WhatsApp victor@gzkunling.com

Cleanroom Qualification per ISO 14644-2 and EU GMP Annex 1: Verification, Monitoring, and the Boundary Between Them

Semiconductor R&D and pilot lines do not have a regulatory mandate to run the qualification and monitoring programs to a defined cadence, but the practice is the same. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the CCS is the document that integrates the two. The cadence is typically stricter for semiconductor than for pharmaceutical, because the cost of an excursion is typically higher for a wafer than for a vial.

A working semiconductor qualification and monitoring program has the same four phases (DQ, IQ, OQ, PQ) and the same four parameters (locations, frequency, volume, action limit) as a pharmaceutical program. The action limit is typically set at 50% of the class limit, because the cost of an investigation is small compared to the cost of a missed excursion. The trend is plotted on the same worksheet, and the requalification trigger is the same. The audit is internal, not regulatory, but the standard is the same.

Closing: The Two Programs Are the Compliance Chain

The mental shift that makes the qualification and monitoring programs work is to stop treating them as separate programs and start treating them as the two halves of the same compliance chain. The qualification is the baseline. The monitoring is the test that the baseline continues to hold. The CCS is the document that links the two, and the link is the requalification trigger. The cost of running both programs is small compared to the cost of an excursion that the program would have caught, and the cost of an excursion is small compared to the cost of a recall.

If you are building a qualification and monitoring program from scratch, or reviewing an existing one, we can share a draft DQ/IQ/OQ/PQ protocol template, a monitoring plan worksheet, and a requalification trigger flowchart, typically within two business days. Reach out with your current zone layout, your current class limits, and the date of your most recent requalification.

The boundary is also where the most common audit findings occur. A program that runs the qualification but does not define the requalification trigger is running a one-time test, not a qualification program. A program that runs the monitoring but does not define the action limit is running a data collection, not a monitoring program. A program that has both but does not link the two through the CCS is running two separate programs, not an integrated compliance program. The CCS is the document that integrates the two, and a working CCS is the document the auditor reads first.

The Semiconductor Equivalent: A Stricter Cadence

Semiconductor R&D and pilot lines do not have a regulatory mandate to run the qualification and monitoring programs to a defined cadence, but the practice is the same. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the CCS is the document that integrates the two. The cadence is typically stricter for semiconductor than for pharmaceutical, because the cost of an excursion is typically higher for a wafer than for a vial.

A working semiconductor qualification and monitoring program has the same four phases (DQ, IQ, OQ, PQ) and the same four parameters (locations, frequency, volume, action limit) as a pharmaceutical program. The action limit is typically set at 50% of the class limit, because the cost of an investigation is small compared to the cost of a missed excursion. The trend is plotted on the same worksheet, and the requalification trigger is the same. The audit is internal, not regulatory, but the standard is the same.

Closing: The Two Programs Are the Compliance Chain

The mental shift that makes the qualification and monitoring programs work is to stop treating them as separate programs and start treating them as the two halves of the same compliance chain. The qualification is the baseline. The monitoring is the test that the baseline continues to hold. The CCS is the document that links the two, and the link is the requalification trigger. The cost of running both programs is small compared to the cost of an excursion that the program would have caught, and the cost of an excursion is small compared to the cost of a recall.

If you are building a qualification and monitoring program from scratch, or reviewing an existing one, we can share a draft DQ/IQ/OQ/PQ protocol template, a monitoring plan worksheet, and a requalification trigger flowchart, typically within two business days. Reach out with your current zone layout, your current class limits, and the date of your most recent requalification.

The monitoring program has four parameters that have to be defined explicitly: the sample locations, the sample frequency, the sample volume, and the action limit. A program that does not have all four parameters defined is the second most common audit finding in monitoring programs. The four parameters are linked, and the link is the contamination control strategy.

