ISO 13379-1:2025

International
Standard
ISO 13379-1
Second edition
Condition monitoring and
2025-10
diagnostics of machine systems —
Data interpretation and diagnostics
techniques —
Part 1:
General guidelines
Surveillance et diagnostic de l'état des systèmes de machines —
Interprétation des données et techniques de diagnostic —
Partie 1: Lignes directrices générales
Reference number
© ISO 2025
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Diagnostics and its relation to condition monitoring . 2
5 Requirements to set-up condition monitoring and diagnostics. 3
5.1 General .3
5.2 Establishing diagnostics needs .3
5.3 Failure mode symptoms analysis (FMSA) .4
5.3.1 General .4
5.3.2 Usage guidance .4
5.3.3 Rating guidance . .5
5.3.4 Assessing FMSA results using a monitoring priority number (MPN) .6
5.3.5 Assessing FMSA results using a diagram .7
5.4 Diagnostics requirements report .8
6 Elements used for diagnostics . 8
6.1 Condition monitoring data . .8
6.1.1 Parameters and measurements .8
6.1.2 Descriptors .9
6.1.3 Symptoms .9
6.1.4 Fault .10
6.2 Machine system data . .11
6.3 Maintenance data and events related to the machine system .11
7 Diagnostic approaches and models .11
7.1 Definition of diagnostic approaches .11
7.2 General guidelines for developing a diagnostic model . 12
7.3 Data-driven approach . . 13
7.3.1 General . 13
7.3.2 Building the model .14
7.3.3 Strengths and weaknesses .14
7.4 Knowledge-based approach. 15
7.4.1 Fault-symptom diagnostics . 15
7.4.2 Causal trees .16
7.4.3 First-principle models . .18
7.5 Confidence factor determination .19
Annex A (informative) Example of diagnostic report .20
Annex B (informative) Failure mode symptoms analysis (FMSA) worksheet .23
Annex C (informative) Examples of ratings used for failure mode symptoms analysis (FMSA) .25
Annex D (informative) Effectiveness of the diagnostics system.26
Annex E (informative) Description of selected methods used to build diagnostic models .28
Annex F (informative) Overview of diagnostic model applicability by monitoring technique .35
Annex G (informative) Example of bearing spalling modelled with a causal tree .36
Annex H (informative) Example of diagnosis confidence level determination .38
Bibliography .39

iii
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out through
ISO technical committees. Each member body interested in a subject for which a technical committee
has been established has the right to be represented on that committee. International organizations,
governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely
with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are described
in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types
of ISO document should be noted. This document was drafted in accordance with the editorial rules of the
ISO/IEC Directives, Part 2 (see www.iso.org/directives).
ISO draws attention to the possibility that the implementation of this document may involve the use of (a)
patent(s). ISO takes no position concerning the evidence, validity or applicability of any claimed patent
rights in respect thereof. As of the date of publication of this document, ISO had not received notice of (a)
patent(s) which may be required to implement this document. However, implementers are cautioned that
this may not represent the latest information, which may be obtained from the patent database available at
www.iso.org/patents. ISO shall not be held responsible for identifying any or all such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions
related to conformity assessment, as well as information about ISO's adherence to the World Trade
Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www.iso.org/iso/foreword.html.
This document was prepared by Technical Committee ISO/TC 108, Mechanical vibration, shock and condition
monitoring, Subcommittee SC 5, Condition monitoring and diagnostics of machine systems.
This second edition of ISO 13379-1 cancels and replaces the first edition (ISO 13379-1:2012), which has been
technically revised. The main changes are as follows:
— the Scope of the document has been extended by the addition of Clause 1 c);
— Clause 4 has been added to outline recommended steps to perform diagnostics;
— new methods for assessing the failure mode symptoms analysis have been added, see 5.3.4 and 5.3.5;
— new examples and descriptions of elements used for diagnostics have been added in Clause 6;
— information provided in 7.1, 7.3 and Annexes E and F has been updated to reflect the state of the art;
— descriptions of data-driven methods have been moved to (informative) Annex E;
A list of all parts in the ISO 13379 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.

