BookPDF Available

Proceedings of the 17th International Conference "TRIZfest 2022"

Authors:
  • ICG Training & Consulting

Abstract and Figures

The collection of papers «Proceedings of the 17th MATRIZ TRIZfest-2022 International Conference». The conference is intended for TRIZ specialists and users: academics, engineers, inventors, innovation professionals, and teachers. The present book of Proceedings includes papers related to the research and development of TRIZ, best practices with TRIZ, cases of practical application of TRIZ, and issues of TRIZ training and education.
Content may be subject to copyright.
Organized by the International TRIZ Association MATRIZ
THE 17th INTERNATIONAL CONFERENCE
TRIZfest-2022
August 31 - September 1-3, 2022
CONFERENCE PROCEEDINGS
Editor: Valeri Souchkov
ISSN: 2374-2275
ISBN: 979-8-218-07055-7
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
2
Proceedings of the 17th MATRIZ TRIZfest-2022 International Conference
August 31 - September 1-3, 2022
Organized by the International TRIZ Association MATRIZ.
212 pages.
ISSN: 2374-2275
ISBN: 979-8-218-07055-7
Editor: Valeri Souchkov, TRIZ Master
The collection of papers «Proceedings of the 17th MATRIZ TRIZfest-2022 International Con-
ference».
The conference is intended for TRIZ specialists and users: academics, engineers, inventors,
innovation professionals, and teachers.
The present book of Proceedings includes papers related to the research and development of
TRIZ, best practices with TRIZ, cases of practical application of TRIZ, and issues of TRIZ
training and education.
Series “Proceedings of the TRIZfest International Conference”. ISSN: 2374-2275; ISBN:
979-8-218-07055-7
© Copyright: 2022, International TRIZ Association MATRIZ.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,
or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording
or otherwise, without the prior permission of the publishers.
Prepared and Published by the International TRIZ Association MATRIZ, 8946 Wesley Place,
Knoxville, TN 37922 USA
www.matriz.org
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
3
TRIZfest-2022 ORGANIZING COMMITTEE
Dr. Mark Barkan, TRIZ Master, MATRIZ Executive Director: The Organizing Committee
Chair.
Dr. Oleg Feygenson, TRIZ Master, MATRIZ President
Mr. Valeri Souchkov, TRIZ Master: Co-Chair of the Program Management Committee.
Dr. Simon Litvin, TRIZ Master: Chairman of the TRIZ Master Certification Council
(TMCC).
Dr. Elena Gredynarova, Level 3, founder and owner of school “Eidos”: The Chair of the Sec-
tion on TRIZ-Pedagogy
TRIZfest-2022 PAPERS REVIEW COMMITTEE
Dr. Robert Adunka, TRIZ Master, Germany
Mr. Dmitry Bakhturin, Russia
Dr. Mark Barkan, TRIZ Master, USA
Dr. Tiziana Bertoncelli, Germany
Mr. Christoph Dobrusskin, The Netherlands
Mr. Lorenzo Duroux, France
Dr. Oleg Feygenson, TRIZ Master, South Korea
Dr. Marta Gardner, USA
Ms. Barbara Gronauer, Germany
Dr. Pavel Livotov, Germany
Dr. Sergei Logvinov, Russia
Mr. Alex Lyubomirskiy, TRIZ Master, USA
Dr. Oliver Mayer, TRIZ Master, Germany
Mr. Horst Naehler, Germany
Dr. Toru Nakagawa, Japan
Mr. Alexey Pinyaev, TRIZ Master, USA
Mr. Tanasak Pheunghua, Thailand
Dr. Runhua Tan, People Republic of China
Dr. Stephane Savelli, Belgium
Mr. Valeri Souchkov, TRIZ Master, The Netherlands
Dr. Christian M. Thurnes, Germany
Mr. Hongyul Yoon, TRIZ Master, South Korea
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
4
Dear TRIZfest-2022 Participants and Readers,
It is a pleasure to present the papers from the 17th International Conference “TRIZfest-2022
which was held on August 31, September 1-3, 2022.
Due to pandemic, this time again it was decided to conduct the conference online.
This year the conference includes papers and presentations focused on the following topics:
Theoretical, research results.
TRIZ-related methods and tools development.
Best practices, business experiences, integration with non-TRIZ methods/tools.
TRIZ-Pedagogy
Educational methods and experiences.
Case studies.
TRIZfest-2022 continued its special section “TRIZ-Pedagogy”. The day of “TRIZ Pedagogy”
was very successful, attracting 870 participants. It was conducted in Russian only, therefore it
will be published a separate publication.
Main conference included discussions on several important topics regarding TRIZ and its ap-
plications. We would like to thank all the authors and co-authors who contributed their works
to include to these Proceedings and therefore provided considerable impact on further develop-
ment of TRIZ and its dissemination around the world.
We would like to express our sincere gratitude to all the members of the TRIZfest-2022 Organ-
izing Committee who provided their help and support as well as to the members of the Papers
Review Committee who invested their precious time to select the best papers and provide au-
thors with comments how to improve their papers.
And at last but not least, we would like to express our thanks to all the participants of the
conference from many countries who contributed to the event by their engagement and their
sometime provoking questions to the speakers.