  1. Sample locations. The sample locations are the points where the air is sampled. The locations are defined by a risk assessment, with more locations at the higher-risk zones (Grade A, the open vial, the open container) and fewer locations at the lower-risk zones (the background, the corridors). The locations are documented on a floor plan, and the floor plan is the input to the monitoring SOP. A program that samples at the same locations as the qualification is not running a monitoring program; it is running a periodic requalification.
  2. Sample frequency. The sample frequency is the rate at which the sample locations are observed. For Grade A, the frequency is continuous; for Grade B, the frequency is typically per shift or per batch. The frequency is defined in the CCS, and the data is reviewed at the documented cadence (typically monthly for the trend plot, quarterly for the QA review).
  3. Sample volume. The sample volume is the volume of air sampled at each location. The volume has to be sufficient to detect a defined concentration at a defined confidence, and the volume is typically 1 cubic meter for ISO 5 and 28 to 100 liters for ISO 7. The volume is defined in the monitoring SOP, and the SOP is the input to the data analysis.
  4. Action limit. The action limit is the concentration that triggers a defined response. The action limit is typically set at the class limit, and the response is a defined sequence: an alert notification, an investigation, a CAPA if the root cause is identified, and a re-monitoring at the same location to confirm the recovery. The action limit is documented in the CCS, and the trend is plotted on the data analysis worksheet.

The Boundary: Where the Two Programs Meet

The boundary between the qualification and the monitoring is the CCS, and the CCS is the single document that defines the relationship. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the action limit is the trigger that calls for a requalification. A requalification is triggered when the trend shows a sustained drift, when an action limit is exceeded repeatedly, or when a change to the design or the operation calls the qualification into question. The CCS defines the trigger, and the trigger is the input to the requalification SOP.

The boundary is also where the most common audit findings occur. A program that runs the qualification but does not define the requalification trigger is running a one-time test, not a qualification program. A program that runs the monitoring but does not define the action limit is running a data collection, not a monitoring program. A program that has both but does not link the two through the CCS is running two separate programs, not an integrated compliance program. The CCS is the document that integrates the two, and a working CCS is the document the auditor reads first.

The Semiconductor Equivalent: A Stricter Cadence

Semiconductor R&D and pilot lines do not have a regulatory mandate to run the qualification and monitoring programs to a defined cadence, but the practice is the same. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the CCS is the document that integrates the two. The cadence is typically stricter for semiconductor than for pharmaceutical, because the cost of an excursion is typically higher for a wafer than for a vial.

A working semiconductor qualification and monitoring program has the same four phases (DQ, IQ, OQ, PQ) and the same four parameters (locations, frequency, volume, action limit) as a pharmaceutical program. The action limit is typically set at 50% of the class limit, because the cost of an investigation is small compared to the cost of a missed excursion. The trend is plotted on the same worksheet, and the requalification trigger is the same. The audit is internal, not regulatory, but the standard is the same.

Closing: The Two Programs Are the Compliance Chain

The mental shift that makes the qualification and monitoring programs work is to stop treating them as separate programs and start treating them as the two halves of the same compliance chain. The qualification is the baseline. The monitoring is the test that the baseline continues to hold. The CCS is the document that links the two, and the link is the requalification trigger. The cost of running both programs is small compared to the cost of an excursion that the program would have caught, and the cost of an excursion is small compared to the cost of a recall.

If you are building a qualification and monitoring program from scratch, or reviewing an existing one, we can share a draft DQ/IQ/OQ/PQ protocol template, a monitoring plan worksheet, and a requalification trigger flowchart, typically within two business days. Reach out with your current zone layout, your current class limits, and the date of your most recent requalification.

ISO 14644-2 defines the monitoring plan that follows the qualification. The standard is explicit: the monitoring plan is not a substitute for the qualification, and the qualification is not a substitute for the monitoring. The two are linked, and the link is the boundary that the standard calls the “monitoring plan” — a defined set of sample locations, a defined sample frequency, a defined sample volume, and a defined action limit that triggers a defined response.

The standard leaves a wide range of choices within the boundary, and a working monitoring plan has to define each of them explicitly. A plan that does not define the action limit, or defines it as “investigate if out of spec,” is the most common audit finding in monitoring programs. The action limit has to be a number, and the response has to be a defined sequence of actions, not a single investigation.

