Every regulatory statement on this site rests on a text that is named, dated and checkable. Here they all are, with their reference and their date.
This is not a bibliography. These are the texts we open when a project touches a dossier: the ones that decide what a procedure has to demonstrate, what enters a control strategy, and what gets reported after a change. Each is cited somewhere on this site, and you can walk back from the statement to the text.
One distinction we hold to: the texts a technical statement on this site rests on, we have read. Those listed here as the frame of a sector are named for what they cover, and nothing further is attributed to them.
One reading key, because it settles a frequent confusion. The US pharmacopoeia numbers its chapters by their standing: below 1000 a chapter carries requirements; above, it informs. Near infrared exists on both levels — <856> binds, <1856> explains — and Raman and mid-infrared follow the same rule.
ICH — the international guidelines
ICH texts are public: they download from the ICH quality guidelines page.
| Text | What it governs |
|---|---|
| ICH Q2(R2) — Validation of Analytical Procedures EMA/CHMP/ICH/82072/2006, step 5, revision 1 | What a validated analytical procedure has to demonstrate. Its section 2.5 covers multivariate procedures: calibration, internal testing, then validation on an independent set. Its section 3.1.2.1 covers identification tests. |
| ICH Q14 — Analytical Procedure Development Guidance for Industry, FDA, CDER and CBER, March 2024 | Procedure development, the analytical target profile, the control strategy. Its section on the life cycle of a multivariate model names the diagnostics expected in routine use: Hotelling’s T², Mahalanobis distance, examination of residuals. |
| ICH Q5E — Comparability of Biotechnological/Biological Products Subject to Changes in Their Manufacturing Process step 4, 18 November 2004 | Comparability of a biological product either side of a manufacturing process change. It asks for “highly similar” attributes rather than identical ones, it asks that in-process controls be confirmed, modified or created after a change, and it provides that nonclinical and clinical studies are not warranted where comparability is established through analytical data alone. |
| ICH Q7 — Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients step 4, 10 November 2000; adopted as EudraLex Volume 4, Part II | Good manufacturing practice for active substances. This is the text that frames an in-process control on a synthesis, crystallisation or drying step of a drug substance. |
| ICH Q8(R2) — Pharmaceutical Development step 4, August 2009 | Quality by Design, the design space, and the definition of real time release testing the other texts refer back to. |
| ICH Q9(R1) — Quality Risk Management EMA/CHMP/ICH/24235/2006, step 5, in force in the EU since 26 July 2023; the revision was adopted at step 4 on 18 January 2023 and supersedes the ICH Q9 of 9 November 2005 | Quality risk management, which sets the level of evidence a measurement owes according to what it decides. |
| ICH Q10 — Pharmaceutical Quality System step 4, June 2008 | The quality system across the whole product life cycle. It is what hosts the change management of a model. |
| ICH Q11 — Development and Manufacture of Drug Substances step 4, 1 May 2012 | Development and manufacture of drug substances, chemical and biotechnological. |
| ICH Q12 — Technical and Regulatory Considerations for Pharmaceutical Product Lifecycle Management EMA/CHMP/ICH/804273/2017, step 5, 4 March 2020 | Post-approval change management: established conditions, comparability protocols. The frame a recalibration is reported within. |
| ICH Q13 — Continuous Manufacturing of Drug Substances and Drug Products final version adopted 16 November 2022 | Continuous manufacturing, where process measurement takes the place of the end-of-batch sample. |
FDA and the US Code of Federal Regulations
| Text | What it governs |
|---|---|
| Formal Meetings Between the FDA and Sponsors or Applicants of PDUFA Products Guidance for Industry, FDA, final version of August 2026, published 13 August 2026 | The framework for formal meetings with the FDA, and their categories. Introducing an analytical technology onto an already registered process belongs to no development milestone: it therefore goes through a Type C meeting, the category meant for subjects that do not fit the narrower formats. What we supply at filing. |
| PAT — A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance Guidance for Industry, FDA, CDER, CVM and ORA, September 2004 | The founding text of the discipline. It defines PAT as a system for designing, analysing and controlling manufacturing through measurements taken during processing, on the critical attributes of incoming and in-process materials. It states that “analytical” there covers chemical, physical, microbiological, mathematical and risk analysis. It sets out a three-part definition of a well understood process: critical sources of variability identified and explained, variability managed by the process, and quality attributes reliably predicted over the design space. And it states that transferring a laboratory method in line is not automatically PAT. The 2021 NIR guidance refers back to it explicitly. Nonbinding recommendations. |