Valeri Souchkov, TRIZ Master
Co-Chair of the TRIZfest-2022
Program Committee
Enschede, The Netherlands
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
5
TABLE OF CONTENTS
Sorted alphabetically
ABOUT THE NEED TO ELIMINATE ERRORS IN TRIZ
8
Victor Minaker
AN APPROACH TO RESOURCE ANALYSIS IN BUSINESS PROBLEM
SOLVING
13
Nikolay Saunin
AN OVERVIEW OF ADVANCED CECA METHOD
25
Jerzy Chrząszcz
APPLICATION OF TRIZ IN IPD
41
Jun Huang
APPLYING TRIZ TO FORECAST THE DEVELOPMENT OF CECA
50
Oleg Y. Abramov, Jerzy Chrząszcz
CASE STUDY: FRACTURE OF SHAFT HEAD OF CENTRIFUGAL
PUMP
64
Jing Wang, Ning Dang, Liwei Cong
CAUSE-EFFECT CHAIN ANALYSIS OF DIFFERENT TYPES OF
ENGINEERING PROBLEMS
70
Chuanwen Wang, Alp Lin, Dongshuang Xu, Claire Liu
DECISION-MAKING METHOD (DMM)
78
Victor Minaker, Mikhail Bykhovskiy
EMPLOYING TRIZ TOOLS TO BUILD “VIRTUAL ECOLOGICAL
NICHE” FOR EMERGING TECHNOLOGY INDUSTRIES
91
Zhang Ting
IDENTIFYING THE BEST TECHNOLOGY FOR SOLID WASTE
MANAGEMENT USING TRIZ-BASED TECHNOLOGY SCOUTING
102
Oleg Abramov, Evgenia Smirnova, Pavel Fimin
INFORMATION IN TRIZ
114
Li Haijun
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
6
INNOVATION DESIGN OF RANGE HOOD NOISE REDUCTION
BASED ON TRIZ THEORY
126
Feixiang Mei, Guoxin Cao, Yanglv, Jiancheng Zong
INNOVATIVE DESIGN OF INTELLIGENT WELDING SYSTEM FOR
PRESSURE VESSEL BASED ON TRIZ
138
Yan Zhao
ON SOME ASPECTS OF TRIZ FLOW ANALYSIS
150
Hans-Gert Gräbe
RESEARCH AND ANALYSIS ON THE INNOVATIVE INTEGRATION
OF TRIZ THEORY AND DFSS METHOD
160
Wei Tong (Tony), Yufu Xi, Bing Xu, Wei Shen, Yaoyuan Zhang, Wei Li, Min Wen,
Lan Zhang
SUSTAINABILITY MODELLING WITH TRIZ
169
Oliver Mayer
TECHNICAL CONTRADICTION ANALYSIS AS A CONVENIENT
TOOL TO RESOLVE CONTRADICTIONS IN EXPRESS WAY
177
Anton Kozhemyako
THE INNOVATIVE APPLICATION OF TRIZ THEORY IN
AUTOMOBILE INDUSTRY
184
Wei Tong (Tony), Jiapeng Nie, Tian Xing, Jianyi Song, Junjie Ren, Fei Liao,
Zhuo Chen, Sheng Li
TRIZ INSPIRED CREATION OF DIGITAL BUSINESS MODELS
Stefan Schaper
191
TRIZ REVERSE – CASE STUDY IN THE FIELD OF BASIC RE-
SEARCH OF PHYSICS
199
Silvia Liubenova Popova, Swen Günther
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
7
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
8
TRIZfest-2022
About the need to eliminate errors in TRIZ
Victor Minaker
Master of TRIZ, Russia
Abstract
This article discusses the problem of errors in TRIZ. Examples of errors are given and one of the most
important errors is considered, which consists in mixing the concept of technical contradiction and the
concept of alternative technical contradiction.
Keywords: TRIZ, technical system, function, functional analysis, function ranking, harmful functions,
importance of functions, auxiliary functions, dialectical contradiction, technical contradiction, alterna-
tive technical contradiction, physical contradiction, ARIZ, multi-criteria ranking, integral estimates.
1 Introduction
The process of renewal and development of TRIZ has not stopped since its birth in the middle
of the last century. The process of creating something new involves making and correcting
mistakes. Therefore, many erroneous statements must have accumulated in TRIZ.
TRIZ is a set of rules, in particular models and algorithms for the use of knowledge. The effec-
tiveness of TRIZ application is determined by the volume, detail, and correctness of these rules.
Using TRIZ, a person performs mental work. Therefore, success or failure is also achieved by
him. If the rules contain mistakes, then a person achieves success despite these mistakes. In this
case, the rules will make it harder, not easier, to find successful solutions.
If mistakes are identified and corrected, then TRIZ will become a science to a greater extent,
and if mistakes are hushed up, and everything created in TRIZ is declared infallible, then it will
gradually approach religious dogma.
Unfortunately, in TRIZ it is difficult to rely on practice as a criterion of truth. Consequently, in
TRIZ misconceptions can exist for a very long time, and new false ideas can spread very widely.
Therefore, it seems to me very important to work on eliminating old and new errors in TRIZ.
