Guidelines: The Feed Analysis Laboratory: Establishment and Quality Control

ISSN 1810-0708
15
FAO ANIMAL PRODUCTION AND HEALTH
guidelines
THE FEED ANALYSIS LABORATORY:

ESTABLISHMENT AND
QUALITY CONTROL
Setting up a feed analysis laboratory, and
implementing a quality assurance system
compliant with ISO/IEC 17025:2005
Cover photographs:
Left and centre: L.H. de Jonge
Right: FAO/Jon Spaull
15
FAO ANIMAL PRODUCTION AND HEALTH
guidelines
THE FEED ANALYSIS LABORATORY:

ESTABLISHMENT AND
QUALITY CONTROL
Setting up a feed analysis laboratory, and
implementing a quality assurance system
compliant with ISO/IEC 17025:2005
Authors
L.H. de Jonge
Animal Nutrition Group
Wageningen University
The Netherlands
F.S. Jackson
Manager, Nutrition Laboratory
Massey University
New Zealand
Editor
Harinder P.S. Makkar
FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS

Rome, 2013
Recommended Citation
de Jonge, L.H. & Jackson, F.S. 2013. The feed analysis laboratory: Establishment and quality control.
Setting up a feed analysis laboratory, and implementing a quality assurance system compliant with
ISO/IEC 17025:2005. H.P.S. Makkar, ed. Animal Production and Health Guidelines No. 15. Rome, FAO.
The designations employed and the presentation of material in this information product do
not imply the expression of any opinion whatsoever on the part of the Food and Agriculture
Organization of the United Nations (FAO) concerning the legal or development status of any
country, territory, city or area or of its authorities, or concerning the delimitation of its
frontiers or boundaries. The mention of specific companies or products of manufacturers,
whether or not these have been patented, does not imply that these have been endorsed or
recommended by FAO in preference to others of a similar nature that are not mentioned.
The views expressed in this information product are those of the author(s) and do not
necessarily reflect the views or policies of FAO.
ISBN 978-92-5-108071-9 (print)
E-ISBN 978-92-5-108072-6 (PDF)
FAO, 2013
FAO encourages the use, reproduction and dissemination of material in this information
product. Except where otherwise indicated, material may be copied, downloaded and
printed for private study, research and teaching purposes, or for use in non-commercial
products or services, provided that appropriate acknowledgement of FAO as the source and
copyright holder is given and that FAOs endorsement of users views, products or services
is not implied in any way.
All requests for translation and adaptation rights, and for resale and other commercial use
rights should be made via www.fao.org/contact-us/licence-request or addressed to
copyright@fao.org.
FAO information products are available on the FAO website (www.fao.org/publications) and
can be purchased through publications-sales@fao.org.
iii
Contents
Contents iii
Foreword vii
Abbreviations viii
Acknowledgements ix
Chapter 1
Introduction 1
1.1Background 1
1.2Aim 1
1.3A road map of the document
Chapter 2
Development of a business plan
2.1Introduction 3
2.2Approach for development of a business plan
2.2.1Type of laboratory to be developed
2.2.2Market analysis for potential customers and their needs
2.2.3Types of analyses
2.2.4Market analysis for available laboratory services
2.2.5Evaluation and decision making
Chapter 3
Setting up and running the laboratory
11
3.1Introduction
11
3.2Physical realization of the laboratory
11
3.2.1The analytical process
12
3.2.2 Methods to be made operational
13
3.2.3Building and facilities
18
3.2.4Equipment
24
3.2.5Organizational structure and responsibilities of personnel
26
3.3Realization of the laboratory Procedures
27
3.4Continuity and improvement of the laboratory
28
iv
Chapter 4
Implementation of a Quality Management System and the

road to accreditation
31
4.1Introduction 31
4.2Basic principles of quality
31
4.2.1Technical level
32
4.2.2Organization level
37
4.2.3Commercial level
40
4.3Reading and interpretation of ISO/IEC 17025:2005.
41
4.4A road map for building a high quality system
50
4.5First situation: Routine stand-alone feed analysis laboratory
50
4.5.1Introduction 50
4.5.2Initial phase
51
4.5.3First year
52
4.5.4Second year
52
4.5.5Third year
54
4.5.6Fourth year
56
4.6Second situation:Routine laboratory connected to a feed

manufacturer
58
4.7Third situation:Laboratory as part of a research organization
59
4.8Fourth situation:Government or reference laboratories
60
Sources used
63
Appendix A
Ensuring quality analytical performance

Appendix B
First line of quality control and the use of Shewhart charts

Appendix C
Validation requirements
65
73
77
Appendix D
Calculation of uncertainty of measurement
79
Appendix E
An example of technical records for a determination
81
Appendix F
An example of a maintenance and calibration document

Appendix G
An example of a training record
83
85
Appendix H
Procedure for traceability of volumetric calibration

Appendix I
Trend analysis
87
89
vii
Foreword
Feed has a fundamental influence on productivity, health and welfare of the animal. Feed
quality influences animal product quality and safety, and the environment. To achieve balance among these parameters, the animals nutritional requirements must be properly met.
Confidence in the nutritional information on any feed or feed ingredient provided by
suppliers is critical for buyers because it provides a guarantee of feed quality. Current reports
from many countries suggest that manufacturers and buyers do not always have confidence
in the data provided from non-accredited laboratories, which can negatively affect market
prices and international trade. It is therefore important that laboratories work towards
adopting a Quality Assurance System for all of their routine feed analyses. This has been
detailed in two FAO Animal Production and Health Manuals: No.14, Quality Assurance for
Animal Feed Analysis Laboratories, and No.16, Quality Assurance for Microbiology in Feed
Analysis Laboratories.
Not only must the methods used be of an internationally recognized standard, but
all steps in the process, from the initial sample submission through to the final report
preparation, must be traceable. An internationally accredited laboratory gives producers
and buyers of feed a great deal of confidence in the data they receive. This provides
wider market possibilities for feed manufacturers. Also, the right nutritional information
about feed ingredients and feeds will enable preparation of balanced diets that meet the
nutritional requirements to match the physiological stage of animals and to satisfy the
farmers husbandry objectives.
This document presents a step-by-step process to guide the laboratory management
team through the various stages, from planning the feed analysis laboratory building and
layout, to hiring suitable staff and choosing which methods to set up, with appropriate
equipment requirements. A detailed guideline for initiating a Quality Management System
starts with validation of methods, personnel and training; addresses systematic equipment
maintenance, calibration, proficiency testing and quality control procedures; and final
reporting and auditing, all culminating in a final accreditation inspection within an estimated
four-year time frame.
The authors have extensive laboratory experience as well as personal experience with
successfully bringing non-accredited laboratories up to an internationally recognized accreditation standard. The content of the document has been peer reviewed by a large number of
experts and their suggestions incorporated. The guidelines presented will assist governments
and feed manufacturers, as well as a range of institutions, including research and education,
to work towards establishment of a feed analysis laboratory whether as an integral unit
or as an independent commercial laboratory with internationally recognized accreditation.
Berhe G. Tekola
Director
Animal Production and Health Division
viii
Abbreviations
A.U.
Absorption Unit
AAS
Atomic Absorption Spectroscopy
ANOVA
Analysis of variance
CRM
Certified Reference Material
CUSUM
Cumulative sum
FAO
Food and Agriculture Organization of the United Nations
FAO/IAEA
FAO/IAEA Agriculture and Biotechnology Laboratory
GC
Gas chromatography
GC-FID
Gas Chromatography-Flame Ionization Detector
GC-MS
Gas Chromatography-Mass Spectrometry
HPLC
High Performance Liquid Chromatography
IAEA
International Atomic Energy Agency
ICP
Inductively Coupled Plasma [Analysis]
ICP-AES
Inductively Coupled PlasmaAtomic Emission Spectroscopy
ILAC
International Laboratory Accreditation Cooperation
ISO
International Organization for Standardization
LC-MS
Liquid ChromatographMass Spectrometer
LIMS
Laboratory Information Management System
LOD
Limit of Detection
LOQ
Limit of Quantification
MS
Mass Spectrometer
MS-MS
Sequential mass spectrometry
MU
Measurement of uncertainty
NIR
Near-Infrared Spectrometry
PR
Public Relations
PSG
Project Steering Group
QA
Quality Assurance
QMS
Quality Management System
R&D
Research and development
ROI
Return on Investment
SD
Standard deviation
SOP
Standard operating procedure
UV
Ultraviolet
ix
Acknowledgements
We thank all peer reviewers, listed below, for taking time to critically read this manual and
for their comments and suggestions that led to its improvement.
Copy editing and preparation for printing was by Thorgeir Lawrence and layout was
coordinated by Claudia Ciarlantini, which are thankfully acknowledged.
Jim Balthrop
Quality Assurance Manager, Office of the Texas State Chemist,

P.O. Box 3160, College Station, Texas 77841, United States of
America
Richard A. Cowie
Senior Quality Assurance Manager, SRUC, Ferguson Building,

Craibstone Estate, Aberdeen AB21 9YA, Scotland, United
Kingdom
Johan DeBoever
Senior Researcher Feed Evaluation, Institute for Agricultural and

Fisheries Research, Animal Sciences Unit, Scheldeweg 68, 9090
Melle, Belgium
E. Fallou Guye
Animal Production Expert, Animal Production and Health

Division, FAO, Viale delle Terme di Caracalla, 00153 Rome, Italy
Gustavo Jaurena
Professor, Animal Nutrition, Centre of Research and Services

in Animal Nutrition, Facultad de Agronoma, Univ. de Buenos
Aires, Av. San Martn 4453 - C1417 DSQ, Ciudad Autnoma de
Buenos Aires, Argentina
Alicia Njera Molina
International Quality Assurance Manager, Masterlab bv, The

Netherlands
E. Odongo
Animal Nutritionist, Animal Production and Health Section, Joint

FAO/IAEA Division, IAEA, Vienna, Austria,
Alfred Thalmann
Elbinger Str. 10c, D 76139 Karlsruhe, Germany

(Formerly: Staatliche Landwirtschaftliche Untersuchungs- und
Forschungsanstalt Augustenberg; later Landwirtschaftliches
Technologiezentrum Augustenberg)
Anja Tpper
Head, Animal Feed Analysis and Microbiology Unit,

Landwirtschaftliches Technologiezentrum Augustenberg,
Nelerstrae 23-31, D76227 Karlsruhe, Germany
Chapter 1
Introduction
1.1Background
The importance of livestock production in developing countries is increasing due to
growing demands for animal products, combined with demands for more sustainable
agriculture. Globalization and increased demand for animal products also offer export
opportunities, leading to improved welfare for people from the exporting countries. Key
elements for success, however, are high productivity and low prices, coupled with high
quality and safety of animal products. Chemical analyses of diets and animal products play
an essential role in achieving optimal production and guaranteeing safety for consumers. In
developed countries, it is common practice to undertake most of the analyses in accredited
laboratories. The availability of this kind of service in developing countries is constrained
due to fewer laboratories and a lack of infrastructure. This can lead to a lack of reliable
data, which can make animal agriculture in developing countries much less competitively
priced when compared with that in developed countries.
The situation in developing countries can be improved by increasing the number of
laboratories and improving the quality of the available analytical services. The provision of
relevant information and training can be used as a tool to improve the quality of the analytical services in these laboratories. For this purpose, an international group of laboratory
experts participated in an FAO working group that led to the publication of two manuals
describing quality assurance, safety issues and analytical methods: Quality assurance for
animal feed analysis laboratories (FAO, 2011), and Quality assurance for microbiology in
feed analysis laboratories (FAO, 2013). This group was also the starting point for an FAO
network of international experts, aimed at improving analytical capability in developing
countries.
Increasing the number of laboratories, however, is a challenging task. The initiative
to start a laboratory lies with local stakeholders and is based on an investment decision.
The high investment and the technical complexity can result in hesitation by stakeholders
because of a lack of expertise.
1.2Aim
The aim of this present document is to present guidelines for starting and running an animal
feed analysis laboratory, including the implementation of quality assurance (QA) systems
compliant with an international standard. To achieve this goal, the relevant information will
be described and illustrated by giving examples wherever appropriate, which should lead to
a better understanding by semi-technical persons and decision-makers.
The feed analysis laboratory: establishment and quality control
1.3A road map of the document

The document is divided into three parts.
The first part (Chapter 2) describes the initial phase of building an animal feed analysis
laboratory, a phase especially important for decision-makers and business developers. Some
critical and crucial decisions have to be made during this phase, with such central questions
as: What should the laboratory do? and What are the chances of success in terms of being
an economically viable unit? To answer these questions, stakeholders should first undertake
a market analysis and identify any potential customers, their particular requirements, and
other business opportunities to make the project viable. This information is necessary to
make a balanced judgement of the potential benefits or profits with respect to the investment required. This document addresses the issue from the point of view of different types
of laboratory. The investments needed for facilities, equipment, consumables and labour
(time and skill) will be given for the different types of analyses, including a global price
indication. Using this information, stakeholders should be in a better position to calculate
Return on Investment (ROI).
The second part (Chapter 3) of the document deals with the physical creation and running of the laboratory, and is especially important for financers and the laboratory staff.
The first section deals, among others, with the physical creation of the laboratory, such
as securing the land; building the laboratory, including the necessary facilities; employing
suitable laboratory personnel; purchasing appropriate equipment; and organizing the
laboratory (placement of various items of equipment and their safe operation). Special
attention is given to pertinent legislation, and health and safety aspects. The second section focuses on the organization of the primary process, from receiving sample material to
sending analysis reports and invoices, and management of the laboratory. Crucial elements,
such as storage, planning, traceability and confidentiality, are highlighted, and examples
of the related procedures described. Management aspects are separated into internal and
external. Internal aspects focus on optimization, and increasing efficiency and quality of
the laboratory processes. This also includes human resource management. External aspects
focus on improving the market position of the laboratory, including enhancing contact
with customers, identifying new customers, better positioning of the laboratory within the
marketplace, establishing contacts with national and international networks on laboratory analyses, participation in conferences that address feed and food analysis issues, and
exhibitions of laboratory equipment. Both aspects are important to ensure the successful
running of the organization.
The third and final part of this document (Chapter 4) describes the implementation of
a QA system, which is especially important for laboratory staff and the QA manager. The
quality of the analytical results produced should be guaranteed to gain the confidence
of customers. The implementation of a complete quality control system can take several
years of experience and refinement before it can be accredited. In this document, activities
and their time schedule are described for the implementation of quality principles with
the aim of achieving laboratory accreditation according to ISO/IEC 17025:2005 General
requirements for the competence of testing and calibration laboratories, which is an internationally recognized standard for quality systems within testing laboratories. Additional
information is also given about implementation of these principles in daily practice.
Chapter 2

2.1Introduction
Preparation of a business plan is essential for the creation of a new laboratory. A business
plan describes aims of the laboratory and a plan to realize them. It also assesses challenges
in achieving the aims and making a profit, as well as identifying potential opportunities for
the laboratory. The plan can be seen as a road map for the future laboratory to achieve
its final goal. The physical creation of a laboratory needs a relatively large investment, and
therefore a good business plan can also help to increase the confidence of potential investors. The realization of a business plan involves a range of experts, including marketing
and technical specialists, to provide a realistic plan for the new laboratory on which the
necessary investments can be based.
The absence of a business plan can lead to incorrect decisions regarding investments, or
to missed opportunities, which could lead to a non-profitable outcome.
Two different areas of expertise are necessary:
knowledge to assess the market situation and to identify opportunities and challenges; and
laboratory knowledge to evaluate types and requirements of potential analyses and
the estimation of laboratory costs.
A group to develop a business plan should be created to include personnel with expertise in both the above two areas. They should regularly interact while developing the plan.
A suggested approach to realize a business plan is described in the next section. This
approach should not be seen as definitive, but rather as a possible tool to create a business
plan. The example approach is limited to chemical analyses.
2.2Approach for development of a business plan

The key elements required to prepare a business plan are given in Figure 2.1.
2.2.1Type of laboratory to be developed

The characterization of the laboratory to be created is the first important step. From a commercial point of view, a distinction should be made between a laboratory as a stand-alone
unit, and one embedded in a larger organization. A laboratory could be one of five types:
Stand-alone commercial laboratories. These laboratories have an independent
judicial position, working on a commercial basis with the aim to make a profit. In
practice this means that they largely analyse samples sent from commercial clients.
Laboratories integrated with feed producing units. These laboratories are part
of a larger organization, but operate as a separate unit within the total chain of
quality control of animal diets and pre-mixes produced. In practice they are used for
figure 2.1
Sequential steps in the development of a business plan

1. Type of laboratory
2. Market analysis for potential customers

and their needs
3. Types of analyses to be conducted
4. Market analysis for available laboratory

services
5. Evaluation and decision-making
the control of feed ingredients and animal diet specifications. Sometimes they also
provide services to others outside their own organization.
Laboratories integrated with research organizations. These laboratories are
used to analyse samples from experiments performed within a research organization.
The analytical work conducted in these laboratories is a mixture of standard analyses
and more specific research analyses in a wide range of matrices. These laboratories
can operate as separate units or be fully integrated within a research division.
Laboratories integrated with educational organizations. The primary function
of these laboratories is to provide training opportunities for students. Students may
also analyse their own research samples.
Government or reference laboratories. The function of these laboratories is to be
a reference for other laboratories and thereby assist in maintaining and improving the
quality of analytical work conducted in the individual laboratories within a country.
They are also used by the government for regulatory analyses regarding feed and
food safety.
In some situations the laboratories connected to research and educational organizations
also analyse samples from commercial clients.
2.2.2Market analysis for potential customers and their needs

The next step is to identify and quantify potential customers and their needs. These could
be producers or users of feed ingredients, such as feed industries, researchers, nongovernmental and governmental organizations. The number and type of clients may vary
between different locations and reflect the level and sophistication of agriculture and
livestock activities in the region. Customers are critical for the revenue of the laboratory
and therefore its commercial existence. Good market analysis is necessary to assess the
opportunities for a new laboratory to be economically viable. This analysis should take
into account the maximum acceptable duration for transport of samples to the laboratory
(ideally no more than one or two days). It should be noted that differences in regulations
between countries may present difficulties in sending samples across international
borders.
The market analysis should be performed using the steps presented below, starting
with an initial analysis based on the current use of analytical services. This should focus on:
Number and type of customers.
Type of analyses and the amount spent on analytical services.
Organization of the analytical services: in-house versus outsourced analysis of
samples.
Customers can be divided into categories relating to the amount they are likely to spend
per year on analytical requirements, such as small-scale farmers; small- and medium-scale
feed manufacturing units; and large feed manufacturing mills. Analytical services should
be divided into different types of analyses. An obvious division is to separate the analytical services into proximate analyses (i.e. the classic animal feed analyses) and advanced
analyses that use sophisticated instruments (examples being minerals, amino acids and
contaminants). The amount spent on each type of analysis can be roughly calculated by
identifying the number of samples and the prevailing market prices.
A division between analytical services performed within the organization and those
outsourced to independent commercial laboratories should be made. Some companies may
have the capability to perform routine analyses in-house, such as nitrogen analysis, which is
likely to continue even if a new laboratory offers the same service. The ratio between both
types of services strongly affects the opportunities of the new laboratory.
The second step should assess additional and potential opportunities in the market that
are likely to utilize the services of the laboratory being established. Attracting new clients
should be a primary goal, and this could be achieved by offering a more comprehensive
testing facility, faster turnaround time or shorter travel distance than is currently available
in the area, or assisting with interpretation of the results and providing recommendations
about ration formulation.
A critical evaluation should be undertaken to establish why these potential clients would
use analytical services in the future, but are not currently using the services. Having the
ability to perform some unique analyses offers clear opportunities for the laboratory and
could strengthen its market position. Its realization, however, will have an impact on investments needed (see Section2.2.3).
The third step is to investigate and predict future market developments. Some important points that can be addressed are:
growth of the animal production sector;
pressure to produce animal products efficiently and sustainably;
national and international legislation for feed and product safety; and
volume of feed or feed ingredient export.
These issues will require the generation of new information and data, and therefore the
need for analytical services. As part of the market analysis it is important to seek advice
from local councils as well as national regulators to understand market trends and possible
changes to legislation. These should be taken into account at the planning stage. This
market analysis should be performed for each type of laboratory.
2.2.3Types of analyses
The types of analyses conducted by the laboratory affect both its market position in terms
of attracting potential customers, and the investment needed. The types of analyses can
be divided into five types:
Type1. Proximate analyses
Type2. Macro-minerals
Type3. Micro-minerals at trace level
Type4. Chromatographic analyses (e.g. amino acids, fatty acids)
Type5. Chromatographic analyses at trace levels (contaminants such as aflatoxins,
pesticides and pesticide residues, antibiotics, etc.).
The types of analyses will determine the investment needed. Proximate analyses are
used for feed characterization for general nutritional parameters, and the capacity to
perform these analyses should be seen as the minimum requirement for every laboratory.
Other types of analysis are more specialized and need specific equipment and facilities. For
minerals and chromatographic analyses, it is important to make a distinction based on the
required detection limit of the samples to be analysed. Determination of trace levels are
mostly performed to establish the presence or absence of a contaminant that could affect
public health, which governments try to protect by legislation (e.g. aflatoxins, pesticides,
pesticide residues and antibiotics). Consequently, these determinations not only require
highly skilled personnel and sensitive and expensive equipment, but also demand a higher
level of purity of chemicals used (including water) and clean work conditions to avoid
contamination.
Types of laboratories can be tentatively categorized as:
Basic nutrition laboratory performing only proximate analyses (Type1 analyses).
Laboratory conducting analysis of nutrients; performing proximate, mineral and chromatographic analyses (Types 1 to 4 analyses).
Laboratory conducting analysis of nutrients and anti-nutrients (Types 1 to 5 analyses).
All animal nutrition laboratories should be able to perform proximate analyses, with the
possibly of extending to analysis of other analytes in the future.
It is also possible for a laboratory to sub-contract some analyses if it is not economically
viable to set up and maintain capability for all types of analyses (this saves customers the
inconvenience of sending multiple samples and ensures the laboratory can still secure a
portion of the work, and hence income).
In order to estimate the investment required to perform different types of analyses, a
calculation of the cost to perform the proximate analyses as well as the cost to implement
other types of analyses should be made. The cost calculation should include costs for facilities, equipment, personnel and consumables.
2.2.4Market analysis for available laboratory services