EU GMP Annex 1: The Overlay for Sterile Manufacturing

EU GMP Annex 1 (2022 revision) overlays the ISO 14644 framework with additional requirements for sterile manufacturing. The two grades that matter most for aseptic processing are Grade A (the critical zone, typically the open vial or the open container at the fill line) and Grade B (the background to Grade A, typically the aseptic suite). The grade limits are equivalent to ISO 5 for Grade A at the critical work area and ISO 7 for Grade B, but the operational expectations are stricter: the monitoring is continuous at Grade A, the action limit is more aggressive, and the recovery expectation is documented.

Annex 1 is explicit that the grade is not the same as the ISO class. The grade is an operational expectation, and the monitoring is the evidence that the expectation is being met. The qualification is the baseline, and the monitoring is the test that the baseline continues to hold. The two are linked, and the link is the contamination control strategy (CCS) — a single document that defines the relationship between the design, the qualification, the monitoring, and the response.

The Qualification: DQ, IQ, OQ, PQ

The qualification has four phases: Design Qualification (DQ), Installation Qualification (IQ), Operational Qualification (OQ), and Performance Qualification (PQ). The four phases are run in this order, and each has a defined scope, a defined protocol, and a defined pass criterion. The pass criterion is a number, not a judgment, and the report is built from the protocol, the data, and the pass/fail observation for each test point.

  1. Design Qualification (DQ). The DQ is the document that proves the design meets the user’s requirement. The user requirement is typically expressed as the cleanliness class, the air change rate, the pressure cascade, the temperature, the humidity, and any process-specific requirement (vibration, lighting, ESD). The DQ has to be completed before the construction starts, because the DQ is the input to the construction specification. A DQ that is completed after the construction is a retroactive document, and an audit will treat it as such.
  2. Installation Qualification (IQ). The IQ is the document that proves the installed equipment matches the design specification. The IQ covers the HEPA filters, the FFUs, the air handling unit, the ductwork, the controls, and any process-specific equipment. The IQ is run at installation and is the input to the OQ.
  3. Operational Qualification (OQ). The OQ is the document that proves the installed equipment operates within the design specification under defined conditions. The OQ covers the empty room, with no equipment, no operators, and no process. The OQ is the test that the room meets the cleanliness class in the as-built state, before any operational use.
  4. Performance Qualification (PQ). The PQ is the document that proves the room operates within the design specification under operational conditions, with the actual equipment, the actual number of operators, and the actual process. The PQ is the test that the room meets the cleanliness class in the as-used state, which is the state that the monitoring program will continue to observe.

The four phases are run once at initial qualification, and re-run at the requalification cadence. The requalification cadence is defined in the CCS, and a typical cadence is annual for the OQ and PQ, with the IQ requalified after any change to the installed equipment. A program that does not requalify at the documented cadence is the most common audit finding in qualification programs.

The Monitoring: Locations, Frequency, Volume, and Action Limits

The monitoring program has four parameters that have to be defined explicitly: the sample locations, the sample frequency, the sample volume, and the action limit. A program that does not have all four parameters defined is the second most common audit finding in monitoring programs. The four parameters are linked, and the link is the contamination control strategy.

  1. Sample locations. The sample locations are the points where the air is sampled. The locations are defined by a risk assessment, with more locations at the higher-risk zones (Grade A, the open vial, the open container) and fewer locations at the lower-risk zones (the background, the corridors). The locations are documented on a floor plan, and the floor plan is the input to the monitoring SOP. A program that samples at the same locations as the qualification is not running a monitoring program; it is running a periodic requalification.
  2. Sample frequency. The sample frequency is the rate at which the sample locations are observed. For Grade A, the frequency is continuous; for Grade B, the frequency is typically per shift or per batch. The frequency is defined in the CCS, and the data is reviewed at the documented cadence (typically monthly for the trend plot, quarterly for the QA review).
  3. Sample volume. The sample volume is the volume of air sampled at each location. The volume has to be sufficient to detect a defined concentration at a defined confidence, and the volume is typically 1 cubic meter for ISO 5 and 28 to 100 liters for ISO 7. The volume is defined in the monitoring SOP, and the SOP is the input to the data analysis.
  4. Action limit. The action limit is the concentration that triggers a defined response. The action limit is typically set at the class limit, and the response is a defined sequence: an alert notification, an investigation, a CAPA if the root cause is identified, and a re-monitoring at the same location to confirm the recovery. The action limit is documented in the CCS, and the trend is plotted on the data analysis worksheet.