Unfortunately, this work has not been carried out intensively enough for a long time. Most of
the new TRIZ members most often do not even suspect that in many cases they use erroneous
recommendations or recommendations with significant gaps. In my opinion, this is one of the
reasons why TRIZ is not applied effectively enough.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
9
It seems to me that quite a lot of small and not very small errors have accumulated over the
years of TRIZ development. Misconceptions can also be found in MATIZ documents, for ex-
ample, in the Glossary. At the same time, sometimes an error that seems small, after a profound
analysis and an attempt to correct it, turns out to be fundamental.
When errors are detected, it becomes possible to eliminate them, i.e., to obtain a new correct
result. If the errors are not detected, the correct result will not appear.
2 Examples of errors
I refer to minor mistakes as the use of terms that are not consistent with the rules of the language
or generally accepted terms in science. An example of a minor error is the term "function car-
rier". Function: what somebody or something is there to do; a specific action that a system can
perform [1]. In TRIZ, this is the action of changing or artificially stabilizing the value of a
function object parameter. [2] Taking this into account, the term function carrier cannot be
considered successful. The action can be performed, but it cannot be carried, so instead of the
term "function carrier" it is better to use the term "function performer". Another similar example
is the use of the terms chemical, biological, etc. field, since there are no such concepts in the
relevant scientific disciplines. [2]
An example of a not very small error is the rule of ranking auxiliary functions during functional
analysis in accordance with their importance. The function ranking procedure includes two un-
proven hypotheses. One of them suggests that it is advisable to rank functions according to their
importance for the functioning of the system and this can be done with sufficient accuracy.
Another hypothesis suggests that the importance of functions is less, the further the function
object is located from the object of the main function in the functional model. In the generally
accepted understanding, if something is important, then it is of great significance to meet some
need or achieve some result. Thus, when stakeholders consider one function more important
than another, they mean the level of satisfaction of the need or the approach to the result, or the
risk of their dissatisfaction or non-achievement. If we take this view as a basis, then the func-
tional analysis will need to be supplemented with a risk analysis. To what extent such an addi-
tional analysis will be justified if it is associated with a functional model is still unclear, it is
likely that risk analysis can be based on parametric trees with great success. But the position of
the function object in the functional model has nothing to do with the importance of the func-
tion. The question may arise whether the total result of functional analysis will not be reduced
too much if functions are not ranked according to their importance? In my opinion, in cases
where the identification of harmful and inadequately performed useful functions gives an in-
significant result, functional analysis contains additional capabilities.
For example, it allows you to additionally identify:
• useful functions whose parameters are insufficiently defined,
• useful functions, the values of which have conflicting requirements,
• useful functions that are only needed in the auxiliary stages,
useful functions whose parameter values for their adequate performance at auxiliary stages
should be higher than at the use stage.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
10
And, of course, in functional analysis, no one forbids the classification of harmful functions,
like the classification of useful functions into adequately and inadequately performed. Moreo-
ver, such a classification of harmful functions is necessary to identify aggravated and non-ag-
gravated technical contradictions.
I consider the confusion of the concept of technical contradiction and comparison of different
systems to be examples of major mistakes, as well as the choice of an alternative based on
multi-criteria ranking using integral estimates. The second of these two examples is considered
in a separate paper. Here we will consider the first example with the formulation of a technical
contradiction.
3 About contradictions
The concept of technical contradiction in TRIZ emerged from the philosophical concept of di-
alectical contradiction. By analogy with the dialectical contradiction, the technical contradic-
tion was a special combination of properties of a technical system, manifested in the process of
its evolution. [3] But attempts to make ARIZ as universal as possible have led to unnecessarily
broad and vague formulations of technical contradiction. At the same time, a technical contra-
diction from a special combination of system properties began to turn simply into a description
of a certain step of the algorithm for solving an inventive problem. In these formulations, vari-
ants have appeared that relate not to one system, but to several. [4-6] This could not but lead to
errors in solving problems.
Later in TRIZ, a clear concept of physical contradiction was developed. This concept would
have to return the original understanding of the technical contradiction as a special combination
of the properties of the system, and pay attention to the formulations relating to different sys-
tems. But, this did not happen.
After the emergence of a clear concept of physical contradiction, the vagueness of the formu-
lations of technical contradiction became noticeable. In this connection, B. Goldovsky made an
attempt to make the formulation of technical contradiction more formalized and to return to
technical contradiction its original meaning of a special combination of system properties. [7]
This undoubtedly made a big difference. It was important, in particular, that as a result, for the
first time it became possible to identify non-obvious technical contradictions using cause-and-
effect diagrams. However, in Goldovsky's formulations, variants relating to different systems
have been preserved.
Subsequently, V. Gerasimov and S. Litvin proposed and described in detail the method of trans-
ferring properties. [8] In this paper, several concepts new to TRIZ were proposed, in particular
the concept of an alternative technical contradiction. At the same time, it was emphasized that
the alternative contradiction in its formulation differs in that the indicators of different systems
are compared, and the usual technical contradiction is inherent in one technical system. Also,
the sets of compared indicators differ in the usual and alternative technical contradictions. In
this article, in particular, attention was drawn to the fact that in ARIZ, when formulating a
contradiction with a missing useful-harmful element, we are not talking about an ordinary, but
about an alternative technical contradiction. However, in many TRIZ publications that appeared
later, the formulations of the "usual" technical contradiction are still confused with variants of
comparing the properties of different systems. It is often heard that since the same set of deci-
sive TRIZ tools (inventive principles, etc.) is used to eliminate the usual and alternative contra-
dictions, then this mixing of different types of contradictions does not matter. One cannot agree
with this, in particular, because sets of indicators that do not coincide in them lead to different
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
11
trajectories and results of the search for solutions. For example, when the problem is that a
technical system reaches the limit of development, it is necessary to use the transfer of proper-
ties and alternative contradictions. And when the problem lies in an aggravated contradiction,
and the limit of development may still be far away, it is necessary to use the apparatus of elim-
inating technical and physical contradictions. In this context, the analogy with surgical opera-
tions and sets of surgical instruments is quite legitimate. The set of surgical instruments is the
same, and the surgical operations of various diseases are different.