The next step in the process is to examine the current market situation for performing analytical services, and the opportunities for changing this situation. The success of the new
laboratory depends mostly on the opportunity to take over part of the existing market, so
it is therefore important to investigate how much flexibility currently exists. This assessment
should focus on the following issues:
Estimation of the number and type of laboratories already present in a specific area.
For the characterization of the laboratories, the division as earlier mentioned (Section
2.2.1) can be used. This analysis will lead to identification of laboratories that can be
seen as competitors for the potential new laboratory.
Identification and assessment of the relationship between the laboratories and the
clients. This should give information on the amount different clients will spend on
obtaining laboratory services. The relationship between clients and a laboratory will
be based on the quality and promptness of services provided to a client. This will also
determine their loyalty and personal preference towards a company. Company policy
may dictate the flexibility each potential client will have to make a change in their
out-sourcing of laboratory business. Some of this information may be commercially
sensitive and difficult to obtain.
Visiting potential clients as part of a Public Relations (PR) exercise can be valuable in
establishing contacts in the industry and establishing an indication of the amount of
analysis work that could be available and the type of service expected by the clients
(e.g. are they dissatisfied with their current suppliers, and if so, for what reason?).
2.2.5Evaluation and decision-making

The last step in the creation of the business plan is to bring together all the information collected in the previous steps to make an evaluation of the different options. For calculating
profitability, the evaluation should estimate the potential revenues and costs. This should
be done for the different types of laboratories in Section 2.2.1 above.
The calculation of the potential revenue should be based on the results of the market
analysis for each of the clients (Section 2.2.2) and their relationship with the existing laboratories (Section 2.2.4). The first step in this calculation is to gather information on the different
types of analysis and the amount a client pays to each of the laboratories in the region. The
second step is to estimate how many potential clients would be willing to make a change
and to procure services from the new laboratory, and estimate this potential revenue. This
likely change can be expressed in terms of a probability factor with a scale from 0 to 1, with
0 being low probability of change to the new laboratory and 1 being a high probability of
change. The total revenue is estimated by multiplying the amount spent by the probability
factor. As there are a number of factors beyond the control of the laboratory, this figure will
only provide an estimate, as an exact figure is difficult to derive. In general, clients that use
more than one laboratory are more willing to switch to a new laboratory, compared with
those that use only one, provided the new laboratory meets all their analytical needs.
Focusing on a new market can be much more profitable than looking at just the existing
market, but it means more uncertainty because it depends on future plans of potential clients. Therefore, it is preferable to focus in the first instance on potential customers within
the present market.
The total costs should also be calculated for all the different types of analyses. A good
approach is to calculate firstly the costs to perform only the proximate analyses, followed
by an estimation of the additional costs to perform other, more advanced, analyses. Costs
will vary widely between countries, depending on freight costs, currency exchange rates,
availability of suitable equipment or of high grade consumables, and labour costs.
The expected profits for the different options can be calculated from the predicted total
revenue and costs. Calculation of profit, however, is based on various input variables that
contain uncertainties. A sensitivity analysis can easily be performed by changing the value
of input variables and assessing the effect on the calculated profit. This analysis should
be limited to those variables that contain the highest uncertainty and therefore have the
greatest effect on the accuracy of the calculation. Some examples of such variables are the
probable costs and the prices of analyses charged by other laboratories. The evaluation
of this uncertainty can be performed by estimating the difference between the predicted
profit and a non-profit situation.
The new laboratory should be careful if using the market price for calculating the revenues because it will be competing with other, established laboratories. The new laboratory
should avoid using prices that are too low in order to capture a part of the market: a reduction in prices might increase the volume of work, but the consequences could be a decrease
in profit as a result of prices with too little profit margin. Also, it could lead to a general
price war amongst laboratories, with destructive consequences. In addition, laboratories
might find it difficult to raise prices after a certain period.
The final result of these calculations are values for the profitability of the laboratory
under different conditions, expressed as an average value with a confidence interval. The
range of this confidence interval reflects mainly the uncertainty in the estimation of the
potential revenues. The profitability is often expressed as Return on Investment (ROI) which
is related to the investments required. The decision-makers and investors will use these
values to make decisions regarding the investment in the new laboratory. Depending on
the type of laboratory (see Section 2.2.1), however, the decisions could be made in different ways:
Stand-alone commercial laboratories. The decision will be based purely on the
level of profitability and its confidence interval. The uncertainty expressed in the range
of the confidence interval will positively affect the margin of profit that investors
demand.
Laboratories integrated with feed producing units. The decision should primarily
be influenced by the profit a feed producer is likely to make from the feed manufacturing activity. In this case, an alternative approach is to compare the costs of the
laboratory against those for outsourcing the analyses to be undertaken for effective
running of the feed manufacturing unit. A laboratory integrated into the feed manufacturing unit has several benefits, such as a quick turnaround time, not dependant
on any outside laboratory, and better quality control of the products. These benefits
should also be quantified and taken into consideration when making a decision.
Laboratories integrated with research institutes. The decision will primarily be
influenced by the additional value the laboratory brings to the research conducted in
the institute. Although such laboratories can also perform analyses for commercial
clients, practice shows that these laboratories often have difficulty competing with
stand-alone commercial laboratories, mainly due to the high throughput of the latter
and thus lower unit costs per analysis. Nevertheless, research laboratories should also
try to generate additional revenue by attracting commercial clients. The additional
revenue could help finance new instruments and analyses required to meet fast
moving research needs.
Laboratories integrated with educational organizations. The decision should
primarily be based on enhancing the quality of education and producing graduates
and researchers having the required skills for addressing the challenges of the industry
and capable of contributing substantially to cutting-edge science. As for the laboratories integrated with research institutions, the educational laboratories should try
to attract commercial clients; however, they should be cautious and not assign such
work to staff lacking the appropriate training and competence, such as trainees and
students.
Government or reference laboratories. The decision to create this type of laboratory should, in contrast to the other laboratories, be based not on commercial benefits but purely on regulatory reasons. Its central position and high quality demands
require a large investment in personnel, equipment and facilities, that should be
funded by public money to guarantee the independence of the laboratory and avoid
conflict with commercial interests.
11
Chapter 3
Setting up and running the

laboratory
3.1Introduction
After a positive decision is made to set up a laboratory, actions need to be taken to put into
practice the business plan developed in Chapter 2. The setting up of a new laboratory involves:
Selection or construction of a building and facilities required for various analyses
(Section3.2.3).
The analytical process and an organizational structure to facilitate this (Sections3.2.1
and 3.2.5).
Selection of analyses to be performed (Section3.2.2).
Selection and purchase of equipment (Section3.2.4).
Attracting and maintaining qualified staff (Section3.4).
Establishing standard operational procedures (SOPs), i.e. formally written controlled
documents outlining all the steps for each of the methods the laboratory decides to
undertake (Section3.3).
In most cases a Project Steering Group (PSG) is constituted to bring the above six stages
to a satisfactory completion within a fixed time schedule. This group should contain a
technical expert with experience in a feed analysis laboratory, a laboratory manager with
experience in quality assurance (or Quality Assurance (QA) manager), and personnel from
finance, procurement and management areas. Throughout the process of setting up the
laboratory, good communication among the members of the PSG is vital to ensure success.
This chapter will focus on the realization of an analytical chemical laboratory. For microbial determinations, which are also important analyses in animal feedstuffs, special facilities
and equipment are needed, such as clean and separate rooms to avoid contamination, and
flow cabinets to work under safe conditions. A detailed description of these requirements
is outside the scope of this manual. Special facilities and equipment to conduct laboratory
research in the field of animal nutrition, such as in vitro or in situ investigations, are also
outside the scope of this manual.
3.2Physical realization of the laboratory

The physical realization of an operational laboratory involves the construction of the laboratory, purchase of equipment and appointment of personnel. All these issues are related
to the analytical work, and more specifically to the methods, that the laboratory intends
to conduct. The choice of methods is therefore a critical step (see Figure3.1). This section
starts with an overview of the analytical process and available methods, followed by their
implications for construction or selection of buildings and facilities, purchase of equipment
and putting in place an organizational structure with defined responsibilities for the personnel involved.
12
figure 3.1
Main steps in the development of an operational laboratory
Business plan
Choice of methods
Equipment
selection and
purchase
Selection or construction of
building and facilities
Development of organizational
structure and assigning
responsibilities to personnel
Technical knowledge and experience should not dictate Choice of methods. It should be
directed only by commercial or scientific criteria, as stated in the business plan. If some technical knowledge and skills are not available, then appropriate technical staff can be recruited.
3.2.1The analytical process

The analytical process is the foundation of the laboratory, and ensures that procedures
and protocols are followed to consistently produce results of a high standard that meet
international QA requirements. The various stages of the analytical process are described
in Figure3.2.
This process starts with the receipt of samples and a request from the client for the
analyses to be performed. On receipt of the samples and appropriate sample preparation
figure 3.2
The various stages of the analytical process

Receiving sample material and information on the
required tests as requested by clients
Acceptance of samples and tests to be conducted
Sample preparation and storage
Performance of required analyses
Acceptance of results
Final reporting and sending invoice
and storage, analyses can begin. The results of these tests are collated and checked, and
once approved by an authorized person, a final report, including an invoice, is sent to the
customer. It is important to make sure that all requests from clients have been noted as
well as the most suitable method chosen (if alternatives are available such as fat with or
without prior hydrolysis). Responsibility for checking these details should be clearly defined.
Sample materials are stored in the laboratory for a fixed time, e.g. one month, from
completion of analyses and either returned (on request), discarded or destroyed.
3.2.2 Methods to be made operational

The methods used in a feed analysis laboratory can be divided into proximate and specific
instrumental analyses. The proximate analyses are used for characterization of feeds based
on macro-nutrients, such as dry matter, protein, fat, fibre and ash, whereas the advanced
(instrumental) analyses focus on specific components, like individual minerals, amino acids
and fatty acids. A detailed description of these methods can be found in a previous FAO
publication (FAO, 2011). Since the level of analytical skills required differs for each method,
it is vital that each technician is signed off as being trained and capable of performing
each specific method, and this should be recorded in the training file.
Near-Infrared Spectrometry (NIR) is sometimes mentioned as an alternative method
to estimate the nutritional composition of feeds, and needs only a power supply and a
grinding machine, and possibly also an internet connection. However, laboratories should
regard NIR only as an additional method that should be based on the results they have
determined by traditional methods, as described in this chapter.
The main analyses and the steps involved in conducting these are given below (for more
specific method details, see FAO, 2011).
For personnel, technical skill levels should be categorized at three levels:
Basic analytical skills. Familiarity with the use of balances, sample mixing and
sub-sampling. No specific background or education required other than an attentive
attitude and attention to detail.
Medium analytical skills. Understanding of basic chemical reactions and the principle of the method being applied; awareness of laboratory safety when working
with solvents and strong acids and bases; computer competency; use of gas cylinders;
spectrophotometer; use of analysis-specific equipment such as Fibertec or Soxtec; and
bomb calorimeter. Laboratory experience is essential.
High analytical skills. Specific instrument training, i.e. High Performance Liquid
Chromatography (HPLC); Ultra HPLC; Gas Chromatography (GC); Gas Chromatography-Mass Spectrometry (GC-MS); Liquid Chromatography-Mass Spectrometry
(LC-MS); and Inductively coupled plasma atomic emission spectroscopy (ICP-AES);
associated software programs; able to maintain required instrumentation. Ability to
make independent decisions regarding peak identification and its area. Awareness
of laboratory safety requirements when working with toxic and carcinogenic compounds. Laboratory experience is essential, along with relevant University or recognized technical qualifications.
Typical tests in a feed analysis laboratory require specific facilities and skills. These are
listed in Table 3.1.
13
14
Table 3.1
Typical tests in a feed analysis laboratory and their technical requirements

Parameter
Description
Sample preparation
Description
Sample preparation is essential for sub-sampling of material prior to a determination.
Activities
Drying and grinding.
Equipment
Low temperature oven dryer (6070 C) or freeze dryer; splitter; mill; sieves.
Facilities
Two- or three-phase electric power; exhaust system.
Personnel
Basic analytical skills.
Dry matter analysis

Description
Dry matter is by definition the part of the sample that remains after drying at 103 C.
Activities
Weighing and drying.
Equipment
Analytical balance (0.1mg); forced-air drying oven (at least 110 C); desiccator.
Facilities
Granite (or similar) table for balance stability; an oven connected to an exhaust system.
Personnel
Crude ash
Description
Crude ash is by definition the part of the sample that remains after incineration at 550 C.
Activities
Weighing and incineration.
Equipment
Analytical balance (0.1mg); desiccator; muffle furnace.
Facilities
Connection to exhaust ventilation system for muffle furnace; granite (or similar) table for balance
stability.
Personnel
Ash insoluble in acid (sand)

Description
Ash insoluble in acid is the ash that remains after boiling in strong acid.
Activities
Weighing, boiling and incineration.
Equipment
Analytical balance (0.1mg); desiccator; muffle furnace; heating and reflux equipment.
Facilities
Granite (or similar) table for balance stability; a fume hood connected to an exhaust system.
Personnel
Crude protein
Description
The term crude protein refers to measuring the total nitrogen content and to calculate the
protein content by multiplying the nitrogen content by an appropriate conversion factor (usually
6.25). If an alternative method such as the summation of amino acids is used, the term crude
protein should not be used. Two methods, Kjeldahl and Dumas, are available for nitrogen
determination.
Crude protein Kjeldahl method

Description
Nitrogen is converted into ammonia which is absorbed in boric acid and titrated against a
standard acid.
Activities
Weighing, digestion, distillation and titration.
Equipment
Analytical balance (0.1mg); digestion unit; distillation unit; titration unit.
Facilities
Granite (or similar) table for balance stability; fume hood connected to an exhaust system.
Personnel
Medium-level analytical skills.

(cont.)
15
TABLE 3.1 (cont.)

Parameter
Description
Crude protein Dumas method

Description
With complete combustion of sample at 950 C in the presence of oxygen, nitrogen is converted
to a gaseous state and reduced to N2, followed by measurement in a thermal conductivity cell.
Activities
Weighing, combustion to N2, and measurement.
Equipment
Analytical balance (0.1mg); Dumas apparatus.
Facilities
Granite (or similar) table for balance stability; helium and oxygen gas supply (high purity; 5.0).
Personnel
Medium- to high-level analytical skills.
Crude fat
Description
Crude fat is by definition the non-polar extractable fraction of the sample. The extraction can
be performed with or without prior acid hydrolysis, both being complementary methods. The
laboratory should offer both options.
Activities
Weighing, hydrolysis, filtration, extraction and drying.
Equipment
Analytical balance (0.1mg); units for heating, filtration, extraction and refluxing; forced-air drying
oven or vacuum oven (preferable).
Facilities
Personnel
Fibre analysis
Description
Fibre analysis is based on boiling of the sample in a special detergent solution and measurement
of the remaining organic fraction. There are two methods available, both are based on digestion
of feeds in detergent solution: (1)digestion of feed directly in the detergent solution and
filtration using crucibles (this is the official standard method); and (2)digestion of sample whilst in
a nylon bag and then washing the bag containing the digested sample to make it detergent free.
Fibre analysis Crucible-based filtration method

Description
Digestion of feed directly in the detergent solution and filtration using crucibles (this is the official
standard method).
Activities
Weighing, boiling, filtration, drying and incineration.
Equipment
Analytical balance (0.1mg); hot plate; reflux and filtration unit; forced-air drying oven; muffle
furnace; crucibles.
Facilities
Personnel
Fibre analysis Nylon bag-based method

Description
Digestion of sample whilst in a nylon bag and then washing the bag containing the digested
sample to make it detergent free.
Activities
Weighing, boiling, washing, drying and incineration.
Equipment
Analytical balance (0.1mg); ANKOM apparatus; forced-air drying oven; muffle furnace.
Facilities
Personnel
Starch
Description
Starch can be measured by the classical Ewers method or with an enzymatic method. The
enzymatic method can be used for all sample types and is therefore preferable.
Activities
Weighing, extraction, incubation, dilution, and spectrometric measurement.
Equipment
Analytical balance (0.1mg); temperature-controlled water bath; autoclave (optional); suction unit;
volumetric equipment; spectrometer.
Facilities
Granite (or similar) table for balance stability; fume hood with vacuum system.
Personnel

(cont.)
16
TABLE 3.1 (cont.)

Parameter
Description
Reducing sugars
Description
Reducing sugars contain the most important sugars, including glucose, fructose and sucrose.
Determination is based on the Luff-Schoorl principle.
Activities
Weighing, incubation, dilution, and titration or spectrometric measurement.
Equipment
Analytical balance (0.1mg); temperature-controlled water bath; volumetric equipment; titration

unit or a spectrophotometer (depending on method).
Facilities
Granite (or similar) table for balance stability.
Personnel
Gross energy
Description
Gross energy represents the total energy value of the sample and is measured by bomb
calorimeter.
Activities
Weighing, instrumental measurement, and titration.
Equipment
Analytical balance (0.1mg); bomb filling system; bomb calorimeter; titration unit.
Facilities
Granite (or similar) table for balance stability; oxygen supply; fume hood connected to an exhaust
system.
Personnel
Medium-level analytical skills, with some experience.
Minerals
Description
Minerals are generally measured by spectrometric methods following incineration and hydrolysis.
Activities
Weighing, incineration (optional), acid digestion, dilution, spectrometric measurement,

and Atomic Absorption Spectroscopy (AAS) or Inductively Coupled Plasma-Atomic Emission
Spectroscopy (ICP-AES) instrumental measurement.
Equipment
Analytical balance (0.1mg); muffle furnace (550 C) with connection to an exhaust system
(optional); heating plate or digestion unit (250 C); volumetric equipment; AAS and spectrometer
or ICP-AES.
Facilities
Granite (or similar) table for balance stability; fume hood connected to vacuum system; acetylene
and air supply for AAS, or argon supply and three-phase current for ICP-AES; high purity water.
Personnel
Amino acids (excluding tryptophan)

Description
The standard method for the determination of amino acids is based on the hydrolysis of
protein to amino acids using a strong acid with or without previous oxidation, followed by
chromatographic separation and detection after derivatization.
Activities
Weighing, oxidation (optional), hydrolysis, evaporation, and chromatographic measurement.
Equipment
Analytical balance (0.1mg); hydrolysis unit; oven (110 C); evaporation equipment; HPLC or
dedicated amino acid analyser.
Facilities
Granite (or similar) table for balance stability; fume hood connected to an exhaust system; cold
water supply; compressed air for autosampler (optional); helium supply for degassing buffer
solutions; high purity water.
Personnel
High-level analytical skills.
Amino acids tryptophan

Description
Determination of tryptophan is based on an alkaline hydrolysis followed by chromatographic

separation.
Activities
Weighing, hydrolysis, and chromatographic measurement.
Equipment
Analytical balance (0.1mg); air-forced oven dryer; HPLC system attached to a UV- or fluorescence
detector.
(cont.)
17
TABLE 3.1 (cont.)

