument was first installed. Inspectors reviewing a site’s compliance will typically check:
The TGA’s implementation of PIC/S GMP mirrors the structure used across European and WHO guidance, so the underlying expectations are broadly consistent between jurisdictions even where the exact wording differs. What varies locally is often the level of documentation flexibility: some regulators accept risk-based justification for extended intervals, while others expect intervals to track manufacturer recommendations unless a formal deviation is raised. Either way, the core principle holds everywhere: calibration has to be documented, traceable and defensible on demand.
A compliant calibration programme rests on a handful of documented components that work together. Missing even one of these tends to show up as an audit finding, because inspectors know exactly where to look.
The CML is the backbone of the whole system. If an instrument is not on the list, it is effectively invisible to the quality system, and that is exactly the kind of gap inspectors are trained to find. Change control should govern any update to the CML, including new instruments, retired equipment and interval changes.
Pro Tip: Review your Calibration Master List against your equipment inventory every quarter. Instruments quietly added to a production line without a corresponding CML entry are one of the most common gaps found during internal audits.
Interval setting is where a lot of calibration programmes either succeed or quietly drift into non-compliance. Manufacturer recommendations are a reasonable starting point, but GMP guidance expects intervals to reflect actual risk, not just a default figure from a user manual. The inputs that matter most are the instrument’s criticality to product quality, its historical drift pattern, how often it is used, the environment it sits in and how tight the process tolerance is around it.
Typical interval bands vary by instrument type, often ranging from a few months to about a year, depending on the criticality and drift sensitivity of the instrument. For example, critical balances may be calibrated roughly biannually to annually, while pH meters and temperature probes in key steps may have shorter intervals, and equipment like autoclaves and HVAC sensors are usually calibrated annually with ongoing monitoring.
Choosing the right number of calibration points matters as much as the interval itself. A single-point check rarely satisfies GMP expectations because it only confirms accuracy at one value on the scale. Balances with non-linear response across their range often need five-point calibration, while critical temperature probes are commonly verified at three points spanning the operating range, a practice echoed in guidance on calibration causes of drift.
Risk-based calibration management means intervals are not fixed forever. Extending an interval requires documented trend data showing the instrument has stayed within tolerance over several consecutive calibrations, plus sign-off from someone authorised to approve the change. Shortening an interval works the same way in reverse: a pattern of drift or an out of tolerance result is the trigger, not a guess.
Measurement uncertainty must be reported on calibration certificates where applicable, a requirement drawn from GMP-compliant calibration guidance, and it is one of the first things an inspector checks when reviewing whether a result can be trusted.
A calibration procedure that cannot stand on its own, without someone explaining it verbally, will not survive an audit. The same goes for a certificate missing key data fields. Both need a defined structure.
A compliant calibration procedure should cover:
A calibration certificate, in turn, needs a unique identifier, the instrument’s serial number, the standards used and their traceability, as-found and as-left data, the stated measurement uncertainty, and signatures confirming review and release. The TGA’s PIC/S guidance expects these records to be retained in a way that supports batch traceability and, if needed, product recall.
Electronic calibration systems bring their own expectations. Raw data and audit trails need to be preserved in their original electronic form, because a printed summary report alone will not satisfy data integrity requirements during an inspection. Our resource on digital calibration certificates covers how electronic records support ALCOA+ principles, attributable, legible, contemporaneous, original and accurate data that holds up under scrutiny.
When an instrument fails calibration or passes its due date unnoticed, the response needs to be fast and documented. Improvising at this stage is one of the quickest ways to turn a calibration gap into a product quality incident.
Overdue calibrations are not automatically a product failure, but treating them as a non-event is where real risk hides. A thorough impact assessment, even when it concludes no product was affected, is what demonstrates control to an inspector.
Pro Tip: Build an escalation trigger into your calibration software or register so that any instrument nearing its due date flags a supervisor five to seven days ahead, not the day it lapses.
A criticality risk assessment, or CRA, turns interval setting from guesswork into a documented, defensible process. It starts by scoring each instrument against a consistent set of inputs: how directly it affects GMP-critical decisions, how easily a fault would be detected downstream, its historical drift pattern, the manufacturer’s stated performance data and the environment it operates in.
The CRA output feeds directly into the Calibration Master List, setting both the interval and the acceptance tolerance for each instrument. Governance matters here: every CRA needs a documented approval, a defined review cadence, usually annual or triggered by a significant process change, and a link to change control if an interval moves.
Statistical trend analysis is what makes interval changes credible rather than arbitrary. Techniques used in process control, like those described in statistical process control for shop floor operations, apply the same underlying logic: consistent, well-documented trend data is the evidence that supports extending or shortening a calibration interval with confidence.
Most calibration-related audit findings trace back to a handful of recurring gaps rather than exotic failures. Missing as-found and as-left data, certificates without stated traceability or uncertainty, and instruments used past their due date without a documented risk assessment account for the bulk of citations we see referenced in GMP compliance discussions.
Practical steps that close these gaps:
Calibration services can address these gaps by supplying certificates with full traceability statements and documented measurement uncertainty, whether the work happens on-site or in a workshop.
Pro Tip: Keep a simple pre-inspection checklist: pull the CML, cross-check it against physical instrument labels, and confirm every overdue flag has a closed-out assessment attached.
Calibration programmes that still rely on paper logbooks and static intervals are carrying more audit risk than they need to. Digital certificates with preserved audit trails make data integrity defensible by default, and trend analytics let a quality team extend an interval on real evidence instead of a hopeful guess. The programmes that handle inspections calmly are usually the ones backed by an accredited calibration partner who can produce traceable data fast.
— Kaz
We provide NATA-accredited calibration through NATA Calibration and traceable calibration services, delivered either on-site or through our workshop, with every certificate carrying full traceability and documented uncertainty. Requesting a quote is straightforward: send the instrument’s make and model, serial number, the measurement range needed, and target due date.
Calibration needs to be governed by a documented programme that defines intervals, acceptance limits and traceability to a recognised measurement standard. Under PIC/S GMP guidance, every instrument affecting product quality must appear on a controlled register with retained certificates and records.
ISO standards require that measuring equipment used to demonstrate conformity be calibrated or verified against traceable standards at specified intervals, with records kept of the results. These requirements sit alongside GMP requirements rather than replacing them, since GMP adds the risk-based interval justification and OOT investigation steps not detailed in ISO guidance.
There is no single fixed frequency: intervals depend on the instrument’s criticality, historical drift and process tolerance, often ranging from a few months to about a year depending on the instrument type and risk. A documented criticality risk assessment, not the manufacturer’s manual alone, should drive the final interval.
GMP is built around a handful of consistent principles rather than a formal numbered list of five, but they commonly include qualified personnel, validated processes and equipment, documented procedures, traceable records and robust quality control checks. Calibration sits across several of these at once, since it depends on trained staff, validated methods, documented SOPs and retained records.
The instrument needs to be removed from service immediately and every batch or result it may have influenced needs retrospective review. The investigation determines whether product should be quarantined, retested or escalated, and a documented corrective action closes out the root cause before the instrument returns to use.