It should be noted that despite Goldovsky's inclusion of variants related to different technical
systems in the formulation of the "usual" technical contradiction, his works contain very pro-
found remarks that are necessary for a correct understanding of issues related to the concept of
technical contradiction. [9-11]
In this regard, I proposed to narrow down the wording of the usual technical contradiction using
only parameters. In this narrowed version, the technical contradiction is formulated as follows:
"A combination of dependencies of the parameters of a technical system, in which a change in
the magnitude of one of its parameters leads to an improvement in one parameter of a technical
system or a supersystem and a deterioration in another parameter of a technical system or a
supersystem." By improving and deteriorating parameters, we mean parameters that are im-
portant for stakeholders. At the same time, not only the parameters of the technical system itself
can deteriorate or improve, but also the parameters of the supersystem associated with these
parameters (for example, the free space in the room around the table, which depends on the size
of this table). This modification of the formulation of a technical contradiction leads to the fact
that it can be detected in a rather formal way using parametric trees. [12] At the same time, it
is important that not only aggravated contradictions are revealed when undesirable effects are
obvious, but also not aggravated contradictions when undesirable effects are not obvious. I re-
gret to note that most of my TRIZ colleagues still do not see the difference between a technical
contradiction and a comparison of the properties of different systems.
4 Conclusions
In addition to methodological errors in publications on TRIZ, a large number of examples of
allegedly successful solutions have accumulated, which in reality are erroneous and create an
extremely negative attitude towards TRIZ among specialists. This naturally happened for sev-
eral reasons, in particular because the patent fund is used as one of the main sources of infor-
mation in TRIZ. As you know, not all technical solutions for which patents are issued are suc-
cessful. [13] There are quite a lot of erroneous ones among them. But the use of patent databases
as one of the main sources of information in TRIZ is not the only reason for a large number of
erroneous examples of the use of its tools. Insufficient in-depth knowledge in specific areas, of
course, also plays a role. [14] However, methodological errors significantly increase the sever-
ity of these problems. It is obvious that all these problems need to be solved. It is impossible to
solve these problems without their open recognition. In particular, it is necessary to speak
openly about specific unsuccessful technical solutions from publications on TRIZ.
Acknowledgements
I express my gratitude to V. Fey, A. Kudryavtsev, S. Litvin, S. Logvinov, A. Lyubomirskiy, V.
Souchkov and A. Zusman for participating in the discussion of issues that resulted in this article.
I additionally thank V. Sushkov for editing the English version of this article.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
12
References
1. https://en.wikipedia.org/wiki/Function
2. https://matriz.org/wp-content/uploads/2012/10/TRIZGlossaryVersion1-2.pdf
3. https://www.altshuller.ru/triz/triz12.asp
4. https://altshuller.ru/triz/ariz77.asp
5. https://altshuller.ru/triz/ariz82b-2.asp
6. https://altshuller.ru/triz/ariz85v-1.asp#2
7. Goldovsky B.I. On contradictions in technical systems 2. http://www.metod-
olog.ru/00001/00001.html
8. V. M. Gerasimov, S. S. Litvin. Why technology needs pluralism (DEVELOPMENT OF ALTER-
NATIVE TECHNICAL SYSTEMS BY COMBINING THEM INTO A SUPRASYSTEM) TRIZ
Magazine vol. 1, N 1`90, https://metodolog.ru/00594/00594.html
9. Goldovsky B.I. Theses in defense of contradictions in technical systems (3). https://www.metod-
olog.ru/node/1827
10. Goldovsky B.I. Some comments on the heuristic possibilities of contradiction in a technical sys-
tem. https://www.metodolog.ru/node/1949
11. Goldovsky B.I. On the laws of building technical systems. https://www.metodolog.ru/node/2164
12. Minaker V.E. Technical contradiction and parameter analysis. Collection of reports of the XVIII
International conference "TRIZ: The practice of application and development of methodological
tools". Moscow, November 11-12, 2016
13. G. Stevens and J. Burley, “3000 Raw Ideas = 1 Commercial Success!” Research•Technology
Management, 40(3): 16-27, May-June, 1997.
14. Cooper R. «Winning at New Products» (Accelerating the Process from Idea to Launch).
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
13
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
14
1
The author deliberately describes the example superficially, without going into detail, as it's provided
here as a general illustration of the principle.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
15

 


 



Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
16
 


Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
17












  

Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
18
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
19
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
20
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
21
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
22
2
Case provided by D. I. Pravkin
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
23
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
24
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
25
TRIZfest-2022
AN OVERVIEW OF ADVANCED CECA METHOD
Jerzy Chrząszcz
Institute of Computer Science, Warsaw University of Technology, Poland
Pentacomp Systemy Informatyczne S.A., Warsaw, Poland
Abstract
This paper briefly presents conceptual extensions to the Cause-Effect Chains Analysis (CECA) method
developed by the author and jointly dubbed Advanced CECA (ACECA). They pertain to model quality
assessment, quantitative parametrization of disadvantages, and support for selecting key disadvantages.