The Boundary: Where the Two Programs Meet

The boundary between the qualification and the monitoring is the CCS, and the CCS is the single document that defines the relationship. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the action limit is the trigger that calls for a requalification. A requalification is triggered when the trend shows a sustained drift, when an action limit is exceeded repeatedly, or when a change to the design or the operation calls the qualification into question. The CCS defines the trigger, and the trigger is the input to the requalification SOP.

The boundary is also where the most common audit findings occur. A program that runs the qualification but does not define the requalification trigger is running a one-time test, not a qualification program. A program that runs the monitoring but does not define the action limit is running a data collection, not a monitoring program. A program that has both but does not link the two through the CCS is running two separate programs, not an integrated compliance program. The CCS is the document that integrates the two, and a working CCS is the document the auditor reads first.

The Semiconductor Equivalent: A Stricter Cadence

Semiconductor R&D and pilot lines do not have a regulatory mandate to run the qualification and monitoring programs to a defined cadence, but the practice is the same. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the CCS is the document that integrates the two. The cadence is typically stricter for semiconductor than for pharmaceutical, because the cost of an excursion is typically higher for a wafer than for a vial.

A working semiconductor qualification and monitoring program has the same four phases (DQ, IQ, OQ, PQ) and the same four parameters (locations, frequency, volume, action limit) as a pharmaceutical program. The action limit is typically set at 50% of the class limit, because the cost of an investigation is small compared to the cost of a missed excursion. The trend is plotted on the same worksheet, and the requalification trigger is the same. The audit is internal, not regulatory, but the standard is the same.

Closing: The Two Programs Are the Compliance Chain

The mental shift that makes the qualification and monitoring programs work is to stop treating them as separate programs and start treating them as the two halves of the same compliance chain. The qualification is the baseline. The monitoring is the test that the baseline continues to hold. The CCS is the document that links the two, and the link is the requalification trigger. The cost of running both programs is small compared to the cost of an excursion that the program would have caught, and the cost of an excursion is small compared to the cost of a recall.

If you are building a qualification and monitoring program from scratch, or reviewing an existing one, we can share a draft DQ/IQ/OQ/PQ protocol template, a monitoring plan worksheet, and a requalification trigger flowchart, typically within two business days. Reach out with your current zone layout, your current class limits, and the date of your most recent requalification.

A cleanroom’s compliance with its design class is not a single test, it is a chain of tests. The chain starts with the qualification, where the room is verified to meet the design class under defined conditions. The chain continues with the monitoring program, where the room is observed to remain in spec during actual operation. The boundary between the two is the most common source of audit findings, because most teams either run too much monitoring without a defined qualification baseline, or run the qualification once at start-up and assume the monitoring will catch any drift.

This article is a working guide to the boundary between qualification and monitoring per ISO 14644-2 and EU GMP Annex 1 (2022 revision). It assumes the cleanroom is already running with a defined contamination control strategy, a working smoke study protocol, and a documented personnel qualification program. The qualification and the monitoring program are the two halves of the evidence chain that proves the room is in spec.

ISO 14644-1: The Air Cleanliness Class

ISO 14644-1 defines the air cleanliness class by the concentration of particles at or above a considered particle size, measured under defined conditions. The classes run from ISO 1 (the cleanest) to ISO 9 (the dirtiest), with the most common semiconductor and pharmaceutical operations at ISO 5 to ISO 8. The class limit is expressed as particles per cubic meter, and the measurement is a defined protocol that takes a defined volume of air at a defined sample location and counts the particles above a defined size threshold.