Parameter
Description
Facilities
Granite (or similar) table for balance stability; fume hood connected to an exhaust system;
compressed air for autosampler (optional); helium supply for degassing buffer solutions; high
purity water.
Personnel
Fatty acids
Description
The standard method for fatty acids is based on isolation and derivatization, followed by gas
chromatographic separation.
Activities
Weighing, derivatization, and chromatographic measurement.
Equipment
Analytical balance (0.1mg); air-forced oven dryer; GC system attached to a Flame Ionization
Detector (FID).
Facilities
Granite (or similar) table for balance stability; fume hood connected to an exhaust system; cool
water supply; compressed air for autosampler (optional); helium supply as carrier gas; hydrogen
and compressed air for the FID detector.
Personnel
Vitamins
Description
Determination of individual vitamins is based on extraction, followed by clean up, concentration if

needed, and chromatographic measurement.
Activities
Weighing, extraction, purification, and chromatographic measurement.
Equipment
Analytical balance (0.1mg); temperature-controlled water bath; unit for solid phase extraction;
volumetric equipment; HPLC system including UV- or fluorescence detector.
Facilities
compressed air for autosampler (optional); helium for degassing elution solution; high purity
water.
Personnel
Mycotoxins
Description
Mycotoxins are undesirable substances produced by fungi (moulds). These present a potential
danger to animal and human health. The maximum levels are nationally and internationally
regulated. The different methods are based on extraction, purification, chromatographic
separation and detection.
Activities
Weighing, extraction, purification, and chromatographic measurement.
Equipment
volumetric equipment; HPLC system including the possibility for pre- or post-column derivatization
and fluorescence detection.
Facilities
Granite (or similar) table for balance stability; biological safety cabinet; fume hood connected
to an exhaust system; compressed air for autosampler (optional); helium for degassing elution
solution; high purity water.
Personnel
High-level analytical skills
Pesticides
Description
Pesticides are undesirable substances whose maximum levels are defined in national and
international regulations. These regulations demand a low detection limit and positive
identification of the pesticides, which is achieved by using mass spectrometric detection. The
methods are based on extraction, purification, derivatization, chromatographic separation and
identification.
Activities
Weighing, extraction, purification, derivatization, and chromatographic measurement.
Equipment
volumetric flasks; GC-MS, including a databank for identification of the individual components.
Facilities
compressed air for autosampler (optional); helium as a carrier gas.
Personnel
18
3.2.3Building and facilities

The next step in the process is the choice of location, and construction of the laboratory,
including facilities. If connected to a large organization, such as a feed manufacturer, the
location of the laboratory is generally pre-determined. If there is a choice as to where the
laboratory could be located, the presence of some basic requirements, including good
infrastructure (i.e. road system) and assured and uninterrupted power and water supply, are
crucial elements in the selection of the location. The location should also be chosen with
a view to possible expansion plans in the future, and ensuring that local legislation allows
the construction of a laboratory on that site.
After confirming the location, a decision regarding the type of building is required. In
some cases there will be the possibility of renovating an existing building, which might
appear to be a financially attractive option. In practice, however, refurbishment of an old
building is often very expensive and complex, which may lead to a sub-optimal situation
requiring possible future expansion of the laboratory. Construction of a new laboratory
building is therefore the preferred option. For either building option, advice from an architectural company specializing in laboratory design will be invaluable, especially with respect
to conforming to the relevant local and national health and safety requirements.
A good approach while designing the laboratory is to base it on the methods to be used,
and possible future analytical requirements. Allow for extra power points, water outlets and
fume hoods wherever possible, and computer network access points as appropriate.
Suggested divisions or sections for a feed analysis laboratory are as follows:
Sample registration. An area convenient for clients and couriers to deliver samples.
Here the samples are logged into a booking system, given a unique identification
number, and the required analyses selected. Samples are then passed on to the
sample preparation section.
Sample preparation. An area where sub-sampling, blending, grinding and, if necessary, pre-drying occur. Pre-drying of samples is necessary if the laboratory deals with
products with a moisture content >15%, such as silage samples or manure. For predrying, freeze drying is preferable to oven drying at 70 C as there is less damage to the
structure of the sample and to the heat-sensitive components such as amino acids and
vitamins. Weighing, drying and incineration may also be conducted in this area. Keep
heat-generating equipment such as ovens and furnaces in one area, and if necessary an
air extraction unit can be utilized to remove odour as well as excess heat.
Digestion, filtration, distillation, dilution and titration. Keep a separate area
for acid use and storage. These areas require access to water and should be close to
glassware supplies, balances, fume hoods and chemical supplies.
Extraction, derivatization and dilution. Keep a separate area for solvent use and
storage. This area requires access to water and should be close to glassware supplies,
balances, fume hood and chemical supplies.
Instrumental measurements. Depending on the choice of instruments, specialized
conditions may be required, such as air conditioning, dust-free work area, no direct
sunlight, special type of power supply, gas supply, etc.
The separation of laboratory space to perform the above activities is primarily required
to avoid cross-contamination with undesirable substances and to maximize the use of
19
figure 3.3
Schematic presentation of the different sections of

a feed analysis laboratory
Sample registration
Sample preparation
Weighing and drying

and incineration
Digestion filtration
distillation dilution
titration
Extraction and
derivatization, and
clean up and dilution
Instrumental
measurements
space and staff time. Sample preparation produces dust and noise, and should be physically separated from other activities. Weighing, drying and incineration are mostly linked
to each other and do not involve working with chemicals. Traditional wet chemistry is split
into two sections based on the type of chemicals (i.e. strong acids and bases versus organic
solvents), and these two sections should be physically separated to avoid health and safety
issues with mixing of flammable solvents and corrosive chemicals. Also, keep separate fume
hoods for each of these sections. Sensitive instruments are placed in a clean environment,
away from other activities. If mycotoxins are to be analysed there will be a requirement for
a biological safety cabinet to be available in the laboratory. Figure3.3 shows a schematic
presentation of different sections of a feed analysis laboratory.
For maximum efficiency, after the samples have been registered and the analyses
assigned to technical staff, laboratory activities should be separated into five different sections, each with different requirements.
Section 1.Sample preparation
Area
ca 24 m2
Equipment and related items Grinding machine and sieves

Brushes for cleaning sieves and grinder

Three-phase electric power

Cubicles connected to a ventilation system for grinding

Drying ovens

Freeze drier

Compressed air

Network connection or data transfer for automatic weight
recording from balances into a spreadsheet (optional).
20
Furniture and related items

Safety items

Work table/bench
Storage facilities at low temperature (for example refrigerator, freezer) as well as at room temperature
Quarantine facility for imported or potentially infectious
samples.
Dust masks
Safety glasses and ear protection
Hand washing facilities
First aid kit
Section 2.Weighing room (including drying and incineration)

Area
ca 36 m2
Equipment and related items Granite (or similar) weighing tables
Balances
Cubicles connected to an exhaust system (care is needed
to ensure there is no draught produced which could affect
the balance accuracy)
Three-phase power (for muffle ovens)
Weighing balances
Network connections or data transfer for automatic
weight recording from balances into spreadsheet
(optional).
Furniture and other items
Work tables and benches, including chairs
Storage facilities
Glassware and other standard laboratory items such as
desiccators, tongs, spatulas, beakers and crucibles.
Safety items
Laboratory coat
Dust mask and gloves
Heat resistant gloves
Fire extinguisher
First aid kit
Section 3.Digestion room
Area
ca 48 m2
Equipment and other items Digestion heating blocks

Fume hoods connected to an exhaust system (suitable for
acid fumes)

Water and gas supply

Vacuum facilities
Furniture and related items Work tables and benches, including chairs

Safety cabinets for storage of chemical solutions (acids and
bases to be stored separately) and chemicals

Glassware, including beakers, crucibles, dispensers,
pipettes and measuring cylinders
Safety

Laboratory coat
Gloves
Safety glasses
Eye wash station
Fire extinguisher
Containers for chemical waste
Chemical spill kits
First aid kit
Section 4.Extraction room

Area
ca 48 m2
Equipment and related items Fume hoods connected to an exhaust system for solvent
extraction
Filtration unit
Reflux system
Acid concentrator
Rotary evaporator
Fibre and fat extraction systems
Centrifuges
Water and gas supply.
Furniture and other items
Safety cabinets for storing chemical solutions and organic
solvents and chemicals
Other laboratory items such as glassware, dispensers,
transfer pipettes, crucibles, extraction thimbles, etc.
Safety items
Laboratory coat
Gloves
Safety glasses
Eye wash station
Fire extinguisher
Containers for chemical waste
First aid kit
Section 5.Instrument room
Area
ca 48 m2
Equipment and related items Gas Chromatograph with or without Mass Spectrometer
detector (GC, GCMS)
High Performance Liquid Chromatograph (HPLC) with or
without Mass Spectrometer detector (LC-MS) and Ultra
HPLC
Atomic Absorption Spectrometer (AAS)
Inductively Coupled Plasma Atomic Emission or Mass Spectrometer (ICP-AES or ICP-MS)
Amino acid analyser
21
22
Furniture and related items

Safety items
Note:
Spectrophotometer
Vortex
Ultrasonic water bath
Filtering system
N-analyser
Uninterrupted power supply
Water and gas supply
Air conditioning
Purified water system for chromatography work
Network connections to allow direct laboratory access to
data generated from the equipment
Equipment manuals
Laboratory coat
Gloves
Safety glasses
Eye wash station
Fire extinguisher
First aid kit
If mycotoxins or residues are to be analysed, all processes
that pose a risk to the health of operators should be performed in a biological safety cabinet.
Additionally, the laboratory should have the following separate facilities:

Section 6.Sample storage room
Area
ca 24 m2
Equipment and related items Air conditioned or cool dark room
Freezer
Refrigerator
Furniture and related items Storage possibilities such as shelves and cupboards for
sample material
Safety
Fire extinguisher
Section 7.Dish washing and drying facility
Area
ca 12 m2
Equipment and related items Hot-air oven (110 C)

Dishwasher and autoclave (both are optional)
Furniture and related items Water supply and drain

Tiled floor and walls

Work table and bench

Storage facilities
Section 8.Administration or office area

Area
ca 20 m2
Facilities
Computers

Air conditioning unit

Scanner

Photocopier

Fax
Furniture and related items Work tables, including chairs

Client information

Storage facilities for results
Section 9.Welfare and rest area
A suitable clean area should be available for staff to take breaks, eat meals etc. This should
include toilets and showers; and a changing area should be available for staff to change
and store their clothing. Appropriate laundry facilities may be required for laboratory coats.
A total of ca 236 m2 is required to conduct the analytical work in a safe and efficient
manner. The exact area of the laboratory depends on the number of instruments and staff
employed. The space mentioned above will be sufficient to accommodate all basic equipment
for the analyses and eight technical staff, this space can easily deal with a few thousand
samples each year. If the laboratory has fewer technical staff and is dealing with a small
number of samples the area required can be reduced accordingly, to a minimum of 40 m2
to conduct the basic proximate analyses (e.g. dry matter, ash, crude protein, crude fibre and
crude fat) and the spectrometric determination of macro-elements. In some laboratories, the
Instrument room is partitioned based on the type of equipment, such as chromatographic
(e.g. HPLC, GC) and spectrometric equipment (e.g. AAS and ICP); a partition to isolate the
noisy equipment such as centrifuges and vacuum pumps is strongly advised.
It is preferable that the various sections of the laboratory are located on the same floor.
This facilitates the safe transport of samples and chemicals. It is also more efficient for technical
staff not to have to carry samples and potentially dangerous chemicals to different areas along
corridors, etc. The different sections should be arranged in a logical manner to minimize distances and to avoid conflict in activities. A suggested option is to cluster the sections as follows:
Sections 2, 3, 4 and 5 close to each other, as they are all part of the analytical process.
Sections 1 and 6 next to each other so that the samples can be conveniently stored
after appropriate preparation and analysis.
Sections 7 and 8 should be placed separately to guarantee clean working conditions
for both the activities.
Section 9 should be separate from 1 to 8, perhaps located adjacent to office and
other clean facilities.
All this information will help the PSG to draw up a draft design for the future laboratory.
To finalize this design, the following steps are needed.
Ensure the draft design, including the technical specifications, such as ventilation, is
evaluated by an external consultant, such as an architectural company specializing
in laboratory design. The technical specifications will vary from country to country,
23
24
depending on legislative requirements. Local construction companies as well as the

local regulatory agency may be able to recommend suitable companies to contact for
this purpose. It is advisable to use their expertise to improve the design, and it may
be prudent to involve them at an early stage of the process.
Evaluate the draft design, focusing on safety and environmental dimensions. As stated
previously, the laboratory should guarantee safe and healthy working conditions for all
staff and avoid pollution or contamination of the surrounding environment. Important
issues are storage of chemicals; safety while working with explosive or toxic gases and
chemicals; ventilation requirements; and waste control and disposal. In most countries,
there is legislation on these issues. The laboratory should ensure that the required
facilities are provided to enable work to take place as per regulatory requirements.
Consultation with national authorities on this issue can be very helpful.
Contact different construction companies to get estimates for the cost of constructing the design. These companies should have experience and expertise in the
construction of laboratories. They must also be familiar with health and safety regulations, and special materials required (for example acid- and alkaline-resistant bench
tops and floors).
The final design, including the projected cost for the completed project, will need to be
approved by the decision-makers, prior to starting the tendering and construction stages.
This process should be carefully monitored by members of the PSG. At each stage of the
design process, check that the basic constituents remain as required and meet the required
quality and safety criteria.
3.2.4Equipment
The equipment for the laboratory depends on the analyses it intends to conduct, as listed
in Sections3.2.1 and 3.2.2.
For proximate analyses, laboratories have the option of performing the analyses manually
or to use specialized equipment. The advantages of the manual methods are lower costs for
equipment and generally lower maintenance costs. These methods are, however, more laborious and time consuming. A list of equipment required is given in Table3.1. Costs given in
Tables3.2 to 3.4 are indicative, based on typical prices in the Netherlands in 2013.
If an equipment item is utilized for several different methods, such as an analytical balance, more than one may be required to ensure maximum efficiency of staff time.
The list in Table3.2 shows that at least 35000 is needed in order to set up a basic
laboratory able to perform the proximate analyses by manual methods. An alternative to
using manual methods is to purchase specialized equipment for the different determinations, as described in Table3.3. This approach requires a much higher financial investment,
and is only justifiable if a high sample throughput can be guaranteed.
For specific instrumental analyses, there are no alternatives other than to use dedicated
equipment. The cost for this equipment varies between suppliers and the technical specifications. Table3.4 gives indicative prices for the most commonly used equipment for each
type of analysis.
The laboratory also requires standard consumables and tools to facilitate the analytical
process.
25
Table 3.2
Equipment needed to manually perform proximate analyses

Equipment
Analyses
Costs (Euro)
Analytical balance
All
1 000
Drying oven
Dry matter, fibre and fat
2 000
Muffle furnace
Ash, acid insoluble ash, and fibre
2 500
Heating device
Acid insoluble ash, fibre, and fat
1 500
Extraction and reflux unit
Fat
Digestion unit
Nitrogen
10 000
Distillation unit
Nitrogen
3 500
Automatic titration unit
Nitrogen and sugar
Filtration unit
Fibre and fat
2 000
Temperature controlled water bath
Sugar and starch
2 000
Spectrometer
Sugar and starch
3 000
Volumetric equipment
Sugar and starch
1 500
3 000
350
Notes: The costs do not include value added tax, and are approximate in 2013.
Table 3.3
Specialized equipment needed for proximate analyses

Equipment
Analyses
Unit comprises
Costs (Euro)
Kjeldahl determination
Nitrogen
Distillation and titration unit
35 000
Fat determination
Fats and oils
Extraction apparatus, semiautomatic
20 000
Fibre determination
Neutral and acid detergent

fibre, lignin
Boiling and filtration apparatus,

semi-automatic
25 000
Dumas determination
Nitrogen (total combustion

method, alternative to Kjeldahl)
Auto-sampler, balance, Dumas

apparatus
45 000
ANKOM fibre
determination
Neutral and acid detergent fibre

(nylon bag method, alternative
fibre determination)
Analytical balance (0.1mg),

ANKOM apparatus, forced-air
drying oven, and muffle furnace
15 000
Bomb calorimeter
Gross energy
Balance, bomb calorimeter
50 000
Notes: The costs do not include value added tax, and are approximate in 2013.
Table 3.4
Dedicated equipment required for instrumental analyses

Equipment
Analysis
Costs (Euro)
Atomic Absorption Spectroscope (AAS)
Minerals
35 000
Inductively Coupled Plasma Analysis (ICP)
Minerals including P
50 000
Gas Chromatograph-Flame Ionization Detector (GC-FID)
Fatty acids
30 000
Gas Chromatograph-Mass Spectrometer (GC-MS)
Pesticides
50 000
High Performance Liquid Chromatograph (HPLC)
Amino acids, vitamins
30 000
Notes: The costs do not include value added tax and are approximate in 2013.
In some cases there may be alternative solutions. Thus for GC work the gas supply has
traditionally been provided by gas cylinders purchased from a bulk gas supplier, but it may
be more economical to install a hydrogen generator, which then requires water and power.
Selection of equipment needs to be based on set criteria, including technical specifications,
26
such as detector, wavelength range, centrifuge rpm (g value), price, etc., and also the level
of post-purchase service available. Service contracts are available and should be investigated thoroughly prior to purchase. Always ensure that a full service manual is provided
(including electrical diagrams). For optimum performance, regular maintenance is critical,
provided either by company staff or external entities.
3.2.5Organizational structure and responsibilities of personnel

For the laboratory, the number of personnel and their educational and experience levels
depends on the analyses to be offered, the methods chosen and the expected sample
throughput. The first step is to create an organized structure for the laboratory and to
define the activities to take place in the laboratory, as illustrated in Figure3.4.
Technical staff may also be referred to as Technicians, Analysts or Scientists.
The horizontal division of the laboratory into different sections depends on the type of
analyses it intends to perform. Similar types of analyses, such as proximate or chromatography (GC, HPLC) are grouped into one section. Ideally, a senior technician will be responsible for each area, such as proximate analyses, with 2 or 3 technicians to rotate around
the various analyses within their section. This enables an overall knowledge to be gained
of the working requirements for each area of the laboratory. It is important to keep the job
interesting and challenging for staff and to avoid monotonous, repetitive work where ever
possible. The vertical division of the laboratory reflects the different positions and responsibilities within the organization. As a guide, some typical positions can be identified:
The Laboratory Manager is responsible for the whole laboratory and the development
of its strategic plan. A key part of this position is the external communication with
clients and potential clients, as well as full responsibility for results reported to clients.
Management systems must be put in place to ensure reliable data are produced and
that the reporting of this data is thoroughly checked prior to releasing reports.
The Quality Assurance (QA) Manager is responsible for quality assurance within the
laboratory, and should have an independent position. The QA Manager may also have
responsibility for Health and Safety Management, or Environmental Management
figure 3.4
An example of a laboratory organizational structure

Laboratory Manager
Administration
Sample preparation
Junior laboratory
technicians
QA Manager
Senior Technician - proximate

analyses
Senior Technician specialized analyses
Senior and junior laboratory

technicians
Senior and junior

laboratory technicians
within the laboratory, or both, or they may be the responsibility of a separate Health
and Safety Manager and an Environmental Manager.
The Section Head or Senior Technician (Proximate analyses) is responsible for the
daily organization of the analytical process, ensuring that daily and weekly deadlines
within their section are met; quality control for each batch of testing meets requirements and is recorded; staff training is up-to-date; and that there are sufficient staff
to meet the workload requirements. Maintaining stocks of the necessary chemicals
and consumables are also the responsibility of the senior technician, who should
inform the Laboratory Manager in sufficient time to enable ordering and delivery prior
to stocks running low.
The Section Head or Senior Technician (Specialized analyses) is responsible for specific
equipment and methods, especially trouble-shooting, maintenance and solving problems, as well as continuing training of junior staff when required. Training records for
staff should be regularly maintained.
Junior technician(s) are responsible for performing analytical work following Standard Operating Procedures (SOPs), under the direction of the Section Head or Senior
Technician.
Note: Depending on the workload, one senior technician could be responsible for both
Proximate and Specialized analyses.
This organization reflects an ideal situation in a mature laboratory. In a new laboratory, however, the structure may initially be quite different. From the outset, the laboratory
should have all expertise needed to perform all the methods it offers. In practice this means
recruiting senior technicians with the background needed for the methods (see Section
3.2.1). This group should be regarded as the backbone of the laboratory, that trains additional personnel in case of an increasing volume of work and as a back-up for each assay.
Ideally, over time, the structure should become similar to that discussed above.
If the laboratory is part of a larger organization, such as a feed producer or a research
organization, its position and relationship with the other units should also be clearly
described within its structure. The laboratory has a responsibility to ensure that the quality
and credibility of its results are of the utmost quality. In an accredited laboratory this would
require an independent QA Manager.
3.3Realization of the laboratory Procedures

Besides the physical completion of the laboratory, procedures such as SOPs, quality control
programme, participation in proficiency (both internal and external) programmes and use
of reference materials need to be put in place to ensure the laboratory consistently produces results to a high standard. These procedures should guarantee that all aspects of the
analytical process are performed efficiently and are traceable. This should be documented
in a set of SOPs that form the basis of the Quality Management System (QMS).
This section briefly describes the QMS procedures. The topic is considered in greater
detail in Chapter4.
Procedures should ensure that the laboratory can prioritize and organize its workload
and guarantee the quality of results produced. The ISO/IEC 17025:2005 standard can be
very helpful to identify which procedures should be prioritized. In the initial phase, however,
27
28
the focus should be on aspects that directly influence the quality of the results. These
procedures are:
Acceptance criteria for samples.
Sample preparation.
Description of methods, including validation of results.
Quality control (first line of control).
Maintenance and calibration records of equipment.
Job descriptions, including responsibilities and continuing competence of individual
technical staff.
Training records of technical staff, covering which methods they can perform, level
of training, whether they can perform a method independently or under supervision,
and their ability to train others, etc.
Traceability and storage of raw data.
Cleaning procedures for the laboratory.
The presence and implementation of these procedures from the initiation of the laboratory will positively affect the quality of the results, and can be used as a starting point for
the implementation of a comprehensive quality system.
3.4Continuity and improvement of the laboratory