iv
Introduction
Effective management of machine systems throughout their life cycles requires maintaining their
performance, reliability and availability. One of the key strategies to support this objective is condition
monitoring, which provides information on the state of the machine system.
Condition monitoring serves two principal roles:
a) to identify trends that indicate the remaining useful life of the machine system, deterioration of its
performance or increased risk of failures; and
b) to detect nonconformities, referred to as anomalies in the context of condition monitoring, by identifying
deviations from baseline values or expected operating conditions. Such deviations, when compared
against predefined criteria, can result in alarms.
Once an anomaly has been detected, it is often needed to identify its cause(s). Identifying the cause(s) of
the anomaly is referred to as diagnostics and supports the determination of appropriate corrective actions.
Stakeholders typically expect a certain level of accuracy in diagnostics, as its output — a diagnosis — can
directly influence machine system operation, maintenance planning and resource allocation. This document
supports users in developing diagnostic procedures and models, and in evaluating their confidence level,
applicability and limitations.

v
International Standard ISO 13379-1:2025(en)
Condition monitoring and diagnostics of machine systems —
Data interpretation and diagnostics techniques —
Part 1:
General guidelines
1 Scope
This document
a) establishes common concepts for condition monitoring and diagnostics of machine systems, simplifying
communication between the users and manufacturers of condition monitoring and diagnostics systems;
b) establishes technical characteristics and describes principles for condition monitoring and diagnostics
of machine systems;
c) gives guidance on developing condition monitoring and diagnostics systems; and
d) gives guidance on selecting an appropriate diagnostic approach in the particular application.
This document is applicable to any machine system whose state can be described by measuring or observing
its operational parameters (or inputs) and responses (or outputs).
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content constitutes
requirements of this document. For dated references, only the edition cited applies. For undated references,
the latest edition of the referenced document (including any amendments) applies.
ISO 13372, Condition monitoring and diagnostics of machines — Vocabulary
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO 13372 apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org
3.1
confidence level
figure of merit, e.g. percentage, that indicates the degree of certainty that the diagnosis or prognosis is
correct.
Note 1 to entry: This figure essentially represents the cumulative effect of error sources on the final certainty or
confidence in the accuracy of the outcome. Such a figure can be determined algorithmically or via a weighted
assessment system.
[SOURCE: ISO 13381-1:2025, 3.3, modified by adding ‘or’ and removing parentheses]

3.2
diagnostic approach
methodology used to perform diagnostics, typically by analysing observed or simulated data (data-driven
approach) or by explicitly representing fault or system behaviour (knowledge-based approach).
3.3
diagnostic model
mathematical or logical construct used to identify, analyse or interpret data and information for the purpose
of supporting diagnostics
4 Diagnostics and its relation to condition monitoring
Condition monitoring involves continuous or scheduled measurement of machine system parameters,
processing these measurements into data and information, and often the presentation and storage of the
processed data and information. Diagnostics is generally triggered by detecting an anomaly during routine
condition monitoring, routine analysis, random analysis or human perception. This detection is carried out
by making a comparison between the present descriptors of a machine system and baseline (also called
baseline values or reference sets) chosen from experience, from the manufacturer's specifications, from
commissioning tests or computed from statistical information (e.g. long-term average for a particular
operational state).
Diagnostics is a systematic approach used to identify the root cause of anomalies and plays an essential role
in decision making for operation and maintenance tasks. It should include the following steps:
a) Anomaly validation: At the start of diagnostics, it should be ensured that the anomaly has not been
caused by the malfunction of a transducer or any other part of the condition monitoring system.
b) Data and information collection: Next, relevant elements used for diagnostics should be gathered. These
elements are described in Clause 6 and include measurements, descriptors derived from measurements,
symptom and fault characteristics, records of the fault history, operation history and maintenance
history, and specific information that describes machine system construction and operational
parameters. ISO 17359 provides further guidelines for data and information collection.
c) Analysis: The gathered elements should be analysed to identify patterns, trends and deviations from
the baseline. The particular analysis method shall depend on the selected diagnostic approach. The
diagnostic approaches are described in Clause 7, together with guidelines for their selection.
d) Hypothesis formulation: Potential causes of the anomaly, also called hypothesis, should be reasoned
using the information generated by the analysis. When formulating the hypotheses, consideration
shall be given to the factors outlined in Clause 5. The hypothesis that no fault is present shall always be
considered.
e) Hypothesis testing: Hypotheses should be tested for by taking additional actions, such as making
targeted measurements or carrying out further tests, inspections or simulations. A hypothesis shall be
rejected if the test results contradict it. If the test results do not contradict the hypothesis, it shall be
considered not rejected. In cases where the test results provide a sufficiently high level of confidence, a
hypothesis may be considered accepted for the purpose of diagnosis synthesis.
f) Diagnosis synthesis: The diagnosis should be formulated using hypotheses that were not rejected. If
a higher confidence level is needed, only accepted hypotheses should be used. Further information on
determining diagnosis confidence level is given in 7.4.
g) Recommendation: Recommendations for taking corrective actions shall be formulated based on
diagnosis. The corrective actions can include maintenance tasks such as refurbishments, repairs
and replacements, adjustments of operational parameters or further monitoring. To develop
recommendations, the diagnostician should be aware of provisions given in Clause 5 and ISO 17359.
h) Follow-up: After implementing the recommended corrective actions, the machine system should be
further monitored to assess whether the expected behaviour has been restored or additional actions