The class is the design intent. A cleanroom designed to ISO 7 means the design assumes the room will operate at or below the ISO 7 limit. The class is a number, not a guarantee. The room can be designed to ISO 7 and still fail the ISO 7 limit if the design assumptions do not hold in operation. The qualification is the test that proves the design intent matches the physical room, and the monitoring is the test that proves the design intent continues to hold over time.

ISO 14644-2: The Monitoring Plan

ISO 14644-2 defines the monitoring plan that follows the qualification. The standard is explicit: the monitoring plan is not a substitute for the qualification, and the qualification is not a substitute for the monitoring. The two are linked, and the link is the boundary that the standard calls the “monitoring plan” — a defined set of sample locations, a defined sample frequency, a defined sample volume, and a defined action limit that triggers a defined response.

The standard leaves a wide range of choices within the boundary, and a working monitoring plan has to define each of them explicitly. A plan that does not define the action limit, or defines it as “investigate if out of spec,” is the most common audit finding in monitoring programs. The action limit has to be a number, and the response has to be a defined sequence of actions, not a single investigation.

EU GMP Annex 1: The Overlay for Sterile Manufacturing

EU GMP Annex 1 (2022 revision) overlays the ISO 14644 framework with additional requirements for sterile manufacturing. The two grades that matter most for aseptic processing are Grade A (the critical zone, typically the open vial or the open container at the fill line) and Grade B (the background to Grade A, typically the aseptic suite). The grade limits are equivalent to ISO 5 for Grade A at the critical work area and ISO 7 for Grade B, but the operational expectations are stricter: the monitoring is continuous at Grade A, the action limit is more aggressive, and the recovery expectation is documented.

Annex 1 is explicit that the grade is not the same as the ISO class. The grade is an operational expectation, and the monitoring is the evidence that the expectation is being met. The qualification is the baseline, and the monitoring is the test that the baseline continues to hold. The two are linked, and the link is the contamination control strategy (CCS) — a single document that defines the relationship between the design, the qualification, the monitoring, and the response.

The Qualification: DQ, IQ, OQ, PQ

The qualification has four phases: Design Qualification (DQ), Installation Qualification (IQ), Operational Qualification (OQ), and Performance Qualification (PQ). The four phases are run in this order, and each has a defined scope, a defined protocol, and a defined pass criterion. The pass criterion is a number, not a judgment, and the report is built from the protocol, the data, and the pass/fail observation for each test point.

  1. Design Qualification (DQ). The DQ is the document that proves the design meets the user’s requirement. The user requirement is typically expressed as the cleanliness class, the air change rate, the pressure cascade, the temperature, the humidity, and any process-specific requirement (vibration, lighting, ESD). The DQ has to be completed before the construction starts, because the DQ is the input to the construction specification. A DQ that is completed after the construction is a retroactive document, and an audit will treat it as such.
  2. Installation Qualification (IQ). The IQ is the document that proves the installed equipment matches the design specification. The IQ covers the HEPA filters, the FFUs, the air handling unit, the ductwork, the controls, and any process-specific equipment. The IQ is run at installation and is the input to the OQ.
  3. Operational Qualification (OQ). The OQ is the document that proves the installed equipment operates within the design specification under defined conditions. The OQ covers the empty room, with no equipment, no operators, and no process. The OQ is the test that the room meets the cleanliness class in the as-built state, before any operational use.
  4. Performance Qualification (PQ). The PQ is the document that proves the room operates within the design specification under operational conditions, with the actual equipment, the actual number of operators, and the actual process. The PQ is the test that the room meets the cleanliness class in the as-used state, which is the state that the monitoring program will continue to observe.

The four phases are run once at initial qualification, and re-run at the requalification cadence. The requalification cadence is defined in the CCS, and a typical cadence is annual for the OQ and PQ, with the IQ requalified after any change to the installed equipment. A program that does not requalify at the documented cadence is the most common audit finding in qualification programs.