Success of the new laboratory depends strongly on its ability to respond to new opportunities and challenges in the commercial market or within the research field. For this purpose,
the laboratory should focus on efficiency, expertise and innovation.
Efficiency is related to the volume of analytical work performed within a specific time
interval. Increased efficiency leads to a lower unit price, making the laboratory more
attractive for customers. Automation, such as the use of sophisticated equipment and
autosamplers instead of manual methods, can be useful tools for increasing efficiency
and sample throughput. Automation of support processes, such as recording of sample
information, raw data and results, creating working lists, production of reports, etc., can
strongly increase the efficiency of the laboratory by saving staff labour time and increasing
the volume of work processed. This efficiency gain can be realized with the implementation and use of a specially designed Laboratory Information Management Systems (LIMS).
The laboratory should be constantly striving to improve efficiency and work throughput
to ensure its continued success.
The second important factor for success is to provide a consistent high standard of
service. This will give customers confidence and ensure repeat business. Laboratory expertise is strongly related to the specific knowledge, skill and experience of the technicians,
which is often described as human capital, and this is directly linked to the quality of the
laboratory. The laboratory should develop a policy to develop and guarantee the continuity
of this knowledge by utilizing training programmes to ensure there is always a back-up
technician for every assay. It is essential that the laboratory make every effort to retain its
experienced staff by creating a positive work environment, with good working conditions,
fair rate of salary, clear career path, training opportunities, etc. This will promote accumulation of expertise, and consequently increase both the quality and also the market position
of the laboratory. The importance of human capital should not be underestimated.
The third important issue is innovation, or research and development (R&D) to improve
the quality and to broaden the scope of analytical possibilities. The first two issues mentioned, i.e. efficiency and expertise, are the basic conditions needed to build a successful
R&D policy, which will enable the laboratory to respond successfully to new opportunities
in the market. In general, a new method will only be developed and validated following
a definite request from a customer for a large number of samples, probably 100. The
laboratory must weigh the risks of developing new methods without any guaranteed
samples against developing new methods for specific requests. To reduce risk, the laboratory manager should keep themselves up-to-date with current market trends by attending
relevant conferences and seminars, as well as monitoring the latest published literature. In
this way, being a market leader for new analyses will give the laboratory a clear advantage
over competitors.
The key to a successful and long-lasting relationship between the laboratory and its
customers is confidence. Customers must have full confidence in the quality of the data
produced. The presence of a QMS, preferably accredited, enhances trust of the customers.
Another important issue in the relationship with the customer is to meet the required
deadlines and agreed prices (quotes should be given to prospective clients). To evaluate and
improve the relationship, the laboratory should have regular contact with customers and
engage in customer surveys to identify opportunities for improvement. This is an essential
part of a QMS (see Chapter 4).
Procedures need to be developed to deal with complaints in a professional manner, and
subsequently to solve them in an efficient and diplomatic way to ensure customer satisfaction with the overall laboratory service.
The level of communication, and also of additional services offered, gives an opportunity for a laboratory to stand out from its competitors. For example, a consultation service
for nutritional information for various animal feed requirements, or an explanation of
results, and personalized reports that might be on a fresh weight or dry matter basis as
determined by the clients specific requirements, can sometimes tip the balance in favour
of the laboratory.
29
31
Chapter 4
Implementation of a Quality
Management System and the
road to accreditation
4.1Introduction
Right from the start, the focus of the laboratory should be the implementation of rules
and procedures to guarantee the quality of the results produced. A QMS is the total set
of rules and procedures that enables the laboratory to assure its quality. For this purpose,
the system should cover each aspect of the laboratory that could influence either directly
or indirectly the integrity of the analytical results. The range of these aspects makes the
implementation complex, especially when starting a laboratory, and therefore international
standards, such as ISO/IEC 17025:2005, have been developed to describe all areas that
should be covered within the QMS. Full implementation of these standards ensures that
all relevant aspects are addressed and that the QMS can be accredited according to an
internationally recognized standard.
The first step in the process of implementing a QMS is to understand its principles and
the correct interpretation of the standard. This knowledge is necessary as the standards only
describe which aspects should be covered in the quality system, not how. Understanding
the principles enables the laboratory to translate the requirements of the standard into
procedures and rules. The next section (4.2) describes these principles and gives examples
of how the different issues can be organized within the laboratory. The subsequent section
(4.3) deals with the content of ISO 17025 in detail by discussing all aspects described and
connecting them to the issues in Section4.2. A road map for the total implementation of
a QMS is the subject of the final section, 4.4. The sequence of implementation not only
reflects the relative importance of the various issues, but also the requirements to prove
that the system is operational and approved, which is essential for accreditation.
4.2Basic principles of quality

Aspects that affect the quality in the laboratory can be categorized on three different
levels. Firstly, the technical level contains all aspects that directly influence the quality
of the analytical process. Secondly, the organization level contains all relevant aspects
within the organization itself that indirectly influence the quality of the analytical process.
Finally, the commercial level contains all relevant aspects of the interaction between the
laboratory and its customers. These three levels will be described in detail in the following
sections.
32
figure 4.1
Factors influencing the quality of the analytical results

1. Sample
2. Method
3. Personnel
QUALITY OF THE ANALYTICAL RESULT
4. Equipment
5. Consumables and
chemicals
6. Quality control
procedures
4.2.1Technical level
The technical level contains elements that directly influence the quality of the analytical
results produced (Figure4.1), and the six parameters discussed in the subsequent sections.
4.2.1.1Sample
The physical condition of the sample material should allow a reliable performance of all
determinations requested. For this purpose, the laboratory should set up procedure(s) to
establish the acceptance criteria and preparation of individual sample types. For acceptance,
the physical state of the samples status on arrival should be noted (i.e. temperature, frozen,
partially thawed, etc.), and also the form in which the sample needs to be for each assay,
whether freeze dried, defatted, fresh, etc. Other factors, such as the minimum amount of
material required for each test, should be listed and made available to customers (see Section 4.2.3). The laboratory should ensure that the test portion used for the determination
is representative of the total sample provided. For animal feeds this is generally achieved
by prior drying (oven or freeze drying) and grinding, leading to fine, homogeneous material that allows a representative sample to be taken. Products with a high fat content may
need a different type of grinder (not forced through a mesh) or extraction of the fat prior
to grinding. International standards for general sample preparation are available (ISO 6498
Animal feeding stuff Preparation of test samples), with additional requirements described
in standards for specific determinations.
It is the laboratorys responsibility to prove that their sample preparation procedure and
storage lead to reliable results within acceptable variation limits. This proof can be part of
the validation studies for specific determinations (see 4.2.1.2 Methods, below).
4.2.1.2Methods
The laboratory should base its selection of a specific method on the requirements of the customer, the technical capabilities of the laboratory (see Chapter 3), the availability of verified
and referenced methods (ISO, AOAC International, etc.) and, very importantly, sample type.
In the case of animal feedstuffs, the methods are often based on international standards
(see Section 3.2.1), which means the laboratory does not have to prove the correctness of
its methodological principle. If the laboratory uses an in-house developed method, or modification of an accepted, verified method, it must validate the method to demonstrate that
it is fit for purpose. To validate the method, a robust SOP must be written and a validation
Implementation of a Quality Management System and the road to accreditation
study performed. The validation study must demonstrate the robust nature of the method
and consistency of results obtained using verified reference material.
The analytical protocol (SOP) describes exactly how to conduct the method and contains
all vital information about its determination. This information should emphasize all critical
steps, which reflects the knowledge and expertise within the laboratory, and data about the
quality of the determination. To demonstrate the traceability and repeatability of the analytical results, it is critical that the document reflects actual practice, using the equipment
and resources available in the laboratory. The format of the SOP should be comparable to
that of international standards, dividing the SOP into different sections that describe principles, scope, limit of detection, repeatability, chemicals, equipment, procedure, calculation
and interpretation of results. The laboratory should have procedures to enable the creation
and maintenance of SOPs (see Section 4.2.2, document control). The SOP should contain
information relating to any health and safety issues and environmental considerations as
appropriate.
The general aim of a validation study is to prove that the principle of the method is
correct and the quality of results is based on accuracy and precision. The protocol of a
validation study, however, varies between methods and depends on its purpose and scope,
so understanding the principle of the method is vital. Validation is achieved by determining
the following parameters delineating the quality of the method:
Limit of detection (LOD) and Limit of quantification (LOQ) are the lowest concentration that can be identified and measured, respectively.
Accuracy describes the difference between the result found by the laboratory and
the true value in a sample.
Precision describes the variation in the results found by the laboratory in the same
sample. Precision can be estimated at the same time and under the same conditions,
which is repeatability, and at different times and conditions, which is intra-laboratory
reproducibility.
Linearity describes the upper limit for which the concentration shows a linear relationship with the measured signal. This parameter is only relevant with a calibration
curve.
Selectivity describes the influence of other components on the measured signal.
Sensitivity describes the quantitative relationship between the analyte and the measured signal.
Robustness describes the effect of variation in the procedure on the measured
signal.
Stability describes the change of concentration of an analyte over time, stored at
specific conditions, such as temperature and pressure.
Calculation of these parameters requires analytical and statistical knowledge, and
depends on the type of methodology utilized. In AppendixA, examples of these calculations are given for different types of methods.
A validation study does not have to include all of the above parameters (see
AppendixC). The determination of selectivity and robustness is generally limited to a newly
developed in-house method or if modifying a standard method, whereas the determination
of linearity is only relevant for spectroscopy and chromatographic determinations. LOD,
33
34
accuracy and precision should always be part of a validation study. In general, the laboratory should estimate these parameters and record these values in its analytical protocol. If
the laboratory, however, claims to work according to a standard method, it should prove
that its values are at least comparable to those stated in the official method. The analytical
protocol should contain the values mentioned in the official method.
Although there can be some interaction between the performance of the validation
study and the analytical protocol, it is important to emphasize that the determination of
parameters such as LOD, accuracy, and precision are based on the final analytical protocol.
If changes are made to the protocol, the laboratory should investigate the effects on these
parameters by conducting an additional validation study. Results for the validation study
should be available on request and should be the basis for calculating the measurement of
uncertainty (see AppendixD).
4.2.1.3Personnel
Knowledge and skills of laboratory technical staff affect the quality of the results produced.
To guarantee this aspect, the laboratory should be confident that the personnel involved
are capable of conducting the methods correctly. The laboratory should have a procedure
for training and authorization of their personnel for each determination. This should be
described in the training and authorization records for each technician (see AppendixG).
This document could contain a check-list with the information required to conduct the
method and on which results the authorization is based. Only staff with training authority
can train others.
An authorization matrix, as given in Table4.1, is a useful tool to provide an overview
of training status. From the example it can be seen that only one person is authorized to
conduct the analysis method D. Ideally, there should be at least two technical staff capable
of conducting each of the laboratorys standard methods.
Temporary personnel should also be trained and authorized for each determination.
Training records can be divided into stages of the method, e.g. digestion only or
spectrometry, as well as status of training level, e.g. proficient, or capable under supervision.
On-going competency must also be demonstrated in training files. This may use participation in external proficiency schemes, Ring Trials, inter-laboratory comparisons, etc.
Should a member of staff fall below the requirements of on-going competency they must
stop performing the analysis until competency is regained.
Table 4.1
An example authorization matrix

Determination
Technical staff
member
P
P
P
P
P
2
3
4
Notes: P = authorized to perform
4.2.1.4Equipment
The equipment affects the quality of the analytical results produced. In general, the specifications of more sophisticated equipment improve the quality parameters, such as a lower
LOD and better precision. These effects are described in the analytical protocol and the
laboratory should ensure that equipment is working according to these specifications. For
this purpose the laboratory should focus on regular maintenance and performance checks.
The aim of maintenance is to avoid future problems with equipment, which could lead
to unreliable results. Maintenance of equipment can be divided into regular checks done
by laboratory technical staff, and more sophisticated servicing performed by external specialist contractors. The frequency of both types of maintenance depends on the type of
equipment. For standard equipment, maintenance by laboratory technical staff is generally
sufficient, with external contractors necessary only in the event of a major malfunction.
For more sophisticated equipment, such as chromatography equipment, high-speed centrifuges, etc., regular servicing by accredited staff from the supplier or a specialist contractor
is essential to guarantee the continued performance of the apparatus.
Performance checks are conducted to prove that the equipment is meeting the
required specifications. Calibration of volumetric equipment, control of balances, and
estimation of the wavelength in a spectrophotometer, are examples of these kinds of
checks, which should be described in protocols containing information about the procedure, frequency of checks, and criteria. The performance check procedures can be
divided into method-dependent and method-independent procedures. Method-independent checks of equipment are undertaken without conducting a specific analytical
method, but rather by measuring general physical properties such as weight, temperature and volume that can be traced back to internationally accepted references. The
checks for most analytical equipment, however, can only be undertaken by conducting a
method, and should focus on issues such as sensitivity (minimum response for a calibration solution) or retention behaviour (retention time for a specific compound) in the case
of chromatographic assays.
For all main pieces of equipment, the laboratory should have two documents: an
operational manual (or User Guide) and a logbook. The operational manual will contain all
relevant information (such as protocols and frequency) about maintenance and performance checks, and information on the safe use and operation of the item of equipment.
The logbook is used to record all maintenance, problems and results of checks (oil change,
breakdowns, etc.) performed for the apparatus involved, and also to record the extent of
use of the equipment and names of the users. The presence of these documents and a
properly filled-out logbook demonstrates that the equipment is working according to the
specifications and is capable of producing reliable results. All critical equipment should be
labelled with a unique code, and a spreadsheet or list should be used to ensure that all
maintenance and checks are performed according to a fixed schedule.
4.2.1.5Consumables and chemicals
Impurities in and contamination of consumables and chemicals can negatively affect the
quality of a determination, and therefore the laboratory must ensure their quality. Critical
chemicals and consumables should be explicitly described in the analytical protocol of the
35
36
determination, including criteria such as HPLC-grade only, containing <0.01ppm lead,

etc. Certificates for purity should be obtained from suppliers. If using commercially available stock solutions, such as for minerals and amino acids, the laboratory should verify that
suppliers of the reference materials are ISO 34:2009 certified. If not, critical levels should
be checked against an independent reference standard.
4.2.1.6Control of the results
The previous aspects focused on the creation of optimal conditions to produce reliable
results. The day-to-day routine, however, is also influenced by random variation introduced
during execution of a procedure, introducing an error. Therefore the laboratory should
constantly check the quality of the results produced by the implementation of a first line
of control procedure. The basic principle is to analyse a control standard within each batch
of samples and use its result to judge the quality of the other results in the batch. The
matrix of the control standard should be representative of the samples generally analysed
by the laboratory. The most important demands for these standards are stability and
homogeneity, which should be guaranteed by the laboratory. The laboratory should have
criteria to judge the results of the control standard and a method to record these values.
In practice, Shewhart-charts are most commonly used to record these results, including
acceptance criteria and relevant statistical information (see AppendixB). Shewhart-charts
are very powerful tools for a continuous check on the quality of the results, and also reflect
changes in the performance of the method over time. The first line of control, however,
only recognizes problems at batch level and is therefore not guaranteed to avoid errors in
individual samples. For this purpose, it is advisable to conduct analyses in duplicate in the
same (repeatable conditions) or different batches (intra-laboratory reproducible conditions).
Analysing samples in different batches is statistically the best, but not always the most efficient approach. Quality of the results should also be monitored using set criteria for error
percentages allowable between duplicates for each sample type.
Participation in collaborative trials (proficiency testing) and the analysis of Certified
Reference Material (CRM) give valuable information about the accuracy of the results of
the laboratory. The results found in collaborative trials are mostly evaluated by a z-score,
which is calculated as the difference between the found value and the average divided by
the overall standard deviation, and should be between -2 and +2. The z-score should be
used to identify the presence of systematic errors, and the determination of the uncertainty
of measurement. For the evaluation of the z-scores, the laboratory should be aware of the
effects of differences among the methods used by other participants.
Two points need to be noted when referring to the estimation of accuracy when using
the results obtained from collaborative trials or from analysing a CRM. Firstly, the average
obtained for a collaborative trial is based on a limited number of laboratories and methods;
it should therefore be regarded as a compromise instead of the true value. The advantage
of certificated materials is that its results are generally based on a range of methods, which
decreases the influence of systematic errors and therefore can more closely represent the
true value. Secondly, the results obtained for most proximate analyses are empirically connected to a specific method. An example is crude protein, which is for most animal feedstuffs by definition N 6.25, whereas the true protein in a sample can systematically deviate
37
figure 4.2
Organizational aspects affecting the quality of analytical work

1. Structure and
Management
2. Facilities
3. Procedures and
Traceability
7. Internal
relationship
HIGH QUALITY ANALYTICAL WORK
4. Human Resource
Management
5. QA services
6. Document
control
from this value. It is important to realize that, for these determinations, accuracy is based
more on convention than on the true chemical composition.
4.2.2Organization level
The analytical work is fundamental within an organization to facilitate its conduct and continuity. Different aspects of the organization are linked to the analytical process and therefore
indirectly influence the quality of results produced (see Figure4.2). These are more general
aspects for organizations rather than being laboratory specific, and can therefore also be
found in ISO 9001:2008 Quality Management Systems Requirements. If the laboratory is
part of a larger organization, it might use existing procedures for these aspects, otherwise it
is required to develop and implement these issues by itself (see Section4.4).
These aspects are described below in more detail.
4.2.2.1Structure and management
Work in a laboratory relies on teamwork, which should be organized in a manner suitable to avoid the production of incorrect results. A description of responsibilities relating
to the validation of methods and authorization of results should be clearly described and
conducted only by skilled laboratory personnel. To guarantee the continuity of an analytical
process, the head of each section should nominate a deputy that can take responsibility for
the work in their absence.
The QA Manager in the laboratory has a special position because of their responsibility for monitoring operational quality within the laboratory. This person should have an
independent position, with direct access to the upper management of the organization.
Sometimes the QA Manager may be part of a separate QA unit (see Section4.2.2.5 below)
to guarantee independence.
4.2.2.2Facilities
Basic and specific facilities are needed to conduct laboratory work and should be located at
a permanent location, as described earlier, in Chapter 3. The laboratory is obliged to prove
38
that the status of these facilities does not negatively affect the quality of its work by isolating
any interfering activities (such as keeping sample grinding separate from analytical work)
and by following a daily cleaning programme to minimize contamination or sample errors.
Specialized equipment will generally require service and maintenance from external
services. The facilities that house these items of equipment should be controlled by the
laboratory to ensure continuity of operation, such as required temperature, uninterrupted
power supply, dust free environment, etc.
4.2.2.3Traceability and procedures
Traceability of all activities in the laboratory is necessary to guarantee the quality of each
individual result. For traceability, it is essential that steps within an analytical process are
clearly described in the appropriate SOP, and that their correct performance can be demonstrated (traceability).
The analytical process should be described in a set of documents that covers general
and more detailed procedures, as described in Section3.3. These procedures should be as
explicit as possible to avoid variation in performance.
To prove correct performance, all vital information must be recorded and stored. An
example of this is a samples worksheet, which as a minimum should include:
Date of analysis
Condition of sample on arrival (frozen, fresh, dry (oven dried or freeze-dried), etc.)
Information about the control samples analysed
All raw data produced, including spectrometry printouts, weighing sheets, etc.
Name of laboratory technician for each analysis conducted
Identification of equipment used (e.g. ID number of balance, centrifuge, etc.)
Signature for acceptance and date the report is sent to client
Any communication from the client
All critical factors should be traceable from the information on the worksheet. If necessary, the laboratory can prove that the technician was authorized to conduct this determination, the equipment had the correct calibration status, calculation of results was correct,
samples were connected to the first line of control, and the results were authorized by
qualified personnel. It is essential that all of this information is available and stored correctly for a specified period by the laboratory. The storage time for information depends on
legal and customer requirements, but, in general, a storage period of 5 years is acceptable.
At times procedures may deviate from the standard protocol. This may be at the clients request or due to insufficient sample amount. The laboratory should implement
a non-standard work procedure that gives flexibility to deal with this kind of situation.
The key elements of this procedure are authorization and recording that a non-standard
method was used due to client specifications, with a record of any deviations from the
standard method. The client must be notified personally, and additionally in the laboratory report it may be necessary to disclaim any accreditation for the test concerned. In
general, the laboratory should have a restrictive policy on such issues, and their decision
about acceptance should be mainly based on the effect of the deviation on the quality
of the results. In the case of non-standard work, traceability will become an even more
important issue.
4.2.2.4Human resource management

The knowledge and skill of the laboratory personnel strongly affects the quality of the laboratorys results, as mentioned in Section3.4. The technical skills of the personnel should be
recorded in the training and authorization records as previously mentioned. At this level,
the focus is more on general aspects of this policy, such as the development and motivation
of personnel. The aim of this policy should be to guarantee the continuity of the laboratory
and encourage continual improvement.
In larger organizations, these activities are concentrated in a special unit that deals with
personnel affairs in general and is also responsible for archiving all relevant documents,
such as training records. For stand-alone laboratories, the management is generally responsible for this aspect.
4.2.2.5Quality Assurance services
The QA service plays a vital role in the process of implementation and maintenance of the
quality system within organizations and laboratories. In a larger organization, these activities can be organized in a separate unit. To guarantee its independent position, direct access
to upper management should be a priority. The first task of this unit is the development of
a quality policy and the implementation of the basic principles of quality within the organization, mostly achieved by the implementation of ISO 9001:2008. Additionally, this unit
should also facilitate the development and implementation of specific procedures within
individual units. In the case of laboratories, the QA unit should help with the interpretation and implementation of ISO/IEC 17025:2005, leading to setting up of laboratory SOPs.
The second task of the QA unit is to check the operational quality system and facilitate its
continual improvement by conducting internal audits that check how the system works in
practice, and identifying where improvements could be made. The unit can also perform
external audits of product suppliers, sub-contractors or services used by the organization.
The great advantage of centralization of QA activities is accumulation of knowledge and
skill in this unit, which is beneficial for the whole organization as well as training internal
auditors. Success of the quality policy, however, also depends on the relationship between
the QA unit and the laboratory itself.
A separate QA unit is usually absent in most small stand-alone laboratories. In this situation, these activities should be conducted by the QA Manager in the laboratory itself or
employ external expertise to create and implement the basic QMS, including conducting
internal audits.
4.2.2.6Document control
The quality control system requires a great number of controlled official documents to
describe all aspects of the process. These documents have to be updated regularly and
the organization must ensure that only the most recent version is available and in use by
personnel. For this purpose, the organization should have a procedure for document control that regulates administration, distribution and revision, to ensure that only the correct
versions of documents are in use.
In larger organizations, document control is mostly centralized and conducted by the
QA unit. Sophisticated electronic systems have been specially developed to facilitate the
39
40
total process of document control. In the case of smaller laboratories, document control
can also be regulated by the use of a database system.
4.2.2.7Internal relationships
If part of a larger organization, the laboratory will have acquaintances and interactions with
other internal units, which could affect the quality of results. An example of a misuse of this
interaction is an internal unit giving short deadlines to the laboratory to complete analyses.
The organization has the responsibility to identify such influencing factors and ensure that
they do not negatively affect quality. An organizational structure that ensures the laboratory has an independent position can help to avoid these issues.
Another example is the relationship between the laboratory and upper management.
The implementation and maintenance of a QMS involves various investments that should
be approved by upper management, otherwise it will risk functioning under sub-optimal
conditions. To guarantee that the laboratory gets the resources required, commitment from
upper management is essential.
4.2.3Commercial level
The commercial level is the relationship between the laboratory and customers, which can
affect the quality of analytical results. This subject is a relatively new element in QMS and
its focus is on the perceived value of the result for the customer as a decision-maker. This
value depends not only on technical qualifications but also on aspects such as time and
price. Customers have expectations on these issues and therefore laboratories are obliged
to be realistic and ensure the customer is aware of constraints such as turn-around time,
price and methodology.
These aspects will be described based on the three phases of communication between
the laboratory and the customer.
4.2.3.1Acquisition phase
The acquisition phase starts with the first contact between both parties, and ends with the
acceptance of the samples by the laboratory. The aim of this phase is to define the customers expectations and laboratory services required. General information about the laboratory services, such as methods, quality, prices, criteria for acceptance, turn-around time
and general conditions should be freely accessible for customers (for example on a Web
site or as part of a quotation). If standard laboratory services are required, this information
will generally be sufficient for the customer to make an informed decision.
It is advisable that laboratories have their own submission document that can be used
by customers, including acceptance criteria for standard services, and test options. The use
of this document can avoid confusion and disappointment regarding expectations of the
services available. This submission document also forms a contract between the client and
laboratory.
For special work outside the routine, it is advisable to have direct contact between both
parties that should lead to a contract that specifies the work and the expectations (methods
used, delivery time and price). Outsourcing of tests should also be explicitly described in this
contract. These contracts should be signed by both parties and stored by the laboratory.
4.2.3.2Operational phase
The operational phase starts after the acceptance of the samples, and ends with sending the
final report to the customer. In the case of problems or deviations from the standard procedures,
the laboratory is obliged to inform the customer and to discuss the significance of the problem
and other options available. These deviations from the standard protocol should be treated as
a non-standard procedure and conducted accordingly (see traceability, Section 4.2.2.3). In this
situation the laboratory remains responsible for the quality of the results and therefore should
make a judgement between its commercial interests and the quality of the results. The laboratory should retain all correspondence with the customer regarding these issues.
4.2.3.3After-sales phase
This phase starts when the customer receives the final report. The report should give a clear
overview of all essential information about the sample, such as the identification code,
sample description, type of analyses conducted and the results, including the relevant units.
Accredited laboratories should also note which determinations were conducted inside the
scope of the accreditation and which were conducted outside the scope of accreditation.
Any tests that were sub-contracted should also be clearly indicated.
The laboratory should deal with customers queries about the results and interpretation
in a professional manner. Communication with customers should be limited to laboratory
management to ensure a consistent policy in answering questions. Direct communication
between customers and technical staff should be avoided.
On completion of the analytical work, storage of data and sample material should be
organized. General standards for the storage of samples (see technical level, sample Section4.2.1.1) and of data (see organization level, traceability Section4.2.2.3) should be
documented. Customers should only have access to data from their own samples to guarantee confidentiality. After a fixed interval (e.g. 1 calendar month post-reporting), samples
should be destroyed or disposed of in a suitable manner.
Two additional important issues in the communication between the laboratory and the
client are complaints and feedback. The laboratory is obliged to have a procedure for dealing
with customer complaints. These complaints are useful to identify opportunities for improvement and to help improve the quality and level of service to the customer. Taking complaints
seriously will also increase customer confidence. The laboratory has a responsibility to seek
feedback from customers regarding its performance. A questionnaire is a good tool to obtain
this information, and is also useful to optimize processes in the laboratory.
4.3Reading and interpretation of ISO/IEC 17025:2005