Additionally, some functions of prototype software applications are described using a cause-effect dia-
gram coming from a real-life project as an example.
The quality assessment approach proposes criteria for evaluating model correctness and to some extent
its completeness. The classic CECA approach uses strictly qualitative cause-effect modeling, followed
by assessing and ranking the importance of disadvantages and expected challenges of their removal.
Proposed quantitative extensions allow for explicit representation of the unequal impact of disad-
vantages reflected in the cause-effect model. The presented approach is based on risk management rules,
and it is designed to be easily implementable in software.
Selection of key disadvantages is supported by using the expanded and reduced logical expressions,
quantitative model extensions and a procedure of aggregating the impact of particular disadvantages.
The calculations take into account the interconnections and logical operators involved to indicate the
cumulated impact of candidate key disadvantages. Depending on model parameters, the forecasted prof-
itability of removing given disadvantages may also be reported.
Keywords: TRIZ, Cause-Effect Chains Analysis, CECA, hazard, vulnerability, risk management.
1 Introduction
Cause-Effects Chains Analysis (CECA) method was developed in GEN3 in the 1990s, and it is
used within the TRIZ community worldwide [2, 3]. The procedure starts with determining un-
desirable effects that should be eliminated from the analyzed system, which are called target
disadvantages. They are derived by inverting the goals of the project. For instance, if the goal
is described as "to increase the flight range of the drone", the target disadvantage should be
formulated as "the flight range of the drone is insufficient". Then the subsequent causes of
particular effects are identified to build chains of intermediate disadvantages until reaching a
cause perceived as remaining beyond control, which usually reflects a law of nature or a busi-
ness constraint. Such a cause is considered a root cause and stops the analysis of a given chain.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
26
The analysis is usually documented with a diagram consisting of linear chains of boxes describ-
ing the disadvantages, connected by arrows indicating the flow of causality. The chains, in turn,
may connect on inputs by common causes or on outputs by logical operators.
When all the input causes are jointly required to trigger a given effect, the operator is indicated
as AND (logical conjunction). If the effect may be triggered by any of the input causes acting
alone, the operator is indicated as OR (logical disjunction), or the operator is not used, and all
the input causes are directly connected with the effect to reflect its alternative triggers. The
model may also contain loops of causes, representing feedback paths identified in the system.
Although the root causes cannot be directly eliminated, as a rule, they indicate the main factors
influencing the system and also define the boundaries of the model. The expected outcome of
the CECA process is to indicate a set of key disadvantages, the elimination of which would
result in the removal of the target disadvantages specified for the project. Key disadvantages
reflect deep, primary causes, having a large impact on the target disadvantages due to their
nature or location in the model. After transforming key disadvantages into key problems, they
are expected to inspire strong solution ideas.
A wide overview of the TRIZ literature related to cause-effect analysis was provided in [36].
The methods proposed in the mentioned works seem to fall into a few main areas (see Fig. 1):
introducing additional constraints without changing the original perspective [3, 14, 17],
introducing patterns or statistics coming from practical experience [6, 13],
introducing additional object types into the model [5, 7, 16, 27, 32, 35],
introducing additional rules or criteria [4, 5, 9, 11, 12, 16, 19, 20, 22, 23, 26, 36],
introducing elements of quantitative analysis [15, 16, 19, 27, 32, 34],
supporting selection of key disadvantages [4, 9, 23, 26, 28, 31],
changing modeling paradigm [5, 7, 8, 10, 18, 19, 24, 25, 29, 30, 31, 33, 35].
The aspects of model correctness or completeness appear in several works, but they are usually
described as examples, remarks, or guidelines rather than explicit criteria which could be as-
sessed in a systematic way.
Fig. 1. Cause-effect analysis in TRIZ literature overview 1995-2020
The proposed extensions address several known CECA shortcomings. The first one, apparently
common to all methods of cause-effect analysis, is the uncertainty about the completeness and
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
27
correctness of the model. Usually, there is no way to be sure that a given disadvantage cannot
be triggered by another cause or a specific combination of causes, which are not covered in the
model. A typical approach to alleviating this problem is to perform the analysis within a team
comprising methodical experts (to ask appropriate questions) and subject matter experts
(to provide competent answers). This way, the risk of overlooking known factors implying sig-
nificant consequences may be decreased to a level, which makes the uncertainty acceptable.
A lot was written in the TRIZ literature about selecting "the right problem" for solving [1, 6, 9,
16, 20, 23, 26], and to support the selection of such a problem, we need "the right model", which
justifies research on quality criteria applicable to CECA models [36]. The TRIZ users would
benefit from practical criteria for evaluating the quality of models during their creation, espe-
cially performed in an automated manner.