are needed. If the recommendations made during diagnostics prove to be accurate when the follow-up
takes place, then this becomes useful knowledge that should be stored for future reference.
Diagnostics is iterative and carrying out multiple cycles through individual steps can be needed to arrive at
an accurate diagnosis. Diagnostics depends on constraints, such as financial, temporal or organisational, and
external dependencies which influence the feasibility of the individual steps. However, stakeholders should
be made aware that missing any of the steps listed in 4 a) to h) can reduce the accuracy of the diagnosis and
consequently, the confidence in the output of the diagnostics.
Diagnostics should be documented in the form of a diagnostic report. An example of the diagnostic report is
given in Table A.1.
Diagnostics may be used in parallel with prognostics. Guidance on how to develop and apply prognostics
and formulate prognosis (output of prognostics) is provided in ISO 13381-1.
5 Requirements to set-up condition monitoring and diagnostics
5.1 General
Condition monitoring and diagnostics have an essential role to play in decision making for operation and
maintenance tasks. To be effective, a condition monitoring and diagnostics system shall be set up according
to the faults that can occur in the machine system. Therefore, a preliminary study should be carried out
when preparing the requirements for the condition monitoring and diagnostics system of a particular
machine system.
5.2 Establishing diagnostics needs
A life cycle of the condition monitoring and diagnostics system is shown in Figure 1. The V-shape represents
the high-level (e.g. maintenance, machine system and risk assessment) and the low-level concerns (e.g.
measurements, monitoring, periodical tests and data processing).
The left branch of Figure 1 corresponds to the preliminary study, which prepares, for a particular machine
system, the necessary information for condition monitoring and diagnostics. The right branch corresponds
to the condition monitoring and diagnostics that are normally undertaken after the machine system has
been commissioned. Vertical layers depict individual phases of the life cycle.
Figure 1 — Typical life cycle of condition monitoring and diagnostics system