This section describes the different aspects mentioned in the standard and makes a connection with the general principle of the quality system as described in Section 4.2. This
information can help with understanding and therefore assist with implementation of the
different elements of the standard.
The most important sections of the standard are in its Section 4, Management requirements, dealing with general demands regarding the organization and quality system at
the organization level, and in its Section 5, Technical requirements, which is more specific
for laboratories. In Boxes 1 and 2 the present authors focus on important items in ISO/IEC
41
42
Box 1
cating its importance and ensuring integrity
Important sections from Section 4 of

Standard ISO/IEC 17025:2005
if changes are made. This active involvement

should prove that quality is an essential part of
company policy and that the resources required
will be made available.
4.Management requirements
This commitment should be in the form of
4.1Organization
a written, signed document that contains a
This addresses the Organization level, its struc-
statement about the importance of the quality
ture and management.
system for the organization as a whole.
General requirements are stated, such as

legal status, extent of management system
4.3Document control
responsibility, capability to perform testing activ-
This addresses document control at the Organ-
ities according to the ISO Standard, quality sys-
ization level other than control of data related
tem and customer requirements, as well as the
to testing and calibration, which is covered
provision of adequate facilities. Special attention
elsewhere.
is paid to the problem of improper influence due
Procedures are required to control all doc-
to conflict of interests, especially where the labo-
uments, in the broadest sense, that are part of
ratory is part of a larger organization.
the management system, including manuals and
The laboratory should have an independ-
software. All documents produced should be
ent position and its structure, responsibility,
approved by authorized personnel, and a master
authority and interrelationships should be fully
list should identify the current version and distri-
defined. Personnel should be familiar with the
bution to preclude the use of invalid or obsolete
procedures and the relevance of the quality
procedures. The documents should be uniquely
system. The laboratory should have a quality
coded and periodically updated. Changes in
manager and appoint deputies for key mana-
documents should be marked in the updated
gerial personnel.
version and be approved by authorized personnel. The standard offers the opportunity to
4.2Management system
make minor changes by hand on the condition
This addresses QA services and internal relation-
that they are signed and dated and a revised
ships at the Organization level.
document is re-issued as soon as is practicable.
The laboratory should have an implemented

management system that covers all relevant
4.4Review of requests, tenders and contracts
aspects to guarantee the quality of the services
This addresses QA services at the Organization
offered. An overall Quality Manual should give
level.
an overview of all aspects of the management
The laboratory should administer requests,
system and refer to more detailed information,
tenders and contracts. This also covers work
with an outline of the structure of the docu-
that is subcontracted to other laboratories.
mentation. Upper management should play an
The standard however distinguishes between
active role in the creation, improvement and
repetitive routine work and more complex
maintenance of the quality system and should
and advanced tasks. In the former case, the
provide evidence of commitment, communi-
laboratory can refer to a general agreement
with the customer. The standard also demands

communication with the customer if deviations
or amendments are required.
43
tory performance.
This transparency must not interfere with
the confidentiality of other customers. Therefore customers should only be able to access
4.5Subcontracting of tasks
their own results and not those of other clients.
This addresses the Commercial level at the
In the case of an inspection, the laboratory
acquisition phase.
should only make available relevant reports
For subcontracting, the work should be
and results.
placed with a competent subcontractor that
The laboratory should seek feedback from
works according to the ISO standard. The labo-
its customers that can be used to improve the
ratory remains responsible for this work unless
management system.
the customer specifically requests a particular

subcontractor. The laboratory should have a
4.8Complaints
list of suitable subcontractors who meet their
This addresses the Commercial level in the after-
criteria.
sales phase.
The use of subcontractors is a delicate issue
A policy and procedure for the resolution of
because the laboratory remains responsible.
customer complaints should be available. Records
Good knowledge of the quality of other labora-
about the complaints and the resulting investiga-
tories is essential to avoid subsequent problems.
tions and corrective actions should be kept.
In the absence of this knowledge, it is prudent to

avoid subcontracting, and rather allow customers
4.9Control of nonconforming testing and/or
to deal directly with other laboratories.
calibration work
This addresses the Organization level regarding
4.6Purchasing services and supplies
traceability and procedures, and also applies to
the operational phase of the Commercial level.
level.
The laboratory should have a procedure to
A policy should be available for purchasing
be followed if a deviation from its own proce-
services and supplies that can affect the quality
dures or from the original agreement with the
of the tests (pH buffers, analytical standards,
customer is required. Examples of these devi-
etc.). These products should be stored in the
ations are technical problems within an assay,
correct manner, and before use checked against
which might lead to less accuracy, or where
the defined requirements. Records regarding
internal procedures have not been implement-
this testing should be kept. A list and an evalu-
ed. To deal with these situations correctly, the
ation of suppliers of critical consumables should
Standard demands that this procedure should
be available.
contain the following elements:

Responsibility and authority for the
4.7Service to the customer
management
This addresses the Commercial level in the after-
work and how to identify these, as
sales phase.
well as the response actions required.
The laboratory should cooperate with customers to clarify any requests regarding labora-
of
non-conforming
Evaluation of the significance of the

changes (impact assessment).
(count.)
44
Corrections should be made, and customer notified.

Responsibility for authorizing the resumption of work should be defined.
The whole process should be documented,

including all results obtained by the corrective
actions. This procedure should also be followed
when dealing with deviations found during
If a non-conformity leads to doubts about the
internal and external audits. It should focus
quality, corrective actions should be undertaken
on the impact of the problem, looking for a
(see 4.11).
solution and to guarantee that the problem is

solved.
4.10Improvement
4.12Preventive action
level.
The laboratory has an obligation to contin-
level.
uously improve the effectiveness of its manage-
A policy is needed to identify potential
ment system through the use of quality policy,
sources of nonconformities in the manage-
audit results and preventive and corrective
ment system or technical areas. After identifi-
actions. The development of a quality system
cation, improvements or preventive actions are
is a dynamic process that should constantly
required that should be conducted according
be evaluated, developed and improved. The
to a procedure, this includes the initiation
demands of the quality of the system will
and application of controls to ensure they are
increase in time, which gives the new laboratory
effective. An example of a preventive action
the opportunity to develop their system within
is the replacement of poorly functioning old
a certain time period.
equipment, which avoids the issue of future

tests being out of specification.
4.11Corrective actions
4.13Control of records
level.
This addresses methods at the Technical level
Corrective actions should be conducted for
and procedures and traceability at the Organi-
all non-conforming tests or if a departure from
zation level, both in general and for technical
the management system or technical operation
records.
has been identified (see 4.9). A procedure for
All relevant quality documents should be
corrective actions should contain the following
uniquely identified and stored in an appro-
information:
priate way by the laboratory to avoid dam-
Determination of the cause of the problem.

Identification of potential corrective
actions.
Determining the efficiency of the corrective actions.
Additional audit (see 4.14) to evaluate the
new situation.
Date and initials of the person(s) initiating
and following up on the corrective action.
age and unauthorized access or destruction.

Records should be held securely and in confidence (see 4.7). Technical records should contain
all relevant information to identify the factors,
including personnel involved, that could affect
uncertainty and enables measurements to be
repeated if necessary. The records should be
stored for a specified period. Changes in records
must be traceable and signed by the person
making them.
The presence and access to this information
The laboratory management should peri-
is essential for traceability and control of the
odically undertake a review of their QMS to
quality of the tests conducted by the laborato-
ensure its effectiveness and introduce chang-
ry, giving customers and auditors confidence in
es or improvements as needed. This review
the QMS.
should address aspects such as: suitability of the

existing policy and personnel; internal audits;
4.14Internal audits
corrective and preventive actions; assessment
conducted by external auditors or accreditation
level.
bodies; results from proficiency tests; changes
The standard demands the presence of a
to the type or amount of work; customer feed-
schedule of audits that covers all elements of
back; recommendations for improvement; and
the QMS and that should be conducted peri-
other relevant factors. Activities for any such
odically. The QA Manager is responsible for the
review should be performed within an agreed
organization of these audits, which should be
timeframe.
performed by trained and qualified personnel

that have, if possible, a position independent
from the activities being audited. The Standard
allows small, stand-alone laboratories to use its
internal personnel to conduct these audits, but
auditors must not audit their own work. If the
audit reveals doubts about the effectiveness
or validity of a test it should immediately be
followed up with a corrective action (see 4.11).
Records should be kept covering the area of
activity audited, together with the findings and
corrective actions. Follow-up audits should verify and record the effectiveness of any corrective
actions required.
Although the audits should cover all aspects,
the laboratory should set priorities for items and
establish the frequency for performing these. In
the beginning, technical audits have a higher priority and should be performed more frequently
than general audits. For smaller laboratories, it
could be wise to hire external expertise to conduct specific audits and train internal personnel.
The laboratory has an obligation to improve the
quality of audit records over time.
4.15Management reviews
This addresses internal relationships at the
Organization level.
45
46
Box 2
that could negatively affect the quality of the
Important sections from Section 5 of

Standard ISO/IEC 17025:2005
results. Incompatible activities should be separated by time or space to avoid cross-contamination, and access to all areas should be controlled. Good housekeeping, including appro-
5.Technical requirements
priate cleaning and monitoring procedures, is
5.1General
needed to guarantee the quality of work and
The laboratory should have the resources and
avoid contamination.
policies in place enabling it to perform the tests

in compliance with the stated methods and
5.4Test and calibration methods and method
therefore guarantee the quality of the results
validation
produced.
This addresses methods at the Technical level.
5.2Personnel
methods and procedures for all tests, including
This addresses the Technical level for personnel
aspects such as sampling and storage, with
and the Organization level for Human Resourc-
relevant statistical information plus, if neces-
es Management.
sary, any estimation of the measurement of
The laboratory should use appropriate
The laboratory should ensure the compe-
uncertainty. Instructions for the use of equip-
tency of its personnel to deal with specific
ment and other relevant documents should
equipment, perform tests and sign reports. This
be available for personnel at point of use. Any
competency should be based on education,
deviation to the method must be documented,
experience and skills. Supervision should be
technically justified, authorized, and accepted
provided during training. Management should
by the customer. The standard allows the use of
set up a programme for the education, training
international or nationally recognized method-
and continuing competency of personnel, and
ologies as protocols as long as they fully cover
programme effectiveness should be evaluated.
the procedure performed.
In the case of temporary personnel, supervi-
The selection of the methods should be
sion should be ensured by the laboratory. Job
based on the demands of the customer. How-
descriptions should be available for all staff
ever, it is preferable to use national or interna-
positions. Laboratory management is respon-
tionally recognized standard methods. The labo-
sible for the authorization of specific issues,
ratory should use the latest version unless there
which should be documented (including date of
are serious constraints. Laboratory-developed
authorization) in the training files.
methods (in-house methods) or modifications

to recognized methods may be used if validated
5.3Accommodation and environmental conditions
and the customer has been notified.

Validation of a method demonstrates that
This addresses facilities at the Organization
particular requirements for a specific intended
level.
use are fulfilled, and validations should also
The facilities should allow tests to be per-
be performed for non-standard and labora-
formed correctly with adequate monitoring
tory-developed methods as well as standard
and control of any environmental conditions
methods used outside their scope, such as a dif-
47
ferent sample type. The scale of the validation
achieve the accuracy required. Calibration pro-
depends on the needs of the application, but
grammes should be installed for key quantities
should always be a balance between cost and
or values that should be checked before use.
benefit. The range and accuracy of validated
Each item of equipment should be uniquely
methods should be determined, and be relevant
identified and only be operated by authorized
to customer needs. Data for validation studies
personnel. For each item of equipment that is
should be stored. The requirements for valida-
significant to the test, a record should be made
tion are described in section 4.2.1.2 of the main
that contains at least the following informa-
text and in AppendixC.
tion: identification; description of apparatus;
The laboratory should have a procedure to
checks of specification; location; manufacturers
estimate the uncertainty of measurement that
instructions; calibration records; maintenance
takes into account all relevant factors (i.e. dif-
plans; and any damage. The laboratory should
ferent technicians or operators, different equip-
have procedures for safe handling and mainte-
ment, etc.), using an approved statistical calcu-
nance of equipment to prevent contamination
lation. This calculation gives information about
or deterioration. Equipment that is outside
the uncertainty in an individual measurement
specified limits should be taken out of service
and is expressed as a 95%-confidence interval.
by removal or clearly labelling as out of use; it
This calculation takes into account random vari-
should only be used after repairing and testing
ation as well as systematic variation obtained for
(as in Section 4.9 of the Standard). If possible,
reference materials. This calculation is complex
the equipment should bear a label that indi-
and the person conducting it should possess
cates the status and expiry date of calibration.
appropriate statistical knowledge. The labora-
If the equipment is temporarily placed outside
tory can use ISO 5725:1994 for additional infor-
the laboratory, the laboratory is responsible for
mation about this issue. Appendix D gives some
checking its performance before being used
examples of this calculation.
again. Procedures should be available for inter-
Calculations and data transfers within the
mediate checks. Special attention should be
method should be checked in an approved way.
paid to the correct use of correction factors for
When using computers for this purpose, the
equipment.
laboratory should demonstrate that any software developed by the user is documented and
5.6Measurement traceability
validated, and that data is protected and the
This addresses equipment at the Technical level.
integrity and confidentiality of data transfer
All equipment that is critical should be
is guaranteed. The use of standard software,
calibrated before being put into service (see
such as spreadsheet programs, is allowed by
Standard section 5.5). Calibration of equipment
the Standard. Computers involved should be
should, if possible, be traceable to the interna-
systematically maintained.
tional system of Units (SI). For chemical laboratories, such calibration is limited to physical
5.5Equipment
parameters such as weight, volume and temper-
This addresses equipment at the Technical level.
ature. Otherwise the use of certified materials
The laboratory shall have the equipment
or participation in inter-laboratory comparisons
to enable correct performance of the tests to
should be implemented.
(count.)
48
Reference standards and materials should
5.8Handling of test and calibration items
be calibrated by a body that can provide trace-
Addresses the Technical level for the sample,
ability. Checks should be performed to main-
and the Organization level for traceability.
tain confidence in the status of these refer-
Procedures for transportation, handling,
ence standards, whilst their storage should
protection and storage of test items (i.e. sam-
be according to supplier recommendations to
ples) should protect their integrity and pro-
protect their integrity.
tect the interests of both the laboratory and

the customer. The laboratory should give each
5.7Sampling
sample a unique code or number that is used
This addresses the Technical level for the sam-
in all test reports; these codes might be auto-
ple, and the Commercial level for the sample
matically generated if a LIMS system is in place.
acquisition phase.
Deviations from the standard protocol should
The laboratory requires a sampling plan
be recorded in case there are doubts about
and procedures based on statistical principles
the suitability of an item for a test or if it does
when it carries out sampling of substances.
not conform to the description provided, and
This plan should be available at the location of
the laboratory should contact the customer
the sampling. Sampling is a defined procedure
before proceeding and record the discussion
whereby the part that is taken is representative
and agreement. The laboratory should also
of the whole, such as a batch of a diet. This
have procedures and facilities to store and
sampling should not be confused with taking
secure test items to guarantee their condition
a test portion from a sample in the laboratory.
and integrity. Storage conditions of laboratory
Deviations wanted by the customer should be
samples depend on the type of sample and the
recorded in detail and be included in all related
tests required, and should be both part of the
documents, including test reports (see Standard
validation study and they should also be stated
section 5.10) and communications to personnel.
in the appropriate methods. For storing sam-
All documents relating to the sampling strategy
ples, a general procedure should be available. In
and performance should be archived.
general, the laboratory should have the capabil-
In practice, sampling can lead to confusion and problems between customers and the
ity to store at -20 C, 28 C and +20 C, and in

exceptional cases also at -80 C.
laboratory. Two situations can be identified.

Firstly, the laboratory receives a sample from
5.9Assuring the quality of test and calibration
the customer and conducts the required tests.
results
In this case, the laboratory should state that
This addresses the Technical level for quality
the results are representative for the received
control procedures.
sample material by using a disclaimer (see
The laboratory should have quality control
Standard section 5.10). Secondly, the customer
procedures for monitoring the validity of tests
asks the laboratory to perform a test from a
undertaken. Data should be recorded in such
larger batch, which includes setting up a sam-
a way that trends are detectable and results
pling strategy. In this case, the laboratory is fully
can be statistically evaluated. This monitor-
responsible and accountable for the sampling.
ing should be planned and reviewed, and it
It is important to discuss both options with
might include regular use of certified reference
customers.
materials, participation in external proficiency
testing, replicate tests, re-calibrations and cor-
the test report should also contain all relevant
relation with different characteristics. If quality
information about this subject, such as the strat-
control data are displayed graphically, trends
egy and performance (see Standard section 5.7).
are quickly noticed, and if outside the pre-de-
Interpretation and opinions should be clearly
fined criteria action should be taken to identify
marked as such in the report.
and correct the problem and prevent incorrect

results being reported.
Electronic reports should also contain the

parameters described above. Amendments to
a test report should be made as a separate
5.10Reporting the results
document that refers to the original report and
This addresses the Commercial level in the oper-
should also meet the demands set.
ational phase.
Results should be reported accurately and
must include all information requested by the
customer and required for the interpretation of
the tests. The standard allows a simplified way
of reporting if this is previously agreed with customers. The test report should contain at least:
Title, name and address of the laboratory.
Unique identification of the report,
include numbering of pages.
Customer.
Identification of method.
Identification of the item tested.
Date the test item was received.
Test results, including units.
State of sample (as received; on dry-matter basis; fresh weight; etc.).
Name and signature of person(s) authorizing the test report.
When required, a statement that the results
relate to the items tested as received can be
made (see Standard section 5.7). Reports that
cover all the above-mentioned items are generally sufficient for routine analyses. Results that
are obtained from accredited tests should be
marked and communicated to the customers.
If additional information is needed for interpretation of the results, this should be added.
The report may also contain information about
deviations, statement of compliance, uncertainty, LOD, and opinions and interpretation. If the
laboratory was responsible for the sampling,
If an additional or supplementary report is

issued, this should be clearly identified as such.
49
50
17025:2005. The section numbers in these Boxes are the same as in the ISO/IEC Standard,
but are presented in italics. The text that is taken from the standard is in normal font and is
the simple description of what is stated in the corresponding section in the standard. The
text in italics is the additional related remarks. Both simple description and related remarks
are from the present authors, with the authors comments in italics.
4.4A road map for building a high quality system

The previous sections describe the principles and elements of the QMS (in Section 4.2) and
their linkages with the demands stated in Standard ISO/IEC 17025:2005 (Section 4.3 of this
document). For accreditation, the laboratory must ensure all aspects are adequately implemented. The aim of this section is to describe a possible road map for the implementation
of a QMS in four scenarios:
A stand-alone feed analysis laboratory.
A feed analysis laboratory in a large organization.
A feed analysis research laboratory.
A government or reference feed analysis laboratory.
Initially the accreditation of a stand-alone laboratory is described, and most of the steps
required for such a laboratory are also valid for the other three types of laboratory. Subsequently, specific aspects and deviations from the first type of laboratory that need to be
addressed for these three types of laboratories are then outlined.
Before starting, managers should realize that implementation is a continuous process
that takes at least 3 to 4 years. During this period the system should be developed and
improved. To make this process a success, a number of conditions need to be fulfilled:
Upper management commits to the activity, with full cognizance of the importance
of the QMS.
Management provides resources required for implementation of the process.
A realistic time schedule is prepared and implemented as per the plan.
Prioritization of quality issues crucial for the results, and related implementation.
Personnel are provided, especially technical staff, to support this process.
4.5First situation:
Routine stand-alone feed analysis laboratory
4.5.1Introduction
Standard ISO/IEC 17025:2005 is intended for routine laboratory work and is a perfect starting
point for putting in place a QMS. Stand-alone laboratories operate in an open commercial
situation and accreditation can be a powerful tool to gain customer confidence. This section
describes a possible road map that leads to accreditation after a period of 4years.
The first two years of this period should be spent on laying the foundations of a QMS
based on ISO/IEC 17025:2005 for the laboratory. The focus of the first year should be
on the technical aspects, while in the second year the focus is on organizational aspects.
The last two years should be devoted to implementing all aspects of the Standard, with
appropriate evaluation, with the aim to improve the QMS. A guideline outlining what to
do when is presented in Table4.2.
Implementation of these aspects will be described in more detail in the following sections.
51
Table 4.2
Timetable for the implementation of the individual aspects of a

QMS compliant with Standard ISO/IEC 17025:2005
Element
Start
4.1
Year 1
Year 2
X1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
X
X
4.14
4.15
5.1
X
X
5.2
5.3
5.4
X
X
5.5
X (5.4.6)
X
5.6
5.7
5.8
5.9
5.10
Years 3 & 4
Notes: Elements refer to the relevant sections in Standard ISO/IEC 17025:2005.