| Development and Submission of Near Infrared Analytical Procedures Guidance for Industry, FDA, CDER, August 2021 | The text closest to what we do: building the calibration set, sample presentation, external validation both qualitative and quantitative, and the reporting route for each change. It states that the calibration set spans a range wider than the specification, so that the model can describe non-conforming material. |
| Process Validation: General Principles and Practices Guidance for Industry, FDA, CDER, CBER and CVM, January 2011, revision 1 | Its section II.B sets out the four expectations on variation — understand it, detect it, understand its impact, control it commensurately with the risk. The second has no answer other than process measurement. |
| 21 CFR part 11 — Electronic Records; Electronic Signatures | Electronic records and signatures. What a “21 CFR Part 11 ready” claim covers, and what it leaves with the operator. |
| 21 CFR parts 210 and 211 | US current good manufacturing practice for finished pharmaceuticals, which the NIR guidance points to for any implementation on a commercial site. |
| 21 CFR 1040.10 and 1040.11 | Performance standards for laser products. A safety reference rather than a data integrity one — worth telling apart when reading a data sheet. |
EMA and the European Union
| Text | What it governs |
|---|---|
| Guideline on Real Time Release Testing EMA/CHMP/QWP/811210/2009-Rev1, adopted 15 March 2012, in force 1 October 2012 | The conditions for real time release testing, its ten prerequisites, and one rule that decides the level of ambition: once approved it is used routinely, and a contingency plan is described in the dossier. |
| EudraLex volume 4, annex 11 — Computerised systems | Computerised systems in a GMP environment. The European counterpart of 21 CFR part 11, and it does not say exactly the same thing. |
| European good manufacturing practice EudraLex volume 4 | The general frame an in-line measurement is installed within on a European site. |
United States Pharmacopeia (USP)
USP–NF is read on uspnf.com, by subscription. There is no public page per chapter: that is why the rows below carry no link of their own.
| Chapter | What it governs |
|---|---|
| <856> Near-Infrared Spectroscopy official since 1 November 2020 — carries requirements | Instrument qualification, procedure, validation and verification for NIR. It names the error of the laboratory method itself and asks for suitable agreement between it and the prediction error. It also states that measurement can be made through the glass or plastic films used for containment. |
| <1856> Near-Infrared Spectroscopy — Theory and Practice informational | Theory and practice. Chapter <856> itself presents it as a helpful but not mandatory resource. |
| <858> Raman Spectroscopy and <1858> <858> official since 1 November 2020 | The same two-level architecture for Raman. |
| <854> Mid-Infrared Spectroscopy and <1854> <854> official since 1 May 2020 | The same architecture for mid-infrared. |
| <1039> Chemometrics official since 1 May 2020 | Sample selection, preprocessing, algorithm choice, performance metrics and the life cycle of a model. It is what <856> points to for methodologies. |
| <922> Water Activity official since May 2021 — carries requirements | The measurement of water activity: sensor types, qualification, calibration, running the test and reporting the result. It is the official method that was missing: until then only the informational <1112> covered the subject. What near infrared sees of it. |
| <905> Uniformity of Dosage Units harmonised with the European and Japanese pharmacopoeias | The uniformity of dosage units test: ten units at the first stage with a calculated acceptance value, twenty more at the second, from a sample of at least thirty. |
| <1225> Validation of Compendial Procedures official since 1 August 2017 | The four categories of procedure and what each one asks for. Category IV, the one for identification tests, asks for specificity and asks for neither accuracy, nor precision, nor linearity, nor range. |
| <1226> Verification of Compendial Procedures official since 1 December 2019 | Verifying a compendial procedure on a given product under given conditions. |
| <1220> Analytical Procedure Life Cycle official since 1 May 2022 | The life cycle of an analytical procedure, from the target profile to routine monitoring. |
| <1058> Analytical Instrument Qualification official since 1 August 2017 | Qualifying an analytical instrument, which is distinct from validating the procedure that runs on it. |
