ISO/IEC 8183:2023
(Main)Information technology — Artificial intelligence — Data life cycle framework
Information technology — Artificial intelligence — Data life cycle framework
This document defines the stages and identifies associated actions for data processing throughout the artificial intelligence (AI) system life cycle, including acquisition, creation, development, deployment, maintenance and decommissioning. This document does not define specific services, platforms or tools. This document is applicable to all organizations, regardless of type, size or nature, that use data in the development and use of AI systems.
Titre manque
General Information
Buy Standard
Standards Content (Sample)
INTERNATIONAL ISO/IEC
STANDARD 8183
First edition
2023-07
Information technology — Artificial
intelligence — Data life cycle
framework
Reference number
ISO/IEC 8183:2023(E)
© ISO/IEC 2023
---------------------- Page: 1 ----------------------
ISO/IEC 8183:2023(E)
COPYRIGHT PROTECTED DOCUMENT
© ISO/IEC 2023
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
© ISO/IEC 2023 – All rights reserved
---------------------- Page: 2 ----------------------
ISO/IEC 8183:2023(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols and abbreviated terms.1
5 Data life cycle overview . 1
6 Data life cycle framework .2
6.1 General . 2
6.2 Stage 1: Idea conception . 3
6.3 Stage 2: Business requirements . 4
6.4 Stage 3: Data planning . 4
6.5 Stage 4: Data acquisition. 5
6.6 Stage 5: Data preparation . 5
6.7 Stage 6: Building a model . 6
6.8 Stage 7: System deployment . 6
6.9 Stage 8: System operation . 7
6.10 Stage 9: Data decommissioning . 7
6.11 Stage 10: System decommissioning . 7
7 Stages and processes within the data life cycle . 7
Bibliography .10
iii
© ISO/IEC 2023 – All rights reserved
---------------------- Page: 3 ----------------------
ISO/IEC 8183:2023(E)
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are
members of ISO or IEC participate in the development of International Standards through technical
committees established by the respective organization to deal with particular fields of technical
activity. ISO and IEC technical committees collaborate in fields of mutual interest. Other international
organizations, governmental and non-governmental, in liaison with ISO and IEC, also take part in the
work.
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 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 or
www.iec.ch/members_experts/refdocs).
ISO and IEC draw attention to the possibility that the implementation of this document may involve the
use of (a) patent(s). ISO and IEC take 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 and IEC
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 and https://patents.iec.ch. ISO and IEC 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. In the IEC, see www.iec.ch/understanding-standards.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology,
Subcommittee SC 42, Artificial intelligence.
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 and
www.iec.ch/national-committees.
iv
© ISO/IEC 2023 – All rights reserved
---------------------- Page: 4 ----------------------
ISO/IEC 8183:2023(E)
Introduction
Artificial intelligence (AI) systems are being adopted by organizations of all types, sizes and purposes.
Data are essential to the development and operation of AI systems.
In the field of AI systems, there are many data life cycles in use and under consideration for different
purposes (e.g. data quality, bias in data, data governance, development and use of AI systems). Without
an overarching framework, these different data life cycles can be challenging to correctly interpret by
those without previous knowledge, context and expertise. There is a risk that these multiple data life
cycles will not be applied as intended.
This document provides a data life cycle overview in Clause 5, describes a data life cycle framework in
Clause 6 and provides more information on the stages or processes of the data life cycle in Clause 7.
v
© ISO/IEC 2023 – All rights reserved
---------------------- Page: 5 ----------------------
INTERNATIONAL STANDARD ISO/IEC 8183:2023(E)
Information technology — Artificial intelligence — Data
life cycle framework
1 Scope
This document defines the stages and identifies associated actions for data processing throughout the
artificial intelligence (AI) system life cycle, including acquisition, creation, development, deployment,
maintenance and decommissioning. This document does not define specific services, platforms or tools.
This document is applicable to all organizations, regardless of type, size or nature, that use data in the
development and use of AI systems.
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/IEC 22989, Information technology — Artificial intelligence — Artificial intelligence concepts and
terminology
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO/IEC 22989 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/
4 Symbols and abbreviated terms
AI artificial intelligence
DPIA data protection impact assessment
JSON JavaScript object notation
ML machine learning
OWL web ontology language
PII personally identifiable information
XML extensible markup language
5 Data life cycle overview
The data life cycle for AI systems encompasses the processing of data from the earliest conception of
a new AI system to the eventual decommissioning of the system and is separated into a number of
distinct stages. Each stage will often, but not always, be part of a data life cycle for an AI system.