1Only Standard section 4.1.5(i).
4.5.2Initial phase
Before the laboratory starts conducting analyses on a commercial basis the method
description and validation (5.4), operational quality control (5.9), control of records
(4.13), and handling of test items (5.8) should be available and operational. These are
the minimum requirements to guarantee the quality of the results produced, and their
implementation should be the primary priority for the laboratory.
At this time, the laboratory can base its methods, including sample preparation, on
available documents, such as Quality assurance for animal feed analysis laboratories
(FAO, 2011) or on internal specifications. Initially the validation can be limited to the
determination of the LOD and LOQ, repeatability, and comparison with other laboratories. Operational quality control should be realized by obtaining standard sample(s) and
the implementation of the first line of control (see Section4.2.1.6 and AppendixB). All
raw data that is essential for the calculation of results should be stored properly (see
AppendixE).
52
4.5.3First year
After becoming operational, the laboratory should further focus on technical aspects,
such as facilities (5.3) and equipment (5.5). The laboratory should develop documentation
regarding maintenance and performance checks (see Section 4.2.1.4 and Appendix F).
Minimum requirements include the setting up of general procedures that guarantee the
functionality of the laboratory as a whole; availability of specific equipment; and ensuring
good housekeeping.
Other aspects the laboratory should implement are the reviews of requests (4.4), complaints (4.8) and control of non-conforming testing (4.9). These issues are related to its relationship with customers (commercial level) and will also help to increase their confidence
in the laboratory.
The final aspect that should be implemented is the management review (4.15). It is
important to involve upper management in this process right from the start, and to ensure
that they are fully committed to the QMS. This review, after the first year, will be quite brief,
and subsequently will become more comprehensive over time.
The laboratory should also initiate operational actions regarding the QMS during the
first year. On the technical level, it should periodically evaluate its performance of tests and
start participating in collaborative trials and proficiency tests. Experience shows that, especially in the initial period, the laboratory can benefit greatly from using this involvement and
the resulting information to improve the quality of its tests.
The laboratory should also appoint a QA Manager who will be responsible for the
implementation, evaluation, and improvement of the QMS (4.1.5(i)). For new laboratories,
this person is often one of the laboratory technical staff with an affinity for quality issues,
which is acceptable within the ISO/IEC 17025:2005 Standard. Experience, however, shows
that quality issues will become more time consuming as time goes on, possibly leading to
a fulltime position. This function also implies an independent organizational status, which
means that the person will be placed outside the laboratory group (see Figure3.4). During
the first year, resources should be spent on training of the QA Manager, unless this person
already has the requisite knowledge and expertise. The QA Manager may also have the role
of Health and Safety Manager, or Environmental Manager, or both, or these might be the
responsibility of another member of staff.
4.5.4Second year
The main issues in the second year are the setting up and implementation of the QMS for
the laboratory (organization level), and the evaluation and improvement of the quality of
its technical performance.
The laboratory staff should start writing a quality control manual that describes the
organization of the laboratory and its procedures. This manual is a general description of
the total system and its documentation, and refers to more detailed underlying procedures.
It should cover all aspects of the Standard (Standard section 4.2). The Standard gives the
laboratory freedom regarding the format of the manual. A simple approach is to format
the quality control manual based on the different sections of the Standard, and describe
how they are addressed. It can be advisable to hire external expertise to help with setting
up and writing the quality control manual. Parallel to this writing, several decisions have
to be made about the organizational structure of the laboratory, including all functions,
responsibilities, authorization, and substitutes for vital management functions (Standard
section 4.1), registration of personnel and training records (Standard section 5.2), and the
distribution of documentation (Standard section 4.3).
The writing of the quality control manual, including related procedures, is a process that
is often complicated and time consuming, and demands good cooperation between the
laboratory management and the QA Manager. It is important to emphasize that the laboratory management should decide how these aspects are organized and implemented, while
the QA Manager should judge them against the demands of the Standard.
The QA Manager has an important responsibility in the process of document control
to facilitate the setting up, labelling, release, distribution and review of documents. The
laboratory should have a document control system that includes a distribution list for all
current procedures, to ensure that only the most recent versions are available for personnel
by removing all previous versions. This system can vary from simple spread sheets to a
dedicated database. The standard offers the laboratory the opportunity to make small
manual changes (hand-written amendments signed by authorized persons) to quality
control documents, which can be a useful tool for the modifications frequently required in
the initial phase. Review of documents and implementation of any manual changes should
be undertaken regularly after consultation with the laboratory technical staff. Minor hand
amendments should be made to all copies of controlled documents at the same time to
avoid contradictory procedures. Documents should be reviewed and hand amendments
made permanent as soon as is practicable.
During the second year, the technical performance of the laboratory should be evaluated. At a minimum, this evaluation should address the following parameters: results for
the first line of control; repeatability; z-scores from collaborative trials; status of equipment; and complaints. The QA Manager should conduct this evaluation in the format of
an internal audit (Standard section 4.14). The QA Manager should set up a schedule of
internal audits that describes items to be investigated and the frequency of audits. In this
second year, the use of a checklist covering all aspects of the standard should be used, and
deviations from the procedure should be recorded in audit reports (see Table 4.2, first data
column). In response, an action report should be written by the laboratory that describes
how the identified problems will be rectified, and includes a timeline to achieve this. The
implementation of these actions should be overseen by the QA Manager.
The laboratory should describe and implement procedures for preventive and corrective
actions (Standard sections 4.11 and 4.12). These actions focus on avoiding the release of
unreliable results. Corrective actions are consequences of non-conforming testing, deviations from internal audits, and complaints. The laboratory should follow an active policy
of corrective actions to solve problems and shortcomings, and, if necessary, communicate
problems to customers. Preventive actions aim to avoid future problems by monitoring
trends in critical parameters that are indicators of the quality of the test. The laboratory
should identify and regularly evaluate these parameters; some analytical knowledge is
required for this. Examples of these parameters are any drift observed in control charts;
values for blanks; and sensitivity analysis, such as the slope of calibration curves or peak
areas. Trends can be graphed by using, for example, the CUSUM chart (see AppendixI) to
53
54
provide a visual monitoring system. In an ideal situation this curve should vacillate around
zero, whereas other shifts may be caused by systematic drift.
Preventive and corrective actions undertaken should be well documented, tested for
efficiency, and signed. It is important that the laboratory can show that such procedures
are clearly implemented in their routine operation.
The laboratory management should improve its service to customers by seeking their
feedback (Standard section 4.7). This can be achieved by sending an annual questionnaire
to customers. The QA Manager should evaluate these results and use them as a tool to
improve the laboratorys level of service and to identify opportunities for improvement.
The results from the internal audit and the responses from customers should be used in
a systematic way to improve the QMS of the laboratory (Standard section 4.10).
4.5.5Third year
At the start of the third year, the laboratory should decide if it is ready to proceed with the
official accreditation. If a positive decision is made, the laboratory should gain information
about the most suitable accreditation organizations and contact them about the process. It
is important that this organization is authorized for ISO/IEC 17025:2005 (i.e. be a signatory
to ILAC, International Laboratory Accreditation Cooperation), which is not the case in every
country. Most accreditation bodies will recommend an initial gap analysis assessment followed by an initial assessment against ISO/IEC 17025:2005.
The third year should be devoted to implementing the remaining aspects that are more
specific in the standard, and improving the existing QMS.
The remaining aspects on the technical level are sampling (Standard section 5.7), estimation of the measurement of uncertainty (MU) (Standard section 5.4.6), and traceability
of measurements (Standard section 5.6), for which procedures should be implemented.
The set up and implementation of a procedure for sampling is only necessary if the laboratory conducts sampling for external clients, e.g. batch control of feed ingredients or diets,
as part of their services. The laboratory should prove that it has sufficient analytical and
statistical knowledge to conduct this in an appropriate manner.
Measurement of uncertainty should be estimated for all tests that will be accredited
and although not required for non-accredited tests, it is good practice to also estimate for
tests that are not covered by the accreditation. The laboratory should combine this estimation with an evaluation of the validation conducted in the first year. For accreditation it is
necessary that all elements of the validation (see Appendix C) are estimated in a traceable
way that is preferably based on an international standard. At this stage the laboratory
should have enough knowledge and data to conduct these estimations. The results of these
estimations can be used to calculate the measurement of uncertainty, which reflects the
confidence interval (mostly 95%) of individual results for the different tests. Practice shows
that laboratories can have serious problems with this aspect of the standard, which involves
both analytical and statistical knowledge. International standards are available that can help
laboratories to understand this requirement and perform the necessary calculations (see
Appendix D). The calculated results should be realistic and available to customers.
For accreditation of a method, method validation and its relationship to an international standard method are important issues. Generally, the accreditation organization will
accept the use of three types of methods, i.e. conforms to, based on or modification of, an
in-house method. If a laboratory decides to use a method that conforms to or is based
on or modification of a specific international standard method, it is obliged to prove that
its performance matches the quality parameters of that international standard method. If
the laboratory uses the in-house method, it does not have this obligation. However, it is
imperative that all quality information, such as limit of quantification, precision and accuracy of the in-house methods, are made available to customers and users of the results.
These demands, however, can vary between countries, and the laboratory should check this
with the accrediting organization.
All these new values should be recorded in the analytical protocol for each test. These
changes may be combined with a new standard format for these documents, although this
is not a requirement of the standard.
Each piece of equipment requiring calibration should have calibration records that can
be traced back to an International Reference, using reference standards, calibrated weights,
thermometers, etc. For physical units, such as mass and temperature, calibrations should
be conducted by accredited calibration providers to demonstrate traceability to reference
standards. All relevant documents, including certificates, should be retained by the laboratory. Appendix H describes a traceable calibration for volumetric equipment that can be
performed by the laboratory itself. Unless chemical calibration solutions are made under
certified conditions, the laboratory has an obligation to prove its traceability by testing it
against an independent solution (Standard section 5.6.3). This can be achieved by comparing solutions from different suppliers or against a solution prepared from pure chemicals
under traceable conditions in the laboratory. The laboratory should explicitly describe the
criteria for this comparison in its protocol.
The last few aspects that should be implemented are the procedures for subcontracting
(Standard section 4.5), purchasing of services (Standard section 4.6), and reporting of
results (Standard section 5.10).
For subcontracting, the laboratory should set up a clear selection process, based on
objective criteria. Suitable laboratories for each test should be listed and made available to
customers. The laboratory is obliged to evaluate the quality and service level of the laboratory used for subcontracting. It is important to emphasize that the laboratory remains
responsible for the quality of the results in the case of outsourcing. These results should
not be reported under accreditation.
A procedure and schedule should be available for the purchase of services and supplies,
which also includes the evaluation of suppliers. For this purpose, the laboratory should
make a list of critical chemicals and equipment and the specifications they should be
tested against. The service level of the most frequently used suppliers should be regularly
evaluated and documented. The QA Manager should take a leading role in this process.
Finally, the laboratory should ensure that the report of test results is confirmed against
the demands stated in the standard. A possible approach is to implement a Laboratory
Information Management System (LIMS) that automatically generates reports and meets
standard specifications.
Internal audits should be conducted according to the previous mentioned schedule.
Compared with the previous year, two changes will occur. The first change is the increase in
55
56
Table 4.3
Examples of two forms of audit report

Old form, very brief comments
New form, with more detail recorded
Personnel records:
5.2.5 Personnel records:
The record of technician X was

incomplete
The personnel records of technical staff X, Y and Z were

examined. The records of Y and Z were complete. For technician
X, the training and authorization file was not available.
Calibration of equipment:
5.5.2 Calibration of equipment:
Calibration of balance X was

not performed within specified
time frame
The calibration of balances X, Y and Z, diluter A, and Kjeldahl

equipment were examined. For all equipment, calibration
procedures including specifications are available. The calibrations
were performed in the previous year according to these
procedures and judged against the specifications. Balance X was
calibrated later than the stated deadline, without any action or
registration: follow up on reason for delay.
the number of audits, which may require more people besides the QA Manager to undertake them. In this case, a select group of personnel should be trained to become internal
auditors. It is important to realize that auditing contains a subjective element leading to
the under- and overestimation of some issues depending on the importance an auditor
gives to those issues. For this reason it is advisable to rotate auditors between the different
laboratories or sections on a regular basis. The QA Manager is responsible for setting up a
training programme to provide an in-depth knowledge of the standard for future auditors.
The first audit should be conducted under supervision, after which they will be authorized
if deemed competent by the QA Manager. Training records of all new auditors should be
completed and stored in an appropriate manner. The QA Manager should initiate regular
consultation between auditors to increase the quality and uniformity of the audit process.
The standards for auditing management systems are in ISO 19011:2011.
The second change is the quality and detail expected in the audit reports. In the previous
year, reports were mostly limited to checklists of investigated items and deviations found. The
laboratory should now switch to a more detailed description of which items are investigated
and what observations and deviations are found. The items should be described and labelled
according to the Standard. Table 4.3 illustrates the differences between reports for two
examples.
The management review written at the end of the year should be fully documented
according to the demands of the standard (Standard section 4.15.1). Compared with the
previous years, this review should also contain information about its market positions (type
and volume of work) and the preventive and corrective actions.
4.5.6Fourth year
In the fourth year the laboratory should prove that all aspects of the QMS are implemented
and operational according to the demands of the Standard. This year ends with an inspection from an official of the accreditation organization, and ultimately leads to the accreditation of the laboratory to ISO/IEC 17025:2005. To achieve this, the laboratory should
monitor and improve the existing QMS.
The first step is to update the Quality Control manual and all procedures to ensure that
all issues are sufficiently covered by these documents. For this purpose, the new Quality
manual should also contain a reference table that links specific aspects of the standard to
the appropriate sections in the Quality Control Manual. If in doubt, the laboratory can hire
an external expert to check all its documents against the demands of the standard.
Internal audits to check the operational efficiency of the QMS should be conducted
more frequently during this year. Observed deviations should be recorded and the relevant
corrective actions completed. These audits should also pay attention to the presence of all
relevant quality documentation, such as validation reports, personnel files, worksheets and
calibration reports.
The laboratory manager and QA Manager should educate laboratory personnel
regarding the quality procedures used and prepare them for future external audits. Experience shows that technical staff focus largely on the analytical work and have less inclination
to follow the concept of a QMS in totality, which can lead to a difference between theory
and practice. The QA Manager, supported by the laboratory management, should try to
avoid this situation by explaining the QMS and the benefits of the different procedures.
Performance of an audit by external expert(s) can be beneficial as preparation for technical
staff, prior to the first official audit by the Accreditation Organization.
At the beginning of the fourth year, the laboratory should approach an authorized accreditation organization with the request for accreditation according to ISO/IEC
17025:2005. The laboratory should at this point make a definitive choice about which
tests, including the matrix types, that are to be accredited. It is advisable to start with a
limited number of tests and gradually increase them over the following years as extensions
to scope.
In general, the accreditation process will start with a pre-inspection of the laboratory with the aim to judge if it is capable of achieving the requirements of the standard,
i.e. the presence of the facilities, equipment, Quality Control Manual, and personnel.
After approval, an official inspection will be organized under the supervision of a team
leader (Assessment Manager) who will focus on the general requirements (Chapter 4 of
the Standard). Analytical activities (i.e. tests) are audited by technical experts, who will
judge the quality of the results produced (Chapter 5 of the Standard). The number of
technical experts will depend on the diversity of determinations that require accreditation. In most cases, these inspections will lead to a number of observed deviations (i.e.
non-conformities) that should be addressed within a fixed period (usually 3 months). If
all deviations are corrected in a satisfactory manner, the laboratory will be accredited for
the tests specified.
The accreditation is always for a fixed term, generally four years. During this period,
the laboratory will receive an annual maintenance inspection to judge different tests and
aspects of the quality system. After four years, the whole system will be evaluated and
judged by a new team of experts. After addressing all deviations and initiating corrective
actions, accreditation will be granted for another period of four years.
After the accreditation, the laboratory should clearly inform customers which tests are
accredited. This should also be marked in the final report of results (Standard section 5.10).
Each year the laboratory can decide to add or remove tests from the accredited list or
57
58
scope. New tests will be inspected during the next audit. It is crucial to inform the auditors,
well in advance of the scheduled audit, if new methods are to be inspected. Methodology
and validation reports will be required prior to this audit.
After achieving accreditation to ISO/IEC 17025:2005, the greatest challenge is to
maintain and improve the QMS. A traditional pitfall is that the laboratory starts to relax on
quality issues and focus on other things. The Accreditation Organization, however, expects
the level of quality to increase with time, and this should be the prime objective of laboratory management.
4.6Second situation:
Routine laboratory connected to a feed manufacturer
The main difference between this type of laboratory and that described in the first situation is that it is part of a larger organization. For the implementation of a QMS based
on ISO/IEC 17025:2005 it is important that the organization already has a quality system
based on ISO standards (i.e. ISO 9001:2008). The presence of this system in the organization indicates the integrity of general procedures that are already implemented and operational, which the laboratory can use. An additional advantage is the presence of a quality
control unit whose knowledge and experiences can be utilized to set up the QMS in the
laboratory. These conditions give the laboratory a clear advantage compared with the first
situation, and allows it to concentrate its attention more on the specific technical issues
of ISO/IEC 17025:2005 (Chapter 5 of the Standard). Although there may be overlaps, it
can be advisable to make a separate Quality Control Manual for the laboratory that refers
to more general laboratory procedures. The advantage of this is that it concisely describes
the total QMS of the laboratory, which makes specific inspections of the laboratory easier
to perform. The presence of a general QMS can mean a reduction to three years for the
period required to become accredited. Many of the requirements of an ISO 9001:2008
QMS will be shared with that of an ISO/IEC 17025:2005 QMS (or ISO 14001:2004 Environmental Management System or BS OHSAS 18001:2007 Occupational Health and Safety
Management System) meaning that many SOPs may be shared, and form an Integrated
Management System. It may be that if an organization has certification to ISO 9001:2008,
ISO 14001:2004 or BS OHSAS 18001:2007, they might receive only one assessment visit
from the certification body, who will assess against all three standards during the same
assessment. Accreditation to ISO/IEC 17025:2005 will always be assessed separately by
an accreditation body.
Integration of the laboratory within a larger organization increases its interaction with
other sections and therefore increases the possibility of being influenced by others outside
the laboratory. This issue is explicitly mentioned in Section 4.1.4 of the Standard as the
potential for conflict of interests between the laboratory and production or commercial
units. Upper management should guarantee and protect the independent position of
the laboratory, which should be demonstrated in the structure of the laboratory and the
authorization of its management. Upper management should demonstrate its commitment to the QMS by expressing this as a statement in the Quality Control Manual, by
evaluation of management reviews, and being present at external audits.
4.7Third situation:
Laboratory as part of a research organization
The main task of the laboratory in a research organization is to facilitate the research by
conducting analytical activities. They are less focussed on routine analyses than in the first
two situations. These laboratories have more flexibility in their scope of tests and there
is a strong linkage with researchers, which may make an official accreditation harder to
achieve. Some aspects of the standard, especially the technical issues such as avoiding
improper influence (Standard section 4.1.4) and the validation of in-house and nonstandard methods (Standard sections 5.4.3, 5.4.4 and 5.4.5), can be useful to improve the
quality of the analytical work.
Improper influence has already been discussed in the second situation and focuses on
the independence and responsibility of the laboratory to guarantee an objective judgement
of the results. In organizations where the laboratory is an integrated part of the research
unit there is a strong interaction between technical staff and researchers. The greatest
threat is that the researchers expectations can influence the analytical judgement of technical staff. An example of this influence is illustrated in Table 4.4 which gives duplicate
results for four samples. In the case of sample C, the difference was higher than permitted
and the measurement was repeated leading to an additional value (i.e. 33.9). From an
analytical point of view, the value 31.6 would be regarded as an outlier and the average of
33.9 and 34.7 should be the final results. The researcher however expects a linear relationship and will delete both values and use 31.6 as the final result.
Avoiding this type of interference is a challenge for research organizations where
researchers try to find relationships to explain their observations. The pressure for commercial funding may increase this kind of interference, with the danger of such a subjective
judgement of results.
A solution for this problem may be to limit direct interaction between technical staff and
researchers and most importantly to ensure that the authorization of analytical methods
and results will be conducted by the laboratory according to the set principles. The management of the unit should be aware of this problem and should implement mechanisms
to avoid reporting of unreliable results. Examples of such mechanisms are the use of set
procedures and criteria to evaluate results and conducting audits to evaluate conformity of
the analytical work, both of which are described in the standard.
The second issue is to ensure prior validation of methods before they become operational. Laboratories within research organizations spend more time working on (new)
Table 4.4
Example of measurements
Sample
Measurements
10.0; 10.9
21.4; 20.9
31.6; 34.7; 33.9
40.5; 41.0
59
60
figure 4.3
Absorption versus wavelength