| <197> Spectroscopic Identification Tests official since 1 September 2021 | Spectroscopic identification tests. |
| <1010> Analytical Data — Interpretation and Treatment | Statistical treatment of analytical data: method comparison, outliers, intervals. |
| <1112> Application of Water Activity Determination to Nonsterile Pharmaceutical Products informational | Water activity and its use on nonsterile products: what it decides about microbiological stability, and why it is not the same thing as water content. What near infrared sees of it, and under what condition. |
| <1097> Bulk Powder Sampling Procedures informational | Sampling of bulk powders: theory, sampling plan, tools. It names segregation error and sampling method error — the compendial chapter closest to the theory of sampling, and the one most often missing from discussions of whether an increment is representative. |
| <1099> Limit on Number of Large Deviations When Assessing Content Uniformity in Large Samples informational, and not official — official date postponed since 1 March 2019 | Handling large sample counts in content uniformity. That is the situation an in-line measurement creates: not ten units but thousands — a regime <905> was not written to cover. This chapter cannot be invoked as a reference text: its official date, set for 1 March 2019, was postponed by a Revision Bulletin of 22 February 2019; a first revision (PF 47(4)) was withdrawn, and a new proposal (PF 51(3)) still sits among the deferred texts of the USP-NF 2026, Issue 3 cycle. It has not been cancelled — and the fact that USP has been working on it since 2017 without settling it says the essential thing: the large-sample regime is an open question, not an available framework. Status verified on 19 September 2026. |
| <616> Bulk Density of Powders carries requirements where a monograph refers to it — retitled: official 1 May 2024, previously Bulk Density and Tapped Density of Powders | Untapped and tapped density, and the compressibility measures derived from them — Carr index and Hausner ratio. It is the standardised method behind the bulk density that enters any calculation of probed mass. |
| <1174> Powder Flow and <1063> Shear Cell Methodology for Powder Flow Testing informational | Powder flow: angle of repose, flow through an orifice, Carr index, Hausner ratio, shear cells. Both chapters state that one method is not enough — several standardised methods should be used to characterise what governs flow. |
| <430> Particle Size Analysis by Dynamic Light Scattering and <1430.3> <430> official since 1 May 2024, harmonised within the Pharmacopoeial Discussion Group — carries requirements; <1430.3> informational | Submicron particle sizing by dynamic light scattering. It transposes ISO 22412 and settles three useful things. DLS cannot be calibrated: its sizes are not relative values drawn from standards but values calculated from first principles. The test report must carry the laser wavelength, the observation angle and the data analysis program itself, since the result depends on it — which is what settles the question of two instruments that do not return the same figure. And concentration must be adjusted to rule out multiple scattering, which presumes a diluted sample: what that implies for an in-line measurement without dilution. |
| <1776> Image Analysis of Pharmaceutical Systems and <776> Optical Microscopy <1776> informational; <776> carries requirements, harmonised with Ph. Eur. 2.9.37 | Image analysis and optical microscopy as measurement methods. <1776> defines image analysis as “primarily a size and shape analysis” and names the descriptors: equivalent circular diameter, Feret’s diameter, elongation, circularity, convexity. <776> covers size and shape, and sets a limit test by counting. Their scope is the powder, the excipient, the suspended particle — not the object moving along a line: what that changes for counting in production. |
| <788> Particulate Matter in Injections, <1788> and its subchapters <788> carries requirements; the <1788> series is informational. European counterparts: Ph. Eur. 2.9.19 (sub-visible particles) and 2.9.20 (visible particles) | Particulate contamination of injections. <788> sets two methods, light obscuration and the Microscopic Particle Count Test. The <1788> series describes the three instrumental routes, including <1788.3> Flow Imaging, where the count comes with morphological parameters — equivalent circular diameter, Feret’s diameter, circularity. It is the chapter closest to counting by imaging, and it addresses the syringe, not the line. |
| <786> Particle Size Distribution Estimation by Analytical Sieving and <811> Powder Fineness carry requirements where a monograph refers to them | Particle size by sieving, and the descriptive classification of powder fineness from its median dimension. The compendial point of comparison for any in-line size measurement. |
European Pharmacopoeia (Ph. Eur.)