Unfortunately, the recommendations regarding the correctness and completeness of a model,
considered by several authors [9, 11, 12, 16, 18, 21], require understanding the meaning of the
disadvantages' descriptions, being language-specific and domain-specific. Therefore, the most
promising semantic level of quality assessment was considered beyond reach, and the research
described below focused on structural model correctness. One of the most important extensions
to the classic CECA method is the condition-action approach [9, 14]. This approach also
described as parameter-function allows detecting a missing link if the analysis yields two
disadvantages of the same type in a row, thus supporting the correctness and completeness of
the model. This postulate normalizes the content of a model, but it does not affect the structure,
and therefore it cannot be evaluated using a structural approach.
Original CECA is a strictly qualitative method, which implies that all disadvantages are re-
flected in the model as if they were equally important. However, it is widely known that some
factors are more frequent or more impactful than others in the real world. This observation led
to the formulation of the Pareto principle, stating that for many events, 80% of effects come
from 20% of the causes. Because the CECA method provides no direct support to distinguish
between more significant and less significant disadvantages (including key disadvantages),
such differentiation must be provided "externally" by scoring or ranking.
This paper recalls, extends, and integrates several concepts of enhancements to the CECA
method presented in the previous author's publications, aimed at:
assessing the quality of models and indicating specific errors,
complementing qualitative approach with quantitative parameters,
supporting the selection of the most promising key disadvantages, and
developing data structures and algorithms.
2 Previous work
The first ideas of the whole research were articulated in [18], and they primarily addressed the
duality of the context-specific content of a model (box descriptions) and context-independent
structure, which may be described with Boolean algebra. This approach assumes that logical
(binary) variables are appropriate to represent active and inactive disadvantages as 1 and 0,
respectively. At the structural level, the model may be described as input variables particularly
key disadvantages affecting target disadvantages through a network of AND/OR operators,
which corresponds to a combinational logic circuit. Therefore, it was proposed to analyze the
model using methods originating from the digital design domain. One of such methods is logical
minimization to reduce a circuit to its minimal configuration providing equivalent function, and
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
28
this technique may be used to select the most impactful key disadvantages. Another method is
using De Morgan's laws to find inverted logical functions reflecting combined partial solutions.
In the second stage of development, this logical approach to CECA modeling was merged with
quantitative model parametrization to reflect the unequal impact of disadvantages and, possibly,
unequal profitability of candidate solutions if the cost of removing particular disadvantages
may be reasonably estimated [19, 24]. Using quantitative estimations to support decisions re-
garding the elimination of unwanted effects was found to be similar to the risk management
process. Indeed, the disadvantages and candidate solutions may be perceived as the counterparts
of risk factors and mitigation activities, respectively.
A general operating rule states that elimination of the influence of a given disadvantage in a
linear chain may be achieved either directly, by removing this disadvantage, or indirectly by
eliminating any of the preceding disadvantages in the chain. This duality leads to a conflict:
there is only one preceding cause indicated in the model for a given disadvantage (so that it is
supposed to control the manifestation of this disadvantage exclusively), while it should also be
possible to remove this disadvantage without eliminating its sole cause (especially a root cause,
usually considered to be irremovable). Hence, the physical contradiction regarding the removal
of disadvantages in linear chains (excluding root causes) may be described as follows:
each disadvantage should have exactly one control to preserve coherence between the model
and the diagram (as we only draw one arrow between the boxes), and
each disadvantage should have more than one control to preserve coherence between the
model and the concept (as we may remove a disadvantage without deleting its predecessor).
Solving this contradiction resulted in distinguishing hazards (external factors triggering the un-
wanted effects) against vulnerabilities (internal properties blocking or allowing the hazards to
affect the system), which confirmed similarities between cause-effect analysis and risk analysis.
These similarities were combined with the quantitative model parametrization to build a proce-
dure supporting the selection of key disadvantages and candidate solution ideas using an ap-
proach adapted from the risk management area (see Table 1 below). The concept of using vul-
nerabilities to inspire candidate solutions was further explored in [24].
Table 1. Generic risk management process and its projection on dealing with key disadvantages coming
from CECA and candidate solutions to key problems derived from these disadvantages
Risk management process
Quantitative CECA extensions
1. identify risk factors,
1. identify key disadvantages (CECA ends here),
2. estimate initial risk magnitude,
2. estimate initial impact of key disadvantages,
3. select risks from ranked results,
3. select key disadvantages from ranked results,
4. analyze risk mitigation options,
4. analyze solution options,
5. estimate residual risk magnitude,
5. estimate residual impact of key disadvantages,
6. estimate cost of risk mitigation,
6. estimate cost of key disadvantage mitigation,
7. calculate profitability of mitigation options,
7. calculate profitability of solution options,
8. select mitigation options.
8. select solution options.
The basic concept of supporting the selection of key disadvantages by analyzing logical func-
tions describing model structure was developed in three directions: towards the calculation of
impact coming to target disadvantages from particular key disadvantages [28], expansion of the
logical expressions [31], and switching to state-machine representation [25, 30, 33]. To support
the automation of these operations in software, a dedicated data structure was developed, called
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
29
causality matrix, providing a computer-friendly representation for CECA models [34]. It stores
information about nodes and edges of the diagram graph as well as other attributes, including
the estimated cost of removing particular disadvantages.