0.2
0.18
Absorption (AU)
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
385
390
395
400
405
410
415
420
Wavelength (nm)
non-standard methods as an innovative aspect of the overall research work compared with
the routine laboratory described in the first two situations. Introduction and implementation of these methods are also based on interaction with researchers and information
available in the literature. Experience shows that these methods are often poorly validated
and the expectations of researchers are used to evaluate the results, which can lead to
unreliable results and conclusions. The laboratory can solve this problem by performing
firstly an independent validation of the tests before use as stated in the standard (Standard
section 5.4.4). The validation of these non-standard methods is however more complicated
than for standard methods because the laboratory has an obligation to prove that the
principle and the associated method are capable of measuring the analyte accurately. The
laboratory should focus additionally on three issues: selectivity; sensitivity or interference in
different matrices; and robustness (see AppendixA). In some cases, the laboratory should
decide which quality parameters have the highest priority. For example, a higher sensitivity
can negatively affect the robustness of the method, as illustrated in Figure4.3, where the
highest sensitivity is achieved at 392nm, leading to the lowest values for LOD and LOQ.
At this wavelength, however, the absorption is highly sensitive to small variations, whereas
between 402 and 408 nm the absorption is more stable, although it is less sensitive,
leading to a higher value for LOD and LOQ. In general, an improved robustness leads to
lower variation in the results and improves the precision.
4.8Fourth situation:
Government or reference laboratories
The important position of government or reference laboratories makes the implementation
of a quality system based on ISO/IEC 17025:2005 mandatory. From the beginning, their
analytical results should have the highest quality level to gain the full confidence of other
laboratories. To achieve this level from the beginning is a great challenge because it does
not give the laboratory time to build its own quality system that can be improved step by
step as described in the first situation. In practice this means that all necessary validations
and quality procedures should be fully implemented before the laboratory can start analysing samples for customers. To meet this challenge, the laboratory should attract quality
manager(s) with suitable experience in this field. It may also be advisable to hire external
experts to assist on special subjects, such as validation and calibration of equipment.
The laboratory should use the international standard methods and participate, if
available, in international collaborative trials for each determination. This means that the
laboratory should have state-of-the-art facilities and equipment, such as GC-MS, sequential in-line MS (MS-MS) and ICP-MS, for the determination of contaminants at trace level.
Constant improvement of the quality of the analytical results and following international
technical developments should be a vital part of their quality policy.
61
63
Sources used
ISO Standards
ISO 3534-2:2006. Statistics Vocabulary and symbols Part 2: Applied statistics. [Reviewed and
confirmed 2010]
ISO 3696:1987. Water for analytical laboratory use Specification and test methods. [Reviewed
and confirmed 2011]
ISO 5725:1994. Accuracy (trueness and precision) of measurement methods and results.
ISO 6498. Animal feeding stuff Preparation of test samples.
ISO 9001:2008. Quality management systems Requirements.
ISO 10012:2003. Measurement management systems Requirements for measurement
processes and measuring equipment.
ISO 14001:2004. Environmental management systems Requirements with guidance for use.
[Reviewed and confirmed 2008]
ISO 19011:2011. Guidelines for auditing management systems.
ISO 21748. Guidance for the use of repeatability, reproducibility and trueness estimates in
measurement uncertainty estimation .
ISO Guide 34:2009. General requirements for the competence of reference material producers.
ISO/IEC 17025:2005. General requirements for the competence of testing and calibration
laboratories. [Reviewed and confirmed 2010]
Other sources
BSI. No date. BS OHSAS 18001:2007. Occupational Health and Safety Management System.
See:
http://www.ohsas-18001-occupational-health-and-safety.com/ohsas-18001-kit.htm
Accessed 2013-08-03.
EURACHEM. 1998. EURACHEM Guide. The Fitness for Purpose of Analytical Methods. A
Laboratory Guide to Method Validation and Related Topics. A document developed by a
EURACHEM Working Group. See: http://www.eurachem.org/images/stories/Guides/pdf/valid.
pdf Accessed 2013-08-03.
EURACHEM. 2007. EURACHEM/CITAC Guide: Use of uncertainty information in compliance
assessment. First edition 2007. Edited by S.L.R. Ellison and A. Williams. See http://www.
measurementuncertainty.org/pdf/Interpretation_with_expanded%20uncertainty_2007_
v1.pdf Accessed 2013-08-03.
FAO. 2011. Quality assurance for animal feed analysis laboratories. FAO Animal Production
and Health Manual, No. 14. Rome, Italy. Available at http://www.fao.org/docrep/014/i2441e/
i2441e00.pdf Accessed 2013-08-29.
FAO. 2013. Quality assurance for microbiology in feed analysis laboratories. Prepared by R.A.
Cowie and H.P.S. Makkar. FAO Animal Production and Health Manual, No. 16. Rome, Italy.
Available at http://www.fao.org/docrep/018/i3287e/i3287e.pdf Accessed 2013-08-29.
64
ILAC-G17 2002. Introducing the Concept of Uncertainty of measurement in Testing in

Association with the Application of the Standard ISO/IEC 17025. See: ILAC Web site www.
ilac.org
JCGM 100:2008. [GUM 1995 with minor corrections]. Evaluation of measurement data Guide
to the expression of uncertainty in measurement / valuation des donnes de mesure Guide
pour lexpression de lincertitude de mesure. Document produced by Working Group 1 of the
Joint Committee for Guides in Metrology (JCGM/WG 1).
JCGM 200:2008. International vocabulary of metrology Basic and general concepts and
associated terms (VIM) / Vocabulaire international de mtrologie Concepts fondamentaux
et gnraux et termes associs (VIM). Available at http://www.bipm.org/utils/common/
documents/jcgm/JCGM_200_2008.pdf. Accessed 2013-08-29.
OIML R 111-1 Edition 2004 (E). International Recommendation. Weights of classes E1, E2, F1,
F2, M1, M12, M2, M23 and M3. Part 1: Metrological and technical requirements. International
Organization of Legal Metrology. See: http://www.oiml.org/publications/R/R111-1-e04.pdf
Accessed 2013-08-03.
Thompson, M., Ellison, S.L.R. & Wood, R. 2002. Harmonized guidelines for single-laboratory
validation of methods of analysis (IUPAC Technical Report). Pure and Applied Chemistry,
74(5):835855.
VIM See: JCGM 200:2008.
65
Appendix A
Ensuring quality analytical

performance
A1.Determination of limit of detection (LOD) and limit of
quantification (LOQ)
LOD is the lowest analyte concentration that can be reliably distinguished from the blank,
while LOQ is the lowest concentration at which the analyte can be reliably detected and
also at which predefined goals for accuracy and precision can be met.
The internationally accepted formulas for the estimation of the LOD and LOQ are based
on the signal of a blank sample (i.e. without the analyte) and its standard deviation (SD)1
LOD = blank + 3SD (blank)
LOQ = blank + 10SD (blank)
Examples of LOD and LOQ calculations for different method types are given in the
following sections.
A1.1Gravimetric analyses
The LOD and LOQ of gravimetric determinations depend on the accuracy of the balance
used, which is often specified by the supplier of the balance. The accuracy of a good quality
four-figure balance should be approximately 0.3mg (supplier specification).
This accuracy is equal to the 95% confidence interval, which is the average 2SD
(standard deviation). The range of this confidence interval or the accuracy equals 4SD,
which means if the accuracy is 0.3mg (in this case) the SD is 0.075mg (i.e. 0.3/4).
For balances, the signal of a blank can be set to zero.
Using the general formulas:
LOD = 0 + 30.075 = 0.225mg
LOQ = 0 + 100.075 = 0.750mg
The LOD and LOQ for determinations are generally expressed on a content basis and
therefore depend on the amount of sample used. For example, if 1g of sample is used,
using the above example, the LOD will be 0.225g/kg (i.e. 0.225mg/1g) and the LOQ will
be 0.750g/kg (i.e. 0.750mg/1g).
The LOD and LOQ can be improved, i.e. values for LOD and LOQ can be lowered, if
required for more sensitive measurements by using more sample material or a more accurate balance. The above-stated LOD and LOQ values for the gravimetric determinations,
such as dry matter, ash, fat and fibre, are generally sufficient to analyse feedstuffs without
the need for further improvement.
MacDougall, D & Crummett, W.B. 1980. Guidelines for data acquisition and data quality evaluation in
environmental chemistry. Analytical Chemistry, 52(14):22422249.
66
If the accuracy information is not available from the supplier of the balance, the laboratory should estimate the standard deviation of the balance by measuring a weight several
times (at least ten times) over several days.
A1.2Titration-based analyses
The LOD and LOQ of titration-based analyses, such as the Kjeldahl determination for
nitrogen (N), depend on the value found for the blank and the standard deviation of the
blank. For Kjeldahl determinations, the blank is the digestion solution, including catalysts
but omitting the sample. These blanks should be analysed at least six times, and used as
in the following calculation.
An example:
Values observed (ml) are 0.15; 0.18; 0.13; 0.14; 0.18; 0.17, which leads to an average of
0.158ml and a standard deviation of 0.021ml.

LOD = 0.158 + 30.021 = 0.222ml

LOQ = 0.158 + 100.021 = 0.372ml
To express the LOD and LOQ as content in the sample, the following formula for the
determination of N should be used:
N(g/kg) = (V - Vblank)M(acid)f14/W
Where: V is the volume of the acid used (i.e. 0.222ml for LOD and 0.372ml for LOQ)
Vblank is the volume of the acid used by the blank (i.e. 0.158ml)

M(acid) is the molarity of the acid used (i.e. 0.1M HCl)

f is the acid factor (i.e. 1 for HCl)

14 is atomic weight of Nitrogen (N)

W the weight of the sample (i.e. 1g)
For LOD: N = (0.222 - 0.158)0.1114/1 = 0.09g/kg
For LOQ: N = (0.372 - 0.158)0.11 14/1 = 0.30g/kg
A1.3Spectrometric analyses
The LOD and LOQ depend on the absorption value found for the blank and the standard
deviation of the blank. The blank is the diluted reagents or colour reagents, without the
analyte. These blanks should be analysed at least six times and used as in the following
calculation.
An example:
Observed absorption units (A.U.) are 0.004; 0.005; 0.003; 0.003; 0.006; 0.008, which
leads to an average of 0.004A.U. and a standard deviation of 0.002A.U.
LOD = 0.004 + 30.002 = 0.010 A.U.
LOQ = 0.004 + 100.002 = 0.024 A.U.
To express the LOD and LOQ as the concentration of the analyte in the measured solution, the calibration curve should be used. For example, the calibration curve is:
Absorption (A.U.) = 0.500concentration (mg/L)
The LOD in the measured solutions is 0.020 mg/L (i.e. 0.010/0.500) and the LOQ is
0.048mg/L (i.e. 0.024/0.500).
Appendix A Ensuring quality analytical performance
To express the LOD and LOQ as a content of the sample, the following formula should
be used, and in general for spectrophotometric determinations, the formula can be
expressed as:
Content (g/kg) = concentration (mg/L)volume (L)dilution factor/sample weight (g).
Where:
Concentration is the content of the analyte in the measured solution.
Volume is the end volume of the extracted solution
Dilution factor is the additional dilution needed to get the value of the measured
solution within the range of the calibration line.
Weight is the amount of sample used.
For example, if 1g of sample is used, the end volume is 0.100L, and the dilution factor
is 25, then
LOD = 0.0200.125/1 = 0.05g/kg
LOQ = 0.0480.125/1 = 0.12g/kg
LOD and LOQ for spectrometric methods can be improved by increasing the sample
amount and decreasing the volume and dilution factor.
A1.4Chromatographic analyses
Different methods for the determination of the LOD and LOQ for chromatographic
methods are available. A good approach is to analyse an analyte at a near background level
and to determine the standard deviation at the peak area. This value should translate to
concentrations and content by using the same approach as for the spectrometric method.
An example:
Values found (Area) are 2500; 2900; 3000; 2700; 2400; 2700, which give an average of
2700 area and a standard deviation of 228 area. Using the general formulas:

LOD = 3228 = 684 Area

LOQ = 10228 = 2280 Area
To express the LOD and LOQ as a concentration of the analyte in the measured solution,
the calibration curve should be used. For example, if the calibration curve is:
Area = 100000concentration (mg/L)
the LOD in the measured solution is 0.00684mg/L (i.e. 684/100000) and the LOQ in the
measured solution is 0.0228mg/L (i.e. 2280/100000).
To express the LOD and LOQ as a content of the sample, the formula used to calculate
content should be used:
Content (g/kg) = concentration (mg/L)volume (L)dilution factor/sample weight (g)
Where: Concentration is the content of the analyte in the measured solution.
Volume is the end volume of the extracted solution.
Dilution factor is the additional dilution needed to get the value of the measured
solution within the range of the calibration line.
Weight is the amount of sample used.
For example, if 0.5 g sample is used, the end volume of the measured solution is
0.050L, with a dilution factor of 10:

LOD = 0.00680.05010/0.5 = 0.0068g/kg = 6.8mg/kg

LOQ = 0.0220.05010/0.5 = 0.022g/kg = 22mg/kg
67
68
To improve the LOD and LOQ, the same approach can be used as for spectrometric
determinations. Chromatographic methods also offer additional opportunity through
increasing the injection volume used, which will increase the sensitivity or slope of the
calibration curve.
A2.Linearity
Linearity is generally determined by using a calibration curve of the analyte and extrapolating to predict the value of the next calibration point. This value should be compared with
the measured value and the difference should not be larger than a fixed value (generally
5%).
An example:
For calibration solutions of 2, 4, 6, 8 and 10 mg/L the measured absorption values are
0.203, 0.405, 0.608, 0.790 and 0.930. The first step is to calculate the calibration line
based on linear regression for the first three points (i.e. 2, 4 and 6mg/L), which leads to
the curve:
Absorption = 0.1012concentration (mg/L) + 0.0003.
The next step is to predict the absorption value for 8 mg/L, which is 0.810 (i.e.
0.10128 + 0.0003). The difference from the measured value is 0.020 (i.e. 0.810 0.790)
or 2.5%, which is within the acceptable range (<5%). Therefore the linearity can be
extended to 8mg/L and the calibration curve is calculated for the first four points, which
leads to the curve:
Absorption = 0.0982concentration (mg/L) + 0.010.
The next step is to predict the absorption value for 10 mg/L, which is 0.992 (i.e.
0.098210 + 0.010). The difference from the measured value is 0.062 (i.e. 0.992 0.930),
or 6.2%, which is outside the acceptable range (>5%). Therefore the linearity is limited to
8mg/L for the measured solution.
For results above this concentration, the measured solution should be diluted so that
the concentration is <8mg/L.
A3.Precision
The precision of a determination is expressed as the repeatability and the intra-laboratory
reproducibility, which are related to the standard deviation (SD). The repeatability equals by
definition 22 (i.e. 2.8)SD if the sample is measured in the same batch (equal conditions).
The intra-laboratory reproducibility equals by definition 22 (i.e. 2.8) SD if the sample
is measured in different batches (non-equal conditions). Table A1 below shows a typical
example of results obtained for the determination of the precision. The same sample is
measured in triplicate on three different days. For each batch, the average and SD is estimated and used to calculate the relative SD (i.e. standard deviation100%/average) and
the relative repeatability (i.e. relative SD 22). The relative repeatability of the batches
varied from 2.92 to 3.42%, leading to an average value of 3.4%.
The intra-reproducibility is calculated using all values from the different batches (in this
case nine values) leading to an average of 39.8 and an SD of 0.79 (see last line of Table
A1). The relative SD is 1.98, leading to a relative intra-laboratory reproducibility of 5.55%
(i.e. 221.98) and is in this case 5.5%.
69
Table A1
Calculation of precision
Measured values
Average
Standard
deviation (SD)
Relative SD
Relative
repeatability
Batch 1
40.3
40.1
39.5
39.97
0.42
1.04
2.92
Batch 2
38.3
39.2
39.3
38.93
0.55
1.41
3.96
Batch 3
41.0
40.2
40.1
40.43
0.49
1.22
3.42
39.8
0.79
1.98
5.55
Total
The value for repeatability is normally lower than for the intra-laboratory reproducibility
because it involves fewer sources of variation.
Generally, laboratories analyse duplicates in the same batch and should use the relative
repeatability as part of their acceptance criteria. The intra-laboratory reproducibility could
be used as criteria for samples, such as the control sample, and this should be analysed in
every batch.
A4.Accuracy
Accuracy can be tested by analysing certified reference materials (CRM) or by collaborative
trials (performance testing). In both cases, the z-score could be used to evaluate the result.
The z-score is the difference between the measured and the stated value expressed in units
of the SD.
For CRMs, the stated value is the reference value given by the supplier, which is mostly
based on a collaborative trial using various techniques. The laboratory should compare this
value with the measured value and divide the difference by its intra-laboratory reproducibility for that method.
An example:
Measured value is 10.0mM and the relative SD is 2%, or 0.2mM in this case. The reference
value is 9.5mM. The calculated z-score is:
Z-score = | 10.0 9.5 | / 0.2 = 2.5
For collaborative trials, the laboratory should compare its result with the average and
divide the difference by the standard deviation of the collaborative trial. In most cases, the
organizer will automatically do this calculation and report the z-scores for each individual
participant.
A z-score <3 means that the measured value is within the 99% confidence interval (i.e.
3SD) or the reference value (in case of CRM) or the average of the collaborative trial. In
this case, no systematic error is observed, meaning that the accuracy of the test is sufficient.
If the z-score is >3, there is a systematic error and the laboratory should try to identify
the cause and rectify the problem. Different methods, such as modifying the protocol or
the use of an internal standard, are available to improve the accuracy. Information about
the accuracy should be used in the calculation of the uncertainty of measurement (see
AppendixD), and mentioned in any communication with customers.
70
A5.Selectivity and sensitivity

Selectivity means that the measured value is only caused by the specific analyte. Consequently, the absence of this analyte should lead to no detectable signal. However, this is
often not the case. Gravimetric methods are non-selective and mostly based on assumptions, as in the case of crude ash, fat and fibre. For chromatography, the selectivity is based
on the quality of the separation and the type of detector, which means that only detectable
analytes can be investigated. The use of a more selective detector, such as fluorescence,
limits the list of compounds that can interfere compared with UV detection. A list of possible interfering compounds should be analysed with the method and evaluated.
Sensitivity is the response of the measured value relative to the concentration of the
analyte, and is mostly expressed as the slope of a calibration line, or the response factor in
case of a one-point calibration. Other components can influence the sensitivity in two ways.
Firstly, their presence can lead to a background signal, which is often found for colour-based
spectrometric determinations. The results should be corrected for this background if this
value is always the same. If these values differ, individual corrections should be made by
measuring the differences (as for kinetic methods). An example of this approach is:
Calibration solution:
Sample solution:
Start: 0.004
Start: 0.069
A = 0.246
A = 0.080
End: 0.250
End: 0.149
Both A values are used for the calculation of the analyte in the sample.
Secondly, the matrix of the sample can interfere with the determination of the analyte
and therefore affects the sensitivity. This effect is demonstrated in TableA2, showing the
responses to the same analyte at different levels, in different matrices.
The results show that the response in water is high when compared with acid. This
means that if the samples have an acid matrix, the use of a calibration curve in water will
lead to an underestimation of the content of the analyte. This problem can be observed in
the determination of minerals by AAS or ICP and colour-based spectrometric determinations, and can be solved by matrix matching, i.e. the matrix of the samples and those of
the calibration solutions should be the same.
A particular problem occurs when there is a matrix difference between samples, leading
to different response factors for the analyte. In this case, the method of standard addition
Table A2
Response variation with concentration of analyte

Analyte (ppm)
Slope
Absorption (A.U.) in
Water
Acid
0.000
0.000
0.100
0.090
0.200
0.180
0.300
0.270
0.100
0.090
71
to each sample should be applied. The method of standard addition is based on adding
a fixed amount of the analyte to the sample and using the increase in the response to
calculate its original content.
An example:
A liquid sample is presented with an unknown content a, of analyte X.
First the sample will be diluted in two ways.
Dilution #1: 1ml sample + 1ml solution without analyte X. The content of analyte
X in this solution becomes 0.5 a. The measured absorption of this solution was
0.200A.U.
Dilution #2: 1ml sample + 1ml solution containing 1mg analyte X (content is 1mg/
ml). The content of analyte X in this solution becomes 0.5a + 0.5 (i.e. the content
of the diluted sample plus the content of the diluted solution containing analyte X).
The measured absorption of this solution was 0.300A.U.
The difference in absorption between both solutions is 0.100 A.U., which is due to the
difference in concentration of 0.5mg/ml (i.e. (0.5a + 0.5) 0.5a). Therefore a concentration of 0.5mg/ml leads to an absorption of 0.100A.U. The absorption of dilution#1
was 0.200 A.U. which corresponds to a concentration of 1.0mg/ml, which equals 0.5a.
Therefore a, or the content in the original sample, is 2.0mg/ml.
A6.Robustness
Robustness is the effect of variation in the conditions on the final result, such as differences
in temperature and time during incubations, extractions, drying and incineration steps.
These effects can be investigated by performing the test and varying these conditions,
such as drying at 105C instead of 103C. The differences can be expressed as a z-score to
evaluate if they are significant.
An example:
Dry matter of a control sample is 910g/kg with an SD of 1g/kg when drying at 103 C.
Table3 shows the results for the control sample obtained at different temperatures.
The z-scores at 102 and 104 C are 2, which means that these results are within the
99% confidence interval of the value found at 103 C and do not lead to a significant difference in the result. The z-scores at 101 and 105 C are 4, which means that these results
are outside of the 99% confidence interval of the value found at 103 C, and therefore
significant. In this case the drying temperature should be between 102 and 104 C, which
should be described in the protocol as 103 1 C.
Table A3
Effect of temperature variations on dry matter results