The European Pharmacopoeia is published by the EDQM and read online on pheur.edqm.eu, by subscription.
| Chapter | What it governs |
|---|---|
| 5.25 Process analytical technology Ph. Eur. 11.0, revision 04/2021 | The only pharmacopoeial chapter devoted to PAT. It defines the four interfacing modes — off-line, at-line, on-line, in-line — and sets what separates them: what is removed from the stream, where the equipment sits, and the delay before the result can be used. It asks that the causal relationship be demonstrated between the PAT measurement, the model and the quality attribute where a trajectory or signature model is used for control. It states that comparison with a reference test procedure would normally be needed for an on-line or in-line measurement: conventional validation procedures assume homogeneous and authentic samples, a condition rarely met on a process that changes during the measurement. And it notes that the qualification criteria of Ph. Eur. techniques were designed for off-line analytical systems, and are not always relevant or practical in a PAT setting. |
| 5.24 Chemical imaging Ph. Eur. 11.0, 01/2020 | The general chapter for chemical imaging, focused on systems based on vibrational spectroscopy — mid-infrared, near-infrared, Raman — and applicable to other techniques that supply images. It provides that where a system is intended for investigative purposes, the performance requirements of 2.2.24, 2.2.40 and 2.2.48 need not be applied: criteria are established case by case, through a risk-based approach. It also names OCT and terahertz among the three-dimensional imaging methods. |
| 5.21 Chemometric methods applied to analytical data Ph. Eur. 11.1, 04/2023 — published for information | The pharmacopoeial introduction to chemometrics: good data analysis practice, established methods, critical aspects and limitations, with a glossary. It is published for information and therefore carries no enforceable requirement — the same standing as USP <1039>, of which it is the European counterpart. |
The chapters 5.25 draws on, named here as it describes them and no further: 2.2.24 infrared, 2.2.25 ultraviolet and visible, 2.2.37 X-ray fluorescence, 2.2.40 near infrared, 2.2.48 Raman, 2.9.47 uniformity of dosage units using large sample sizes, 5.24 chemical imaging, 5.28 multivariate statistical process control.
Food and animal feed
Here the regulation covers the product and the quality system. It requires critical points to be monitored and the means of monitoring to be justified, without prescribing a method: that is what leaves room for an in-line measurement, and what obliges you to show it is fit for the job.
| Text | What it governs |
|---|---|
| ISO 12099 — Animal feeding stuffs, cereals and milled cereal products: guidelines for the application of near infrared spectrometry 2017 edition, adopted as European standard EN ISO 12099:2017 | The sector’s NIR text: building the calibration set, validation, accuracy and precision checks, transfer between instruments, routine monitoring. |
| Regulation (EC) No 852/2004, hygiene of foodstuffs Regulation (EC) No 183/2005, feed hygiene | The requirement for a system based on HACCP principles. An in-line measurement can become the means of monitoring a critical point, provided the choice is justified and documented. |
| Regulation (EC) No 152/2009, of 27 January 2009 EUR-Lex serves the consolidated version in force | Methods of sampling and analysis for the official control of feed. The reference an in-line measurement is compared against when it sets out to inform a control. |
| Regulation (EU) No 1169/2011, food information to consumers | The nutrition declaration and its tolerances. An in-line measurement of moisture, fat or protein then becomes a labelling compliance tool rather than only a steering one. |
| Regulation (EC) No 178/2002, general food law | Principles, traceability and operator responsibility. It is what grounds the archiving of process data. |
| Codex Alimentarius CXC 1-1969, general principles of food hygiene ISO 22000, food safety management | The international frame and the voluntary one, which structure the approach outside the European Union or alongside it. |
Alongside these texts, it is usually the private audit schemes — IFS, BRCGS, FSSC 22000 — and the customer’s specification that set the real level of demand. That is a substantive difference from pharma, and it changes the question to ask: not what an authority expects, but what your customer expects, and by when.