The generic steps of the preliminary study should include:
a) analysing the machine system availability, maintainability and criticality within the context of the
entire functional process;
b) listing major components and auxiliaries and their functions;
c) analysing components and auxiliaries failure modes and their causes;
d) expressing the failure mode criticality, taking into account the significance (safety, availability,
maintenance costs, production quality) and the likelihood of fault occurrence;
e) deciding in accordance with 5.2 d) which failure modes should be covered by diagnostics;
f) analysing operating conditions under which the individual faults can be best observed;
g) define the operating conditions for which the baselines should be established;
h) setting out symptoms that can help in assessing the condition of the machine system and that can be
used for diagnostics;
i) listing descriptors that can be used to evaluate (recognize) the different symptoms; and
j) identifying measurements to be taken from which the descriptors are derived or computed.
The steps given in 5.2 a) and b) are outlined in ISO 17359. The steps given in 5.2 a) to d) may be followed
using maintenance optimization such as failure modes and effects analysis (FMEA) or failure mode effects
and criticality analysis (FMECA), which are outlined in IEC 60812. They also may be accomplished within a
more general process of maintenance optimization, such as reliability-centred maintenance.
NOTE Guidelines for developing failure management policies using reliability-centred maintenance are set out in
IEC 60300-3-11.
The steps given in 5.2 c) to f) and 5.2 h) to j) may be followed using the methodology explained in 5.3.
5.3 Failure mode symptoms analysis (FMSA)
5.3.1 General
The purpose of the FMSA is to support the selection of monitoring techniques, technologies and strategies
to maximize the confidence level in the detection, diagnosis and prognosis of each identified failure mode.
FMSA focuses on the ability of a monitoring technique to detect symptoms produced by each identified
failure mode and assess their rate of change. When confidence in the sensitivity of a monitoring technique
and the resulting accuracy of diagnosis or prognosis is questionable, it is recommended that additional
monitoring technique(s), known as correlation monitoring technique(s), be used.
FMSA is a methodological extension of the FMEA and should therefore be used in conjunction with any
existing FMEA or FMECA that has already identified and ranked possible failure modes.
5.3.2 Usage guidance
The information needed to perform the FMSA is best compiled into a worksheet that should include:
a) list of the components and auxiliaries involved;
b) list of the possible failure modes for each component or auxiliary;
c) list of the effects of each failure mode;
d) list of the causes of each failure mode;
e) list of the symptoms produced by each failure mode;

f) list of the most appropriate monitoring techniques;
g) list of the estimated monitoring frequency;
h) rating of each failure mode by severity, detectability, diagnosis confidence level and prognosis
confidence level;
i) list of the most appropriate correlation monitoring techniques; and
j) list of the monitoring frequency for the correlation monitoring techniques.
For each item listed, the failure mode, effect, and cause should read logically across the worksheet. An
example of an appropriate FMSA worksheet is given in Table B.1.
The greatest difficulty is in establishing the correct terms for failure modes, effects, and causes. Care shall
be taken to avoid duplication of failure mode and cause on the same line of the worksheet. However, a term
can appear as a cause of failure in one line and as a failure mode in another line. A term can also appear as
an effect in one line when dealing with a component and as a failure mode when dealing with an assembly. It
may be helpful to remember that a failure mode can result in an effect due to a cause.
When considering monitoring strategies, these forms may also be used:
— a failure mode produces symptoms, which are best detected using a primary monitoring technique resulting
in a high diagnosis and prognosis confidence when monitored at a given monitoring frequency; and
— diagnosis and prognosis confidence level increases by using correlation monitoring techniques when
monitored at a given monitoring frequency.
Continuous reassessment should be carried out when experience with a newly installed machine system is
gained or when either the machine system or the condition monitoring system is modified.
5.3.3 Rating guidance
5.3.3.1 General
Ratings which estimate the failure severity (SEV), the likelihood of detection (DET), the diagnosis confidence
level (DGN) and the prognosis confidence level (PGN) should be assigned to each failure mode. All four
ratings can be combined in accordance with 5.3.4 to give a monitoring priority number (MPN). DET, DGN
and PGN can be combined in accordance with 5.3.5 to give a monitoring confidence number (MCN).
Note that there is no universal definition for individual ratings, as the scale for each rating depends on
specific needs of the user of this document. It is recommended to use a consistent scale for DET, DGN and
PGN ratings to ensure that these ratings hold equal weight when establishing MCN or MPN.
5.3.3.2 Failure severity (SEV)
SEV indicates the potential worst-case impact of a failure mode on the machine system or its function, where
lower values represent minor or negligible consequences and higher values reflect increasingly serious
consequences for safety, performance, or operation. The assigned SEV rating should align with the outcomes
of any performed FMEA or FMECA.
An example of the SEV rating using a scale of 1 to 5 is provided in Table C.1. Other examples can be found in
IEC 60812:2018, Annex B.
5.3.3.3 Likelihood of detection (DET)
DET indicates the likelihood of successfully detecting the symptoms of a failure mode, regardless of DGN
or PGN. Lower values represent a low likelihood, such as when symptoms are undetectable, unrepeatable,
not measurable in practice, or potentially masked by other failure mode symptoms. Higher values reflect a
greater likelihood that the failure mode can be reliably detected during monitoring or diagnostics.