1
© ISO/IEC 2023 – All rights reserved
---------------------- Page: 6 ----------------------
ISO/IEC 8183:2023(E)
A data life cycle represents all the stages through which data can pass within any system that uses data
of any kind. It is designed to support the achievement of objectives related to system governance, system
utility, data quality and data security, by ensuring that data processing is given due consideration
during the planning, development, use and decommissioning of the system.
The detailed purpose and timing of use of these stages throughout the life cycle are influenced by
multiple factors, including societal, commercial, organizational and technical considerations, each of
which can vary or at times be combined with other stages during the life of a system. This document
describes the following 10 stages:
— stage 1 – idea conception;
— stage 2 – business requirements;
— stage 3 – data planning;
— stage 4 – data acquisition;
— stage 5 – data preparation;
— stage 6 – building model;
— stage 7 – system deployment;
— stage 8 – system operation;
— stage 9 – data decommissioning;
— stage 10 – system decommissioning.
1)
For information about a data life cycle for data usage, see ISO/IEC 5212: — .
6 Data life cycle framework
6.1 General
The data life cycle framework, shown in Figure 1, identifies a set of conceptually distinct stages that
data used in an AI system go through from data planning to data decommissioning. Figure 1 also
includes idea conception, business requirements and system decommissioning, which are system-
level life cycle stages. For information regarding data sets, refer to ISO/IEC 23053:2022, 6.5. Life cycle
processes appropriate to a defined task can be assigned to each stage. Life cycle processes describe the
actions taken on the data within the life cycle stage.
Stage 9 (data decommissioning) and stage 10 (system decommissioning) both pertain to
decommissioning but stage 9 specifically covers what happens to the data (e.g. secure deletion,
archiving, repurposing) while stage 10 covers what happens to the system irrespective of what happens
to the data that is being processed.
1) Under preparation. Stage at the time of publication: ISO/IEC DIS 5212:2023.
2
© ISO/IEC 2023 – All rights reserved
---------------------- Page: 7 ----------------------
ISO/IEC 8183:2023(E)
Key
primary development pathway
feedback pathway
stage with data processing
stage outside the data processing boundary
NOTE 1 The single-headed arrows depict a linear path through the life cycle stages, while the double-headed
arrows show feedback paths between life cycle stages.
NOTE 2 The verification and validation of the model refers to the internal development process, whose output
is a model. The validation and verification of the system refers to the system as a whole, extending through its
entire period of operation.
Figure 1 — Data life cycle framework
6.2 Stage 1: Idea conception
Idea conception is when a need or requirement for a new or revised AI system is recognized. The AI
system can be used as a partial or complete solution to an existing or potential problem or opportunity
faced by the organization.
Idea conception can also be driven by broader organizational context needs (e.g. economic, technical,
strategic, market or legal requirements). Ultimately, this idea can be expressed as one or more questions
that the AI system can answer. The questions to which the AI system provides answers should be
mapped to and aligned with business objectives and metrics.
3
© ISO/IEC 2023 – All rights reserved
---------------------- Page: 8 ----------------------
ISO/IEC 8183:2023(E)
6.3 Stage 2: Business requirements
The business requirements stage can include one or more stakeholders with appropriate authority or
influence deciding to investigate whether the idea can be turned into a functioning system and deciding
whether to invest further in the idea. This stage involves:
— determining the ambition of the project (e.g. vision, goals and strategy);
— determining assets, including those available and those that need to be acquired;
— specifying the data requirements, a key element for AI systems, based on the business goals and end
user requirements;
— identifying enablers for the project, including in-house skills and knowledge, organizational
architecture, technology and external resources;
— ensuring the project can be developed in line with organizational policies and procedures (or
processes), including:
— compliance (e.g. privacy requirements);
— ethics (e.g. fairness of outcomes);
— culture;
— leadership;
— governance processes.
The business requirements stage can conclude with a determination of whether or not the project is
feasible.
NOTE The business requirements stage does not include processing of the data.
6.4 Stage 3: Data planning
The data planning stage involves deciding upon the scope of the data required to address the questions
identified in the busin
...