Chemicals, polymers and cosmetics
| Text | What it governs |
|---|---|
| Directive 2014/34/EU, equipment and protective systems intended for use in potentially explosive atmospheres with directive 1999/92/EC on the worker side and the IEC 60079 series on the equipment side | The hard constraint of a chemical installation. A probe placed in a classified zone is certified for that zone, and that is settled together with the choice of technology rather than after it. |
| Regulation (EC) No 1907/2006 (REACH) Regulation (EC) No 1272/2008 (CLP) | Registration, evaluation and classification of substances. They frame the product rather than the measurement, and the distinction is worth making when reading a specification. |
| Regulation (EC) No 1223/2009 on cosmetic products, of 30 November 2009 | Responsible person, product information file, safety report. Its Article 8 requires manufacture in accordance with good manufacturing practice, and presumes compliance where manufacture follows the harmonised standards whose references are published in the Official Journal of the European Union. |
| ISO 22716 — Cosmetics: good manufacturing practices 2007 edition | The standard that plays that part: production, quality control, handling of out-of-specification product, documentation. The cosmetics counterpart of pharmaceutical GMP, where an in-line measurement sits as an in-process control. |
None of these texts has an equivalent of ICH Q2(R2): they ask for control methods that are appropriate and documented, without saying how to validate a multivariate procedure. The gap is filled by borrowing from pharma, and that is what a serious file does. Pharmaceutical rigour transfers; it simply is not compulsory.
USP technical guides
These guides are not USP–NF chapters and carry no requirements: each says so on its own first page — “This document is intended for informational purposes only. The document does not reflect the opinion of any USP Expert Bodies”. They download from usp.org. We cite them for what they are: the state of the art as USP describes it, with its sources.
| Guide | What it brings |
|---|---|
| 1. Control Strategies for Continuous Manufacturing of Solid Oral Dose Drug Products © 2023 | The control strategy for a continuous solid oral dose line. It covers the continuum approach to critical attribute classification, process control systems, PAT development and model validation. |
| 2. Process Modeling in Pharmaceutical Continuous Manufacturing © 2024 | The residence time distribution and how it is measured, by pulse input and by step change. This is the guide that shows the NIR probe used as the tracer detector, and that ties the line model to material traceability and diversion. |
| 3. In-vitro Dissolution Modeling for Oral Solid Drug Product Continuous Manufacturing © 2024 | Dissolution modelling, and the conditions under which it can stand in for the test. |
| 4. Control strategy for continuous manufacturing of drug substances © 2025 | The counterpart of the first guide, on the drug substance side. It gives the in-line, on-line, at-line and off-line definitions, and a table mapping each synthesis step to its attributes and to the techniques that follow them. |
| 5. Powder characterization for continuous manufacturing applications © 2025 | The powder properties that decide flowability, and the USP chapters that measure them: <616>, <786>, <811>, <1063>, <1097> and <1174>. It also names the PAT techniques used on dry granulation ribbons. |
Measurement standards and good practice guides
| Text | What it governs |
|---|---|
| ASTM D6122 — Standard Practice for Validation of the Performance of Multivariate Online, At-Line, Field and Laboratory Infrared Spectrophotometer, and Raman Spectrometer Based Analyzer Systems edition in force: D6122-25; the title was extended to Raman and to field analysers across successive revisions | In-service validation of a multivariate on-line analyser: periodic performance checks, comparison with the primary method, drift criteria. The text closest to process measurement outside pharma, and the one analyser specifications cite. |
| ASTM E1655 — Standard Practices for Infrared Multivariate Quantitative Analysis edition 17 (2017), reapproved in 2024 with no technical change — E1655-17(2024) | Building a multivariate infrared calibration: sample selection, validation set, out-of-domain sample detection, performance metrics. The equivalent, outside pharma, of USP <1039>. |