An example of the DET rating using a scale of 1 to 5 is provided in Table C.2. Further examples of DET
ratings are provided in IEC 60812:2018, Annex B, which uses a decreasing scale for detectability rankings.
Therefore, when using DET values from any FMEA or FMECA performed in accordance with IEC 60812, the
scale shall be reversed to maintain consistency with this document.
5.3.3.4 Diagnosis confidence level (DGN)
DGN indicates the expected accuracy of identifying a specific failure mode once its symptoms have been
detected. Lower values represent a low diagnosis confidence level, such as when symptoms are ambiguous,
non-distinctive, unknown or unrepeatable with respect to the specific failure mode. Higher values reflect a
high confidence level.
An example of the DGN rating using a scale of 1 to 5 is provided in Table C.2.
5.3.3.5 Prognosis confidence level (PGN)
PGN indicates the expected accuracy of predicting the remaining useful life or the future fault or failure
probability associated with a specific failure mode. Lower values represent a low prognosis confidence level,
such as when symptoms are unrepeatable, insensitive to degradation progression, ambiguous, or when
failure rates are unknown. Higher values reflect a high confidence level. The prognosis confidence level can
also be influenced by the monitoring frequency, i.e. a higher prognosis confidence level can be expected with
the higher monitoring frequency.
An example of the PGN rating using a scale of 1 to 5 is provided in Table C.2.
5.3.4 Assessing FMSA results using a monitoring priority number (MPN)
MPN is typically defined as the product of four ratings: MPN = SEV × DET × DGN × PGN.
MPN is assigned to each combination of the failure mode and monitoring technique. The assessment of these
values shall consider the following:
a) The range of MPN values depends on the scales used to rate SEV, DET, DGN and PGN;
b) The numerical ratio between two MPN values has no specific meaning;
c) Low MPN values, in relation to the range defined in 5.3.4 a), do not imply that condition monitoring is not
necessary but rather that a low confidence level for detection, analysis, and prognosis can be expected
with the nominated monitoring technique and frequency; and
d) When two or more monitoring techniques are evaluated in relation to a single failure mode, the highest
MPN value indicates that the associated monitoring technique is the most appropriate for the detection,
diagnosis, and prognosis of that failure mode.
The assessment of whether the MPN value is sufficiently high for a single failure mode shall take into
consideration additional factors:
e) the safety risks associated with that failure mode;
f) the expected deterioration rate of that failure mode;
g) the mean time between failures for that failure mode;
h) the presence of secondary or subsequent failure modes;
i) interrelationships with other failure modes;
j) the maintenance lead time needed for corrective or preventive actions;
k) the availability and delivery time of spare parts; and
l) the targeted reliability and availability of the machine system.

5.3.4.1 Alternative method to calculate MPN
If the scales used to express SEV, DET, DGN and PGN are logarithmic, an alternative calculation of MPN may
be used: MPN = SEV + DET + DGN + PGN.
This alternative calculation reflects the additive nature of logarithmic scales, where multiplication of linear
values corresponds to addition in the logarithmic domain. The use of this method is particularly appropriate
when rating systems are designed to express orders of magnitude.
5.3.5 Assessing FMSA results using a diagram
This assessment method uses an FMSA diagram shown in Figure 2. The horizontal axis of the FMSA
diagram represents the range of possible SEV values, while the vertical axis represents the range of possible
monitoring confidence levels (MCN), which is typically defined as the product of three ratings: MCN = DET ×
DGN × PGN.
The FMSA diagram is divided into a set of regions. Typically, four regions, denoted as R1 to R4 in Figure 2,
are used; however, a different number of regions may be defined. The number of regions and the boundaries
between them shall be determined taking into consideration the scales used to rate SEV, DET, DGN and PGN,
as well as the factors listed in 5.3.4 e) to l). The boundaries between the regions may take the form of simple
lines, as shown in Figure 2 a), curves, as shown in Figure 2 b), or a combination of both. If a discrete scale is
used for any of the ratings, the use of step-wise boundaries, as illustrated in Figure 2 c), may be appropriate.
An example of how to construct an FMSA diagram is provided in Table C.3.
Each combination of the failure mode and monitoring technique is allocated to the FMSA diagram based on
the corresponding pair of SEV and MCN values. The assessment of this combination should take into account
the region in which it is located. The boundaries of the regions, as shown in Figure 2, reflect the principle
that failure modes with higher SEV typically require higher monitoring effectiveness to mitigate the risk of
undetected or late-detected faults.
Key
X SEV
Y MCN
Figure 2 — Examples of FMSA diagrams with four regions R1 to R4 delimited by a) linear
boundaries, b) boundaries formed by curves and c) step-wise boundaries
EXAMPLE An example of how the combination of the failure mode and monitoring technique can be assessed
depending on the region from Figure 2 in which it is located:
— R1 — The combination is not acceptable. In general, an alternative monitoring technique or additional correlation
monitoring technique(s) need to be used.
— R2 — The combination is acceptable in some cases, e.g. if the likelihood of fault occurrence is significantly lower
than once during the machine system life, or measures to reduce SEV are feasible. Otherwise, an alternative
monitoring technique or additional correlation monitoring technique(s) need to be used.