DRAFT INTERNATIONAL STANDARD
ISO/IEC DIS 8183
ISO/IEC JTC 1/SC 42 Secretariat: ANSI
Voting begins on: Voting terminates on:
2022-07-26 2022-10-18
Information technology — Artificial intelligence — Data
life cycle framework
ICS: 35.020
THIS DOCUMENT IS A DRAFT CIRCULATED
FOR COMMENT AND APPROVAL. IT IS
THEREFORE SUBJECT TO CHANGE AND MAY
This document is circulated as received from the committee secretariat.
NOT BE REFERRED TO AS AN INTERNATIONAL
STANDARD UNTIL PUBLISHED AS SUCH.
IN ADDITION TO THEIR EVALUATION AS
BEING ACCEPTABLE FOR INDUSTRIAL,
TECHNOLOGICAL, COMMERCIAL AND
USER PURPOSES, DRAFT INTERNATIONAL
STANDARDS MAY ON OCCASION HAVE TO
BE CONSIDERED IN THE LIGHT OF THEIR
POTENTIAL TO BECOME STANDARDS TO
WHICH REFERENCE MAY BE MADE IN
Reference number
NATIONAL REGULATIONS.
ISO/IEC DIS 8183:2022(E)
RECIPIENTS OF THIS DRAFT ARE INVITED
TO SUBMIT, WITH THEIR COMMENTS,
NOTIFICATION OF ANY RELEVANT PATENT
RIGHTS OF WHICH THEY ARE AWARE AND TO
PROVIDE SUPPORTING DOCUMENTATION. © ISO/IEC 2022
---------------------- Page: 1 ----------------------
ISO/IEC DIS 8183:2022(E)
DRAFT INTERNATIONAL STANDARD
ISO/IEC DIS 8183
ISO/IEC JTC 1/SC 42 Secretariat: ANSI
Voting begins on: Voting terminates on:
Information technology — Artificial intelligence — Data
life cycle framework
ICS: 35.020
COPYRIGHT PROTECTED DOCUMENT
THIS DOCUMENT IS A DRAFT CIRCULATED
FOR COMMENT AND APPROVAL. IT IS
© ISO/IEC 2022
THEREFORE SUBJECT TO CHANGE AND MAY
This document is circulated as received from the committee secretariat.
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
NOT BE REFERRED TO AS AN INTERNATIONAL
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on STANDARD UNTIL PUBLISHED AS SUCH.
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
IN ADDITION TO THEIR EVALUATION AS
or ISO’s member body in the country of the requester. BEING ACCEPTABLE FOR INDUSTRIAL,
TECHNOLOGICAL, COMMERCIAL AND
ISO copyright office
USER PURPOSES, DRAFT INTERNATIONAL
CP 401 • Ch. de Blandonnet 8
STANDARDS MAY ON OCCASION HAVE TO
BE CONSIDERED IN THE LIGHT OF THEIR
CH-1214 Vernier, Geneva
POTENTIAL TO BECOME STANDARDS TO
Phone: +41 22 749 01 11
WHICH REFERENCE MAY BE MADE IN
Reference number
Email: copyright@iso.org
NATIONAL REGULATIONS.
Website: www.iso.org ISO/IEC DIS 8183:2022(E)
RECIPIENTS OF THIS DRAFT ARE INVITED
Published in Switzerland
TO SUBMIT, WITH THEIR COMMENTS,
NOTIFICATION OF ANY RELEVANT PATENT
RIGHTS OF WHICH THEY ARE AWARE AND TO
ii
© ISO/IEC 2022 – All rights reserved
PROVIDE SUPPORTING DOCUMENTATION. © ISO/IEC 2022
---------------------- Page: 2 ----------------------
ISO/IEC DIS 8183:2022(E)
30 Contents
31 Introduction . v
32 1 Scope . 1
33 2 Normative references . 1
34 3 Terms and definitions . 1
35 4 Symbols and abbreviated terms . 1
36 5 Data life cycle overview . 2
37 6 Data life cycle framework . 2
38 6.1 General . 2
39 6.2 Stage 1: Idea conception . 3
40 6.3 Stage 2: Business requirements . 4
41 6.4 Stage 3: Data planning . 4
42 6.5 Stage 4: Data acquisition . 5
43 6.6 Stage 5: Data preparation . 5
44 6.7 Stage 6: Building a model . 6
45 6.8 Stage 7: System deployment . 7
46 6.9 Stage 8: System operation . 7
47 6.10 Stage 9: Data decommissioning . 8
48 6.11 Stage 10: System decommissioning . 8
49 7 Stages and processes within the data life cycle . 8
50 7.1 General . 8
51 Bibliography . 11
52
© ISO/IEC 2022 – All rights reserved iii
---------------------- Page: 3 ----------------------
ISO/IEC DIS 8183:2022(E)
53 Foreword
54 ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
55 Commission) form the specialized system for worldwide standardization. National bodies that are
56 members of ISO or IEC participate in the development of International Standards through technical
57 committees established by the respective organization to deal with particular fields of technical activity.