Temperature (C)
Dry matter
z-score
101
914
102
912
104
908
105
906
73
Appendix B
First line of quality control and

the use of Shewhart charts
B1. First step: choice of the control standard
The principle of this procedure is based on the assumption that the deviation in the results
found whilst using the control standard is similar to that of individual samples. To fulfil this
assumption, the matrix and content of the control standard(s) should be comparable with
the samples to be analysed. This comparison is also necessary because the results of the
control standards are used to calculate the precision and uncertainty of measurement for
samples.
For animal nutrition laboratories, the most obvious choices for control samples are feed
ingredients or complete diets. Special attention should be focused on the level of individual
nutrients in these products. Maize for instance, cannot be used as a control standard for
fat, whereas fat-rich soybeans can be used for starch.
The physical requirements of the control standard are stability and homogeneity, which
allows results to be compared over time and between batches (see third step, below). The
requirements of a control sample are primarily precision and repeatability, rather than accuracy. The laboratory can produce its own control sample at relatively low cost.
B2. Second step: preparation of the control standard

The laboratory can produce its own control standard as long as the requirements regarding
stability and homogeneity are fulfilled. The standard procedure is to select a large batch
(at least a few kilograms) of material (such as a feed ingredient or diet), dry, and grind
to a specific size (usually 1mm). The material should be divided into smaller portions of
approximately 300g by using a splitter or a rotation sample divider (see ISO6498). These
portions should be numbered and stored in plastic bags or bottles under appropriate conditions (generally at +2 to +8 C).
Particularly in the case of contaminants, the laboratory can create control standards by
addition of a fixed amount of the analyte(s) to be determined into the control standard
material. Special storage conditions (low temperature, under vacuum, in the dark, etc.) may
be required if the analyte to be measured is unstable.
B3. Third step: determination and control of homogeneity of

the control standard.
The level of each analyte for which the control standard will be used should be determined
by chemical analysis. This determination can be combined with the test for homogeneity
of the control standard. For this purpose, the square root of the number of bags or bottles
(in which the sample was distributed) should be analysed in triplicate at least. These initial
74
samples should be divided at the time of the sampling process. Results should be statistically analysed by Analysis of Variance (ANOVA) to show that the variations within portions
are not significantly different. After passing this test, the results should be combined to
calculate the overall average and SD, and these form the basis for the limits in the first
control or Shewhart chart.
B4. Fourth step: set up and use of the Shewhart chart

The Shewhart chart is designed based on the results obtained from the previous step and
contains five lines:
The first line represents the average value found for the analyte.
The second and third lines are the boundaries of the 95% confidence interval. The
values for these boundaries are calculated by average 2SD (lower line) and average +2SD (upper line).
The fourth and fifth lines are the boundaries of the 99% confidence interval. The values for these boundaries are calculated by average 3SD (lower line) and average
+3SD (upper line).
An example:
If the average is 100g/kg and the standard deviation is 5g/kg, the first line (average value)
will be at 100g/kg, the second and third 95% confidence lines will be at 90g/kg (i.e. 100
25) and 110g/kg (i.e. 100 +25), and the fourth and fifth (99% confidence) lines will
be at 85g/kg (i.e. 100 35) and 115g/kg (i.e. 100 +35). This means that if considering
only random variation, 95% of all results should be between 90 and 110g/kg and 99.5%
of all results should be between 85 and 115g/kg.
In this chart, the results of the control standard found in the analysed batches are
chronologically drawn, as illustrated in FigureB1. The middle line represents the average,
the lines above and below that are the boundaries of the 95% confidence interval, and the
uppermost and lowermost lines are the boundaries of the 99.5% confidence interval. The
dots represent the results found for the control standard in the batches.
The quality of the results found in different batches are judged based on the result of
the control standard in that batch. Although there are no strict internationally accepted
criteria for this judgement, the following criteria are generally used to identify batches that
do not meet the acceptance specifications:
1. Outside the 99.5% confidence interval, which means in this case a content lower
than 85g/kg or higher than 115g/kg.
2.
Twice in sequence outside the same side of the 95% confidence interval, which means
in this case twice between 85 and 90g/kg or twice between 110 and 115g/kg.
3. More than ten times in sequence on the same side of the average.
These criteria are selected based on the chance that this situation occurs in the case of
random variation. In the first situation, this chance is 0.5%, or 1 in 200 batches, whereas in
the second and third situations these chances are even smaller. These situations are therefore a strong indication of a problem within the analyses of that batch, which could lead
to unreliable results. In FigureB1, this situation occurs in batch 2 (outside the 99.5% confidence interval), batch 6 (twice outside on the same side of the 95% confidence interval),
and batch 17 (more than 10 times on the same side of the average). The results obtained
Appendix B First line of quality control and the use of Shewhart charts
75
figure B1
Example of Shewhart chart with 95% and 99.5% confidence levels

140
130
Content (g/kg)
120
110
Content
Average
-3s
-2s
2s
3s
100
90
80
70
60
0
10
15
20
25
30
35
Batch number
in these batches are unreliable and should be rejected, and action should be taken. In most
cases, the batch will initially be repeated. If the results for the control standard remain
outside the expected control values, the test should be stopped and a more structured
investigation instigated (see Standard sections 4.9 and 4.10).
After 30 measurements, or after a specific time interval, the average, the standard deviation, and consequently the boundaries of the confidence intervals, should be re-calculated
based on previous and new data. A chart with new lines should be drawn and used for
the next values for the control standard. In more sophisticated approaches, an F-test and
a t-test can be performed to check the equality between the old and new Stewarts chart.
77
Appendix C
Validation requirements
For all tests, the laboratory should first set up and conduct a validation study.
The first step is to define the matrices and typical concentration level of the analyte,
which should be based on the routine samples the laboratory usually analyses. Although
a smaller number of matrices will reduce the time spent on the validation, the laboratory
should note that this also reduces the types of sample that it can analyse with this test.
The second step is the choice of the method. For validation requirements, a distinction
should be made between standard methods that are officially published, and laboratorydeveloped (i.e. in-house) or non-standard methods.
C1. Standard methods

The validation study should contain a determination of the following items:
Limit of Detection;
Limit of Quantification;
Accuracy; and
Precision (repeatability and reproducibility).
If the method uses a calibration standard curve, the linearity should also be examined.
The values found can be used in two different ways. Firstly, these values can be
compared with those mentioned in the standard method. Values must be equal to the
standard values as a minimum requirement. The laboratory can also claim conformity if its
performance is better than that stated by the standard method.
Secondly, the laboratory may use values from the validation study to describe the quality
of the test without comparing them with those from the standard method.
C2.Non-standard methods
Aside from the items mentioned in the standard methods, a validation study should also
address selectivity, sensitivity and robustness. The laboratory should use the results from the
validation study to describe the quality of the test.
79
Appendix D
Calculation of uncertainty of
measurement
The uncertainty of measurement is the dispersion of the value that is related to the concentration of the substance being measured. This dispersion is mostly expressed as a 95%
confidence level around the value found. The principle of the calculation of the uncertainty
is described in Eurachem/CITAC guide. Quantifying uncertainty in analytical measurement
(available at: www.eurachem.org/index.php/publications/guides/quam), which also gives
examples to illustrate this complex measurement. The aim of this present appendix is to
give some general information that can help laboratories to understand the principle and
perform the calculations.
Methods can be divided into two groups:
1.
Rational methods: the values of the analyte should be independent of the method
used. Examples are mineral analyses or individual organics, such as amino acids and
fatty acids.
2.
Empirical methods: the value of the analyte depends on the method used. Examples are the determination of crude protein, crude fat and crude fibre.
Rational methods can contain a bias or systematic error leading to a difference for the
true value, whereas for empirical methods this bias is by definition absent.
There are two approaches to calculate the total uncertainty:
1. Evaluate the uncertainty from each individual source and combine them. This is a
theoretical approach.
2. Estimate the combined uncertainty from method performance data by combining
only those individual sources of error if they contribute to more than one-third of
the total uncertainty.
The second approach is closer to actual practice and is therefore preferable. This
method however requires performance data, and a certain time period before it can be
performed accurately. This approach will now be discussed in more detail. In general the
uncertainty is calculated from:
Estimation of the precision.
Bias study, which can be performed by analysing certified reference material (CRM),
comparison with a reference method, or by measuring the percentage recovery.
Additional factors outside the standard analyses that can contribute to the
uncertainty, such as sampling or storage.
For empirical methods, the bias is considered to be zero and the laboratory should control the method parameters, such as time or temperature. For rational methods, the laboratory should estimate the significance of the bias compared with the combined uncertainty.
If this value is significant, the laboratory should make a correction or report this value to
80
Table D1
Example of the estimation of the uncertainty

Source
Average
Uncertainty
Relative uncertainty
0.1
Precision
50
Bias
45
4.5
0.1
Combined
0.14
Expanded
0.28
the customer. Information obtained from collaborative trials or CRM, can for both method
types be used directly for the estimation of the uncertainty.
The combined uncertainty is calculated as the root of the sum of the squares of the individual uncertainties. This combined uncertainty is the SD of the determination, and covers
all possible sources of variation in its result. The 95% confidence interval is calculated by
multiplying the combined uncertainty by a factor of 2, which gives the expanded uncertainty and is commonly reported. The uncertainty is expressed as a maximum of 2 digits,
which also determines the number of digits for the result.
The contribution for precision can be estimated by performing duplicate analyses and
calculating the SD of the normalized differences (difference divided by the mean) and divide
by 2. The contribution for bias can be estimated by performing recovery experiments and
calculating the average SD, which should be divided by (number of measurements).
The calculation of the combined and expanded uncertainty is given in TableD1.
In this case, the relative expanded uncertainty is 0.28, which means that the 95% confidence interval (20.14) of the measured value is the value 0.28value. Suppose that the
result is 200g/kg, then the 95% confidence interval is between 144 (i.e. 200 0.28200)
and 256g/kg (i.e. 200 +0.28200).
Additional individual sources should be evaluated if they contribute more than one-third
of the combined uncertainty, which in this case is approximately 0.05.
81
Appendix E
An example of technical records

for a determination
Technical records should contain all relevant information that is necessary for traceability of
the performance of the test and the end results. A standard format for the determination
of nitrogen in samples is given below.
Determination of N according to
List number:
Date:
Technician/Operator:
Titrant:
Volume blank (V0):
code of the test

NNN
DD/MM/YYYY
name
0.05M H2SO4
0.15ml
Sample list
No.
Label
Weight (g)
2012-100
0.7012
2012-101
0.6934
2012-102
V1 (ml)
V1 -V0 (ml)
N (g/kg)1
5.34
5.19
10.4
6.72
6.47
13.1
0.7056
14.99
14.84
29.4
2012-103
0.6912
18.74
18.59
37.7
2012-104
0.7134
15.43
15.28
30.0
2012-105
0.6911
8.79
8.64
17.5
2012-106
0.7123
12.34
12.19
24.0
2012-107
0.6975
4.35
4.20
8.4
2012-108
0.7044
9.78
9.63
19.1
10
Control standard
0.6943
13.45
13.30
26.8
Notes: (1)N (g/kg) calculated as (V1 V0)titrant (0.05)214/weight
Technician/Operator:
Control by:
Approved:
Remarks:
Date:
Signed:
Signature of technician/operator
Name of authorized person
Yes or No
Special observations or other important issues
Date of control
Signature of authorized person
83
Appendix F
An example of a maintenance
and calibration document
A possible format for a maintenance and calibration document could be:
Maintenance and calibration document

Category:
Manufacturer:
Type:
Serial number:
Identification code:
Position:
Analytical balance
Name of manufacturer
Model of balance
Identification Number from manufacturer
Registration code given by the laboratory
Registration code of the laboratory room
User maintenance and calibration

Daily performance
Clean outside of equipment.
Perform internal calibration check.
Every three months

Calibration using mass weights of NNN gram. Measured values should be within
N.Nmg of the reference value.
Supplier maintenance and calibration.
Annually
Cleaning and check of equipment, including electrical safety.
Calibration of balance traceable to International Standards. Measured values should
be within N.Nmg of the reference value.
Adjustment of equipment in case of deviation.
85
Appendix G
An example of a training record
Training record
Name of staff member:
Name
Name of supervisor:
Name
Determination:
Description of the test (internal code)
Training activities:
Explanation and discussion about the method, including the use of the equipment and
control standards
Analysis of list XX under supervision:
Differences between duplicate analyses were within the specifications. Results for the
control standard were within the specification (see appended sheet for data).
Performance without supervision of a list of control standards. Nine out of 10 within
specifications (see appended sheet).

Authorization of Name of the staff member from date.
Signed by
Signature of supervisor
Date: DD/MM/YYYY

Notes:
1. Different categories of training status can be included, e.g. has read and understood the method, some experience with supervision, capable of performing
the method without supervision or ability to train others.
2. Attach appended sheets.
87
Appendix H
Procedure for traceability of

volumetric calibration
This procedure describes a method to perform a traceability calibration of volumetric
equipment used by the laboratory. In practice, this could be used for diluters, pipettes and
volumetric flasks.
General principle
Boil demineralized water to remove CO2 gas and allow it to cool down to room
temperature.
Determine the absolute temperature of the water with a calibrated thermometer
(take into account the measured uncertainty in the thermometer).

Determine the mass of the volume of water obtained by the tested volumetric
equipment. For this purpose, use a certified balance (take into account the measured
uncertainty in the balance).
Evaluate the results against the described specifications.
An example
Calibration of a diluter with fixed syringes of 1.0ml and 5.0ml.
Specifications: Deviation of volume of syringes <0.5%.
Temperature of the boiled water, after cooling: 20.0C
Systematic deviation of thermometer used against reference: +0.5C
Temperature of boiled water after correction: 19.5C
Density of water at 19.5C: 0.9983g/ml
For 1.0ml syringe:
Disperse 5 times the volume of the syringe and record weight.
Weight: 4.9958g
Systematic deviation of balance used against reference: -0.0003g
Corrected mass: 4.9961g
The volume should be calculated by the mass divided by the density, which equals
5.0046 ml (i.e. 4.9961 g/0.9983 g/ml). The deviation against the stated value (i.e.
5.0000ml) is 0.09%, which is within specifications.
For 5.0ml syringe:
Disperse 5 times the volume of the syringe and record weight.
Weight: 24.7861g
Systematic deviation of balance used against reference: -0.0003g
Corrected mass: 24.7864g
The volume should be calculated as the mass divided by the density, which equals
24.8286 ml (i.e. 24.7864 g/0.9983 g/ml). The deviation against the stated value (i.e.
25.0000ml) is 0.68%, which is outside specifications.
In this case, further action is needed.
89
Appendix I
Trend analysis
Cumulative sum (CUSUM) charts are a powerful tool to detect small shifts in the mean of
a process and are more suitable for this purpose than the Shewhart charts described in
AppendixB. In these charts the deviation for the approved average is cumulative, plotted
against time and therefore visible if a drift appears in the assay.
An example
Assume the same determination and Shewhart chart as in Appendix B. Compare two data
sets of results for the control standard given in TableI1. In both cases, the results found are
within the 95%-confidence interval (i.e. between 90 and 110 g/kg), which means that no
further action is required according to the criteria as described in Appendix B.
The CUSUM value is calculated as the difference between the result minus the average
plus the previous CUSUM value. In the case of situation 1, the first CUSUM value is -5 (i.e.
95 100) and the second is 0 (i.e. 105 100 +(-5)). The CUSUM estimation however reveals
a remarkable difference between both situations. In situation 1, the CUSUM value varies
around zero, which is a normal result if only random variation occurs. In the second situation, however, the value clearly increases with time with the cumulative negative effect
very apparent. This is an indication that a possible drift is present in the test, leading in this
case to a systematic underestimation in the results found.
Although this method is easy to perform, deciding when to act is more complicated. In
the literature, a complicated statistical method, known as V-masks, is described, but this is
less suitable for use under normal laboratory conditions. A more pragmatic approach is to
set a maximum value for each test, based on its SD and practical experience.
TableI1
Example of the estimation of the CUSUM values

Situation 1
Situation 2
Result (g/kg)
CUSUM
Result (g/kg)
CUSUM
95
-5
99
-1
105
97
-4
99
-1
93
-11
97
-4
101
-10
107
94
-16
93
-4
99
-17
105
97
-20
FAO ANIMAL PRODUCTION AND HEALTH GUIDELINES

1. Collection of entomological baseline data for tsetse
area-wide integrated pest management programmes, 2009 (E)
2. Preparation of national strategies and action plans for
animal genetic resources, 2009 (E, F, S, R, C)
3. Breeding strategies for sustainable management of animal genetic resources, 2010
(E, F, S, R, Ar, C**)
4. A value chain approach to animal diseases risk management Technical foundations
and practical framework for field application, 2011 (E)
5. Guidelines for the preparation of livestock sector reviews, 2011 (E)
6. Developing the institutional framework for the management of
animal genetic resources, 2011 (E, F, S)
7. Surveying and monitoring of animal genetic resources, 2011 (E, F, S)
8. Guide to good dairy farming practice, 2011 (E, F, S, R, Ar, Pt)
9. Molecular genetic characterization of animal genetic resources, 2011 (E)
10. Designing and implementing livestock value chain studies, 2012 (E)
11. Phenotypic characterization of animal genetic resources, 2012 (E, F)
12. Cryoconservation of animal genetic resources, 2012 (E)
13. Handbook on regulatory frameworks for the control and eradication of hpai and other
transboundary animal diseases A guide to reviewing and developing the necessary
policy, institutional and legal frameworks, 2013 (E)
14. In vivo conservation of animal genetic resources, 2013 (E)
15. The feed analysis laboratory: establishment and quality control, 2013 (E**)
Availability: November 2013
Ar Arabic
C Chinese
E English
F French
Pt Portuguese
R Russian
S Spanish
Multil Multilingual
* Out of print
** In preparation
e E-publication
The FAO Animal Production and Health Guidelines are available through the authorized FAO
Sales Agents or directly from Sales and Marketing Group, FAO, Viale delle Terme di Caracalla,
00153 Rome, Italy.
Find more publications at

http://www.fao.org/ag/againfo/resources/en/publications.html
Animal feed quality is crucial in the livestock sector. This document presents the
sequence of activities for establishing a Feed Quality Analysis Laboratory from
initial planning, building and layout; through hiring suitable staff and selecting
methods and equipment; and culminating in accreditation, based on an
estimated four-year time frame.
The Quality Management System is stressed, and the document highlights
validation of methods, personnel and training; systematic equipment
maintenance and calibration; proficiency testing; quality control procedures;
reporting; and auditing.
ISBN 978-92-5-108071-9
7 8 9 2 5 1
ISSN 1810-0708
0 8 0 7 1 9
I3535E/1/11.13

Guidelines: The Feed Analysis Laboratory: Establishment and Quality Control

Uploaded by

Copyright:

Available Formats

Guidelines: The Feed Analysis Laboratory: Establishment and Quality Control

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Guidelines: The Feed Analysis Laboratory: Establishment and Quality Control

Uploaded by

Copyright:

Available Formats

ISSN 1810-0708

FAO ANIMAL PRODUCTION AND HEALTH

THE FEED ANALYSIS LABORATORY:

THE FEED ANALYSIS LABORATORY:

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS

Development of a business plan

2.2.1Type of laboratory to be developed

2.2.2Market analysis for potential customers and their needs

2.2.4Market analysis for available laboratory services

2.2.5Evaluation and decision making

Setting up and running the laboratory

3.2Physical realization of the laboratory

3.2.1The analytical process

3.2.2 Methods to be made operational

3.2.3Building and facilities

3.2.5Organizational structure and responsibilities of personnel

3.3Realization of the laboratory Procedures

3.4Continuity and improvement of the laboratory

Implementation of a Quality Management System and the

4.3Reading and interpretation of ISO/IEC 17025:2005.

4.4A road map for building a high quality system

4.5First situation: Routine stand-alone feed analysis laboratory

4.6Second situation:Routine laboratory connected to a feed

4.7Third situation:Laboratory as part of a research organization

4.8Fourth situation:Government or reference laboratories

Ensuring quality analytical performance

First line of quality control and the use of Shewhart charts

Calculation of uncertainty of measurement

An example of technical records for a determination

An example of a maintenance and calibration document

An example of a training record

Procedure for traceability of volumetric calibration

Atomic Absorption Spectroscopy

Certified Reference Material

Food and Agriculture Organization of the United Nations

FAO/IAEA Agriculture and Biotechnology Laboratory

Gas Chromatography-Flame Ionization Detector

Gas Chromatography-Mass Spectrometry

High Performance Liquid Chromatography

International Atomic Energy Agency

Inductively Coupled Plasma [Analysis]

Inductively Coupled PlasmaAtomic Emission Spectroscopy

International Laboratory Accreditation Cooperation

International Organization for Standardization

Liquid ChromatographMass Spectrometer

Laboratory Information Management System

Sequential mass spectrometry

Project Steering Group

Quality Management System

Research and development

Standard operating procedure

Quality Assurance Manager, Office of the Texas State Chemist,

Senior Quality Assurance Manager, SRUC, Ferguson Building,

Senior Researcher Feed Evaluation, Institute for Agricultural and

Animal Production Expert, Animal Production and Health

Professor, Animal Nutrition, Centre of Research and Services

Alicia Njera Molina

International Quality Assurance Manager, Masterlab bv, The

Animal Nutritionist, Animal Production and Health Section, Joint