— R3 — The combination is acceptable.
— R4 — The combination is acceptable. Moreover, a less effective monitoring technique can be used or a correlation
monitoring technique(s) can be removed, provided that this does not adversely affect the assessment of other
failure modes.
5.3.5.1 Alternative method to calculate MCN
If the scales used to express DET, DGN and PGN are logarithmic, an alternative calculation of MCN may be
used: MCN = DET + DGN + PGN.
This alternative calculation reflects the additive nature of logarithmic scales, where multiplication of linear
values corresponds to addition in the logarithmic domain. The use of this method is particularly appropriate
when rating systems are designed to express orders of magnitude.
5.4 Diagnostics requirements report
It is recommended to store the synthesis of the preliminary study in a diagnostics requirements report. This
report should include:
a) the adopted breakdown of the machine system into components and auxiliaries;
b) failure modes associated with these components or auxiliaries;
c) potentially observable symptoms for each failure mode;
d) parameters that should be monitored;
e) measurements that should be taken to monitor the parameters; and
f) descriptors and algorithms used to derive or calculate the descriptors.
Not all critical faults can be detected by condition monitoring and there is a possibility that they are also not
diagnosable. Therefore, the diagnostics requirements report should clearly highlight which failure modes
are addressed and which are not. It should be reconsidered whether it is worth adding the ability to detect
or diagnose critical faults that cannot currently be detected or diagnosed.
Formally, the diagnostics requirements report should be composed of two parts:
— a machine system description corresponding to 5.2 a) and b); and
— list of failure modes, symptoms, descriptors and measurements that are to be used for diagnostics
corresponding to 5.2 c) to i), which may be realized using the FMSA worksheet given in Table B.1.
It is also recommended that the theoretical effectiveness of the diagnostics system is calculated. An example
of how to evaluate the effectiveness of the diagnostics system is provided in Annex D.
6 Elements used for diagnostics
6.1 Condition monitoring data
6.1.1 Parameters and measurements
In condition monitoring, a parameter denotes a specific quantity characterising the state or behaviour of
a machine system. A measurement is the result of acquiring the value of a parameter using a particular
measuring system at a defined location within the machine system. A measurement may also be obtained by
means of a virtual sensor, that is, a calculated or derived quantity based on one or more real measurements
or a processed part of a measured signal.