58 ISO and IEC technical committees collaborate in fields of mutual interest. Other international
59 organizations, governmental and non-governmental, in liaison with ISO and IEC, also take part in the
60 work. In the field of information technology, ISO and IEC have established a joint technical committee,
61 ISO/IEC JTC 1.
62 The procedures used to develop this document and those intended for its further maintenance are
63 described in the ISO/IEC Directives, Part 1. In particular the different approval criteria needed for the
64 different types of document should be noted. This document was drafted in accordance with the editorial
65 rules of the ISO/IEC Directives, Part 2 (see www.iso.org/directives).
66 Attention is drawn to the possibility that some of the elements of this document may be the subject of
67 patent rights. ISO and IEC will not be held responsible for identifying any or all such patent rights. Details
68 of any patent rights identified during the development of the document will be in the Introduction and/or
69 on the ISO list of patent declarations received (see www.iso.org/patents).
70 Any trade name used in this document is information given for the convenience of users and does not
71 constitute an endorsement.
72 For an explanation on the voluntary nature of standards, the meaning of ISO specific terms and
73 expressions related to conformity assessment, as well as information about ISO's adherence to the World
74 Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT) see the following
75 URL: www.iso.org/iso/foreword.html.
76 This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology,
77 Subcommittee SC 42, Artificial intelligence.
78 Any feedback or questions on this document should be directed to the user’s national standards body. A
79 complete listing of these bodies can be found at www.iso.org/members.html.
iv © ISO/IEC 2022 – All rights reserved
---------------------- Page: 4 ----------------------
ISO/IEC DIS 8183:2022(E)
80 Introduction
81 AI systems are being adopted by organizations of all types, sizes and purposes. Data is essential to the
82 development and operation of AI systems.
83 In the field of AI systems, there are any number of data life cycles in use and under consideration for
84 different purposes (e.g. data quality, bias in data, data governance, development and use of AI systems).
85 Without an overarching framework, these different data life cycles can be challenging to correctly
86 interpret by those without previous knowledge, context and expertise. There is a risk that these multiple
87 data life cycles will not be applied as intended.
88 The purpose of this document is to provide a data life cycle framework, including terms and concepts,
89 that can be referenced by specialized data life cycles. The aim is to make it easier for users in different
90 roles to understand and correlate specialized data life cycles and how they apply to their organization’s
91 needs by describing a set of high-level data life cycle stages.
© ISO/IEC 2022 – All rights reserved v
---------------------- Page: 5 ----------------------
ISO/IEC DIS 8183:2022(E)
92 Information technology — Artificial intelligence — Data life cycle
93 framework
94 1 Scope
95 This document is applicable to the data processing throughout the AI system life cycle including the
96 acquisition, creation, development, deployment, maintenance and decommissioning. This document is
97 applicable to the acquisition, creation, development, deployment, maintenance and decommissioning of
98 data in AI systems. This document does not define specific services, platforms or tools. This document is
99 applicable to all organizations, regardless of type, sizes and nature, that use data in the development and
100 use of AI systems.
101 2 Normative references
102 The following documents are referred to in the text in such a way that some or all of their content
103 constitutes requirements of this document. For dated references, only the edition cited applies. For
104 undated references, the latest edition of the referenced document (including any amendments) applies.