| ASTM E1840 — Standard Guide for Raman Shift Standards for Spectrometer Calibration 1996 edition, reapproved in 2022 — E1840-96(2022) | The Raman shift standards that check a spectrometer’s spectral axis. The starting point of any periodic check of a RAMAN analyser, outside pharma. |
| ISO 13322-1 and ISO 13322-2 — Particle size analysis — Image analysis methods part 1 (static): 2014 edition, confirmed in 2025; part 2 (dynamic): 2021 edition | Determining a size distribution by image analysis, with particles at rest or in motion. The frame for imaging when the quantity is a size; counting or classing is validated against your reference method. |
| ASTM E2968-14 — Standard Guide for Application of Continuous Processing in the Pharmaceutical Industry the edition cited by the USP technical guides; since revised as E2968-23, Standard Guide for Application of Continuous Manufacturing (CM) in the Pharmaceutical Industry | Concepts and principles of continuous processing in pharma. It is the source of the definition of residence time distribution and of the typology of control strategies — feedback, feed forward — that the USP technical guides take up by citing it. |
| ISO 22412 — Particle size analysis — Dynamic light scattering (DLS) 3rd edition, 2025; the 2017 edition was withdrawn on 5 September 2025 | The normative text for DLS: what an instrument must report, and the tolerances that go with it. We have not opened it — it is a paid standard — and we therefore quote no value from it. It is, however, the text to obtain when a particle size specification has to rest on something other than common practice: see two instruments, two figures. |
| ISO 21920-2 and ISO 21920-3 — Geometrical product specifications (GPS) — Surface texture: Profile 2021 edition; they replaced the withdrawn ISO 4287 and ISO 4288 on 20 December 2021 | The roughness parameters (part 2) and the specification operators (part 3). A roughness figure means nothing without its sampling length, evaluation length and filter: that triplet, not Ra alone, is what makes two measurements comparable. The withdrawn numbers are still widely quoted in the field, including on manufacturer documents predating 2021. |
| ISO 5725, accuracy — trueness and precision — of measurement methods and results a multi-part standard, each part revised on its own schedule: part 1 in 2023, part 2 in 2025 — cite the part and its edition, not the number alone | The statistical ground under comparing two methods, and the source of the metrological vocabulary used on this site. On the regulatory validation pages it is ICH Q2 vocabulary that applies: what ICH Q2 calls accuracy is what ISO 5725 calls trueness. |
| JCGM 100, the GUM — evaluation of measurement data and expression of uncertainty | The reference text on measurement uncertainty, which grounds what we write about the gap between two instruments. |
| DS 3077 — Representative sampling — Horizontal standard 3rd edition, DS 3077:2024, released in October 2024, Dansk Standard | The only standard devoted to the theory of sampling. It grounds what we write on representativeness and the sampling plan. |
| Esbensen & Abu-Khalaf — Before reliable near infrared spectroscopic analysis — the critical sampling proviso, Parts 1 and 2 Journal of Near Infrared Spectroscopy 30(6), 2022 | What the theory of sampling requires before any near infrared measurement, and the order of magnitude of sampling error against analytical error. It grounds the sampling module of our training. |
| GAMP 5 an ISPE good practice guide, not a standard | The classification of software into categories, which proportions the validation effort that falls to the operator. |
| ALCOA+ principles | The attributes expected of a datum: attributable, legible, contemporaneous, original, accurate, and the four that follow. They apply to a model as to a value. |
ASTM and ISO standards are revised regularly. We cite their number without the year: the edition in force at the time of the project is the one that applies, and it is checked against the issuing body’s catalogue.
This list is kept current. If a text you expect is missing, it is because no page rests on it — tell us, the question is a fair one.
Tell us which text frames your project. You will know what it asks of your measurement.
A dossier does not carry the same requirements whether it aims at steering, at documentary evidence, or at replacing a release test. Forty-five minutes are enough to place yours and to spot what an auditor would find missing.