Monitored parameters are used in diagnostics for:
a) establishing descriptors; and
b) identifying the operating conditions at which the symptoms appear.
All the measurements used taken during monitoring are generally suitable for use in diagnostics.
ISO 17359:2018, Annex A, provides examples of specific parameters recommended for condition monitoring
of selected machine systems.
Care shall be taken when considering operational parameters. If a parameter serves as a descriptor or is
used in calculating a descriptor, it should be regarded as an output. If it characterises an operating condition,
it should be regarded as an input. Care shall be taken to distinguish descriptors from operating conditions,
as confusing the two may lead to incorrect diagnostic conclusions.
EXAMPLE The turbine body temperature is a descriptor when diagnosing the body. It becomes an operating
condition when diagnosing a bearing because it affects the work of the bearing but it is not affected by bearing faults.
6.1.2 Descriptors
Descriptors can be obtained from the condition monitoring system, either directly or after the processing
of the measurements. Descriptors are often used in preference to measurements for reasons of selectivity.
The more selective the descriptors are, the more selective the symptoms and therefore, the easier the
diagnosis. The descriptor selectivity reduces the number of possible fault hypotheses when trying to get
from symptoms to faults.
EXAMPLE 1 Descriptors that can be derived from waveforms measurements include:
a) statistical features of the waveform, such as statistical moments (e.g. mean, variance, skewness and kurtosis),
root mean square and shape factor (r.m.s. divided by the by the mean of the absolute value);
b) impulsive metrics, such as maximum absolute value of the signal (peak value), impulse factor, crest factor,
clearance factor and number of peaks per unit of time; and
c) signal processing metrics, such as signal-to-noise ratio, total harmonic distortion and signal to noise and
distortion ratio.
Examples of applying descriptors to specific waveform measurements include peak-peak amplitude of the first
harmonic of the shaft vibration displacement, root mean square (r.m.s.) of the bearing vibration velocity and crest
factor of the vibration acceleration and rotational speed.
EXAMPLE 2 Descriptors can also be assigned to measurements obtained from non-waveform sources, such as
image-based data or scalar measurements. Examples include lubricant total acid number derived from lubricant
chemical analysis, wear particle count, temperature gradient determined from a thermal image and mean brightness
level extracted from an image or its part.
6.1.3 Symptoms
A symptom should be expressed using the terms:
a) Timestamp: information, e.g. time and date, identifying when a symptom occurred or was observed.
The timestamp may be included in the header of a diagnostics report;
b) Location: where the symptom can be observed or measured in the machine system;
EXAMPLE 1 Shaftline at bearing No. 3 vertical direction, bearing pedestal No. 4, high-pressure body (front
left) and TE2SE05.
c) Time characteristic: the time constant of the descriptor evolution;
EXAMPLE 2 Abrupt, slow, permanent, 1h, 10 days.
d) Evolution characteristics: type of the descriptor evolution during the time characteristics;
EXAMPLE 3 Increase, decrease, stable, cyclic evolution.

e) Descriptor quantification: descriptor value or change of the descriptor value;
EXAMPLE 4 Presence, absence, dark (colour), >10, <200, 40.
f) Descriptor: the descriptor used including units, if applicable; and
EXAMPLE 5 Temperature in °C, substance colour, first harmonic of the vibration in μm (peak-peak).
g) Circumstance: operating conditions under which the symptom is seen.
EXAMPLE 6 During run down; within 1 h after cold start-up; at 100 % power; any circumstance.
When preparing the selection of symptoms for a fault, care shall be taken to avoid taking two or more
symptoms that are too dependent (highly correlated). Dependent symptoms are those that occur under
the same circumstances, share similar time and evolution characteristics and whose descriptor has been
obtained by measuring the same physical quantity. The evaluation of dependent symptoms does not provide
further information and, thus, does not allow the diagnostics to progress.
EXAMPLE 7 Examples of the properly described symptoms are (timestamps are omitted for simplicity):
— slow and regular evolution of first harmonic vector of shaft displacement at bearing No. 4 at nominal operation;
— temperature of bearing No. 2 is 10 °C above usual value at nominal operation;
— 2 mm/s instantaneous change in horizontal vibration of bearing pedestal No. 2 at nominal operation;
— unusual permanent noise near the high-pressure body at nominal operation; and
— dark colour of the lubricant oil that was collected during the nominal operation from sampling point No. 2.
6.1.4 Fault
A fault is a condition that occurs when a component or auxiliary degrades or exhibits abnormal behaviour
(anomaly). It should be expressed in the terms:
a) Timestamp: information, e.g. time and date, identifying when a fault occurred or was observed. The
timestamp may be included in the header of a diagnostics report;
b) Location: name or identifier of the component or auxiliary on which the fault occurs;
EXAMPLE 1 Bearing No. 3, shaft, piston, low pressure body, seal No. 2 and TE2SE05.
c) Description: characterisation of the component or auxiliary degradation; and
EXAMPLE 2 Incorrect clearance, cracks, unbalance, misalignment, opening of contacts, leakage.
d) Severity: numeric or categoric representation of the
...