105 ISO/IEC 22989:-1), Information technology — Artificial intelligence — Artificial intelligence concepts and
106 terminology
107 3 Terms and definitions
108 For the purposes of this document, the terms and definitions given in ISO/IEC DIS 22989:— apply.
109 ISO and IEC maintain terminological databases for use in standardization at the following addresses:
110 — ISO Online browsing platform: available at https://www.iso.org/obp
111 — IEC Electropedia: available at http://www.electropedia.org/
112 4 Symbols and abbreviated terms
113 AI artificial intelligence
114 DPIA data protection impact assessment
115 JSON JavaScript object notation
116 ML machine learning
1) Under preparation. Stage at the time of publication: ISO/IEC FDIS 22989:2022
© ISO/IEC 2022 – All rights reserved 1
---------------------- Page: 6 ----------------------
ISO/IEC DIS 8183:2022(E)
117 OWL web ontology language
118 PII personally identifiable information
119 XML extensible markup language
120 5 Data life cycle overview
121 The data life cycle for AI systems encompasses the processing of data from the earliest conception of a
122 new AI system to the eventual decommissioning of the system and is separated into a number of distinct
123 stages. Each stage will often, but not always, be part of a data life cycle for an AI system.
124 A data life cycle represents all the stages through which data can pass within any system that uses data of
125 any kind. It is designed to support the achievement of objectives related to system governance, system
126 utility, data quality and data security, by ensuring that data processing is given due consideration during
127 the planning, development, use and decommissioning of the system.
128 The detailed purpose and timing of use of these stages throughout the life cycle are influenced by multiple
129 factors, including societal, commercial, organizational and technical considerations, each of which can
130 vary during the life of a system. This document describes the following 10 stages:
131 Stage 1 - Idea conception ;
132 Stage 2 - Business requirements ;
133 Stage 3 - Data planning;
134 Stage 4 - Data acquisition;
135 Stage 5 - Data preparation;
136 Stage 6 - Building model;
137 Stage 7 - System deployment;
138 Stage 8 - System operation;
139 Stage 9 - Data decommissioning;
140 Stage 10 - System decommissioning.
141 6 Data life cycle framework
142 6.1 General
143 The data life cycle framework, shown in Figure 1, identifies a set of distinct stages that data used in an AI
144 system go through from data planning to data decommissioning. Figure 1 also includes idea conception,
145 business requirements and system decommissioning which are system level life cycle stages.
2 © ISO/IEC 2022 – All rights reserved
---------------------- Page: 7 ----------------------
ISO/IEC DIS 8183:2022(E)
146 For information regarding datasets refer to FDIS ISO/IEC 23053:-2) [3]. Life cycle processes, appropriate
147 to a defined task, can be assigned to each stage. Life cycle processes describe the actions taken on the
148 data within the life cycle stage.
149 Stage 9 (data decommissioning) and stage 10 (system decommissioning) both pertain to
150 decommissioning but stage 9 specifically covers what happens to the data (e.g. deletion, destruction,
151 return) while stage 10 covers what happens to the system irrespective of what happens to the data.
152
153 Figure 1 — Data life cycle framework
154 NOTE 1 The single-headed arrows in Figure 1 depict a linear path through the life cycle stages while the double-
155 headed arrows show feedback paths between life cycle stages.
156 NOTE 2 The verification and validation of the model refers to the internal development process whose output is
157 a model. The validation and verification of the system refers to the system as a whole extending through its entire
158 period of operation.
159 6.2 Stage 1: Idea conception
160 Idea conception is when a need or requirement for a new or revised AI system is recognized. The AI
161 system can be used as a partial or a complete solution to an existing or potential problem or opportunity
162 faced by the organization.
2) Under preparation. Stage at the time of publication: ISO/IEC FDIS 23053:—
© ISO/IEC 2022 – All rights reserved 3
---------------------- Page: 8 ----------------------
ISO/IEC DIS 8183:2022(E)
163 Idea conception can also be driven by broader organizational context needs (e.g. economic, technical,
164 strategic, market or legal requirements). Ultimately, this idea should be expressed as one or more
165 questions that the AI system can answer.
166 6.3 Stage 2: Business requirements
167 The business requirements stage involves one or more stakeholders, with appropriate authority or
168 influence deciding 1) to investigate whether the idea can be turned into a functioning system and 2)
169 deciding whether to invest further in the idea. This stage involves:
170 determining the ambition of the project (e.g. vision, goals and strategy);
171 determining assets including those available and those that need to be acquired;
172 specifying the data requirements, a key element for AI syste
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.