Every disadvantage in the model is allocated a row and a column with the same index, defining
a square data table. Every data row indicates which causes directly influence a particular effect
and which logical operator (AND / OR / none) models their relation in the diagram. The influ-
ence of a particular cause is indicated by a non-zero value (weight) in a respective column of
the data row describing the given effect. The weights reflect the relative influence of the con-
tributing causes, and if no information is available about the differences, the influence is pre-
sumed to be equal and indicated as 1 for each cause. Additional rows and columns store de-
scriptions of disadvantages, counters of causes/effects, sums of weights, and other attributes.
The third part of this research addressed the problem of CECA model quality, and its outcome
was identifying reasonable quality criteria and feasible ways of applying them [36]. Following
the rules stated in [2], the CECA models contain boxes reflecting disadvantages, arrows indi-
cating causality flow from causes to effects, AND operators reflected with explicit objects, and
OR operators shown as explicit objects or implied by converging arrows. An arrow may connect
two boxes, two operators, a box with an operator or an operator with a box. Proposed quality
criteria referred to the numbers of the edges coming to and from a node of a specific type.
Possible interconnection variants are described below in Table 2 and Table 3; "2+" stands for
"2 or more", and "X" stands for "any number", which is equivalent to "don't care".
Table 2. Box type and correctness as a function of the number of inputs and outputs if AND and OR
operators are explicit objects (adapted from [36])
in
out
status
description
0
0
incorrect
error: an isolated object
0
1
correct
a root cause
0
2+
correct
a common root cause
1
0
correct
a target disadvantage
1
1
correct
an intermediate disadvantage possibly a linear chain segment
1
2+
correct
a common intermediate cause
2+
X
incorrect
error: no more than one input is allowed
Table 3. Operator type and correctness as a function of the number of inputs and outputs if AND and
OR operators are explicit objects (adapted from [36])
in
out
status
description
X
0
incorrect
error: at least one output is required
0
X
incorrect
error: at least two inputs are required
1
X
incorrect
error: at least two inputs are required
2+
1
correct
an operator sourcing a single cause
2+
2+
correct
an operator sourcing a common cause
3 Latest work
The concepts announced above were further developed during the author's cooperation with
GEN TRIZ LLC. and Algorithm Ltd. This cooperation resulted in involvement in real-life pro-
jects and access to complex CECA diagrams for experiments. A fragment of a model coming
from one of these projects is used as an example in section 4 (with changed parametrization).
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
30
The first enhancement consisted in extending the basic quality criteria to use additional infor-
mation provided by the analysts as box colors (blue for target disadvantages, orange for key
disadvantages, etc.). This additional attribute allowed to strengthen the quality assessment by
checking if the connection pattern of a disadvantage box complies with the type flagged by its
color (e.g. a disadvantage with no successors is considered a target disadvantage, and therefore
it should be blue). Furthermore, the considerations regarding the categorization of disad-
vantages led to the conclusion that in addition to hazards initiating harmful effects and vulner-
abilities, allowing the harms to affect the system, we should also distinguish "scaling factors",
determining the magnitude of an impact rather than causing it. The following physical contra-
diction describes the development and removal of such scaling factors or influence-related dis-
advantages:
influence-related disadvantages should be reflected as OR-connected to indicate that the
value of ANY of contributing parameters may be the cause of the problem, and
influence-related disadvantages should be reflected as AND-connected to indicate that a
change in ANY of contributing parameters may solve the problem.
The first part means that scaling factors model these arguments of a formula describing an
effect, which may actually change the output. For continuous mathematical functions, this con-
dition appears equivalent to requiring a non-zero first partial derivative by a particular argu-
ment. The second part, in turn, reflects the general operating rule stating that to eliminate the
effect, one has to eliminate ALL its preceding disadvantages connected with logical OR, while
it is sufficient to eliminate ANY of disadvantages connected with logical AND. This rule is
derived directly from De Morgan laws, saying that negation of conjunction (AND) is equivalent
to alternative (OR) of negated arguments and vice versa.
Solving this contradiction resulted in the specific handling of the key problems derived from
such influence-related disadvantages. Contrary to hazard-type and vulnerability-type causes,
which manifest themselves qualitatively (present-absent) and obey De Morgan laws, the scaling
factors are inherently quantitative in nature. Therefore, switching from combined contributing
disadvantages to combined partial solutions (described by an inverted function) does not require
changing the OR operator between the inputs into AND as we do for qualitative disadvantages.
Let us consider the kinetic energy of a moving object as an example. We know from physics
that it depends on the mass and squared velocity (E mv
2). Although we do not consider any of
these parameters as a cause of the energy, excessive kinetic energy may result either from the
excessive mass of the object or its excessive velocity or both, so we should use OR operator to
connect respective disadvantages in the model. However, to reduce the kinetic energy, we also
may decrease the mass, the velocity, or both, which describes logical OR operation again.
As can be seen, the rule of eliminating ALL input disadvantages combined with an OR operator
does not apply to influence-related disadvantages, which should be processed differently. This
explanation also provides a formal justification of a "shortcut" practice, consisting in modeling
such parametric disadvantages in an intentionally incorrect way as AND-connected to indicate
that changes of ANY of the involved parameters may be sufficient to eliminate the effect. Alt-
hough a model is formally defective, such manipulation allows for a uniform approach to all
logical operators with De Morgan laws when switching to the solution generation phase.