The Monitoring: Locations, Frequency, Volume, and Action Limits

The monitoring program has four parameters that have to be defined explicitly: the sample locations, the sample frequency, the sample volume, and the action limit. A program that does not have all four parameters defined is the second most common audit finding in monitoring programs. The four parameters are linked, and the link is the contamination control strategy.

  1. Sample locations. The sample locations are the points where the air is sampled. The locations are defined by a risk assessment, with more locations at the higher-risk zones (Grade A, the open vial, the open container) and fewer locations at the lower-risk zones (the background, the corridors). The locations are documented on a floor plan, and the floor plan is the input to the monitoring SOP. A program that samples at the same locations as the qualification is not running a monitoring program; it is running a periodic requalification.
  2. Sample frequency. The sample frequency is the rate at which the sample locations are observed. For Grade A, the frequency is continuous; for Grade B, the frequency is typically per shift or per batch. The frequency is defined in the CCS, and the data is reviewed at the documented cadence (typically monthly for the trend plot, quarterly for the QA review).
  3. Sample volume. The sample volume is the volume of air sampled at each location. The volume has to be sufficient to detect a defined concentration at a defined confidence, and the volume is typically 1 cubic meter for ISO 5 and 28 to 100 liters for ISO 7. The volume is defined in the monitoring SOP, and the SOP is the input to the data analysis.
  4. Action limit. The action limit is the concentration that triggers a defined response. The action limit is typically set at the class limit, and the response is a defined sequence: an alert notification, an investigation, a CAPA if the root cause is identified, and a re-monitoring at the same location to confirm the recovery. The action limit is documented in the CCS, and the trend is plotted on the data analysis worksheet.

The Boundary: Where the Two Programs Meet

The boundary between the qualification and the monitoring is the CCS, and the CCS is the single document that defines the relationship. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the action limit is the trigger that calls for a requalification. A requalification is triggered when the trend shows a sustained drift, when an action limit is exceeded repeatedly, or when a change to the design or the operation calls the qualification into question. The CCS defines the trigger, and the trigger is the input to the requalification SOP.

The boundary is also where the most common audit findings occur. A program that runs the qualification but does not define the requalification trigger is running a one-time test, not a qualification program. A program that runs the monitoring but does not define the action limit is running a data collection, not a monitoring program. A program that has both but does not link the two through the CCS is running two separate programs, not an integrated compliance program. The CCS is the document that integrates the two, and a working CCS is the document the auditor reads first.

The Semiconductor Equivalent: A Stricter Cadence

Semiconductor R&D and pilot lines do not have a regulatory mandate to run the qualification and monitoring programs to a defined cadence, but the practice is the same. The qualification is the baseline, the monitoring is the test that the baseline continues to hold, and the CCS is the document that integrates the two. The cadence is typically stricter for semiconductor than for pharmaceutical, because the cost of an excursion is typically higher for a wafer than for a vial.

A working semiconductor qualification and monitoring program has the same four phases (DQ, IQ, OQ, PQ) and the same four parameters (locations, frequency, volume, action limit) as a pharmaceutical program. The action limit is typically set at 50% of the class limit, because the cost of an investigation is small compared to the cost of a missed excursion. The trend is plotted on the same worksheet, and the requalification trigger is the same. The audit is internal, not regulatory, but the standard is the same.

Closing: The Two Programs Are the Compliance Chain

The mental shift that makes the qualification and monitoring programs work is to stop treating them as separate programs and start treating them as the two halves of the same compliance chain. The qualification is the baseline. The monitoring is the test that the baseline continues to hold. The CCS is the document that links the two, and the link is the requalification trigger. The cost of running both programs is small compared to the cost of an excursion that the program would have caught, and the cost of an excursion is small compared to the cost of a recall.

If you are building a qualification and monitoring program from scratch, or reviewing an existing one, we can share a draft DQ/IQ/OQ/PQ protocol template, a monitoring plan worksheet, and a requalification trigger flowchart, typically within two business days. Reach out with your current zone layout, your current class limits, and the date of your most recent requalification.