The categorization of disadvantages used for CECA modeling in the mentioned projects distin-
guishes target disadvantages, key disadvantages, intermediate disadvantages, and disad-
vantages being hard to eliminate or insignificant (shown as gray). This category is assigned to
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
31
disadvantages not promising as candidate key disadvantages because they are considered too
problematic to eliminate OR insufficiently impactful. Such interpretation makes this category
complementary to key disadvantages, expected to be impactful AND removable.
Although each branch should end with an irremovable root cause in a complete CECA model,
it seems to be a typical project practice to focus on indicating key disadvantages and hiding
fragments that are not worth further attention (especially for complex models). Therefore dia-
grams are often only developed until each branch ends with a hard to eliminate or insignificant
disadvantage or with a candidate key disadvantage. Such an approach is handy for traditional
(human) analysis, but it poses a significant limitation in automated model processing. Indeed,
hard to eliminate refers to the inability to act upon a given disadvantage and says nothing about
its impact, while insignificant refers to little impact and says nothing about the removability.
To overcome this limitation, it was decided to differentiate these two cases and build the model
description in the form of logical expressions expanded up to primary causes in each branch,
skipping those indicated as insignificant. The resulting expressions should only refer to key
disadvantages and hard to eliminate disadvantages. On top of this, the form of the expanded
expression indicates if a given set of key disadvantages covers all variants of triggering the
target disadvantages. If this condition is not met, then the current set of key disadvantages is
incomplete because their elimination would not guarantee to remove the target disadvantages.
4 Example
To support the proposed processing of cause-effect models, the author developed a dedicated
software tool in the form of several VBA (Visual Basic for Applications) procedures embedded
in an Excel workbook. They automate most of the process depicted in Fig. 2 and, used in tandem
with the free yEd editor, they serve as a proof of concept for the Advanced CECA method.
This tool was applied to analyze the preliminary CECA model in a real cost reduction project
regarding a water distribution system. The analysis revealed 18 key problems derived from the
key disadvantages. This list contained all 12 key problems eventually presented to the customer
as coming from CECA, which confirms the practical usability of the approach. The mentioned
model contained 74 disadvantages and 105 nodes in total, making it medium-sized compared
to other analyzed models, yet still too complex to present here in a readable form. Therefore,
the following illustrations use a fragment of the original diagram with the "sanitized" descrip-
tions to briefly demonstrate the process and its outcomes.
Fig. 2. An overview of the CECA model processing scheme
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
32
The CECA diagrams were developed as PowerPoint slides, reflecting disadvantages differenti-
ated by the colors of boxes (e.g. orange denotes key disadvantage), explicit logical operators
(&, OR), and connectors indicating links between the slides (as in Fig. 3).
Fig. 3. A single slide from an input file constituting the example
Information about the objects and their connections was extracted from the presentation, veri-
fied to detect structural errors, and formatted in an Excel file for import. After importing to yEd,
the model was manually adjusted (see Fig. 4). For complex models, this phase also comprises
removing repeated boxes and fixing possible errors, and its outcome should be a complete dia-
gram, consistently integrating the contents of all slides. The errors encountered so far ranged
from duplicated or absent connections resulting from stacking arrows one on another or missing
the connection spots of the objects to equivalent disadvantages shown on different slides with
different labels, e.g. X is excessive vs. X is too high. The former are automatically detected
during import using the criteria of diagram correctness, but the latter requires user involvement.
Fig. 4. A sample CECA model in yEd after manual adjustments
Next, the model representation was imported from the yEd file and further processed in Excel.
A causality matrix was built from the model description automatically, and the categorization
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
33
of disadvantages as hazards, vulnerabilities, or scaling factors was indicated manually together
with other parameters (Fig. 5). Initially, all connections are reflected in the matrix with 1s, but
these weights may be changed before further calculations using the outcome of the cost analysis
or other data we replaced 1s with 80 and 20 in row 4. Then the structural impact factors were
calculated (Fig. 6 and 7), and the plotted results are shown in Fig. 8. The expanded and reduced
logical expressions are presented in Fig. 9, and the model representation with categorization,
calculated impact factors, and other attributes was finally re-imported to yEd (see Fig. 10).
Fig. 5. A screenshot of the causality matrix generated from the input data with additional parameters
Fig. 6. A screenshot of the calculated impact factors table (minimal expression reduction)
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
34
Fig. 7. A screenshot of the calculated impact factors table (maximal expression reduction)
Fig. 8. The plots of impact factors and profitability of removal (maximal expression reduction)
Fig. 9. A screenshot of a table with expanded and reduced logical expressions (maximal reduction)
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
35
Fig. 10. A sample CECA model re-imported to yEd with additional attributes (maximal reduction)
As shown in Fig. 5, the first item, labeled as Two X4 surfaces are grinded, has one direct cause
indicated by 1 in column 3 (X3 design uses two valves). For the second data row, the number
of causes is 0, and the label is shown in red, indicating that a respective disadvantage is a root
cause. Similarly, item 4 (X4 cost is too high) reflects the target disadvantage having no effects,
and so the label of the respective column is also shown in red.
The impact factors are calculated using qualitative data regarding model structure (connection
scheme and logical operators) as well as quantitative data regarding the influence (weights).
The factors are evaluated from the target disadvantages down to the primary causes, and they
reflect the relative contribution of given disadvantages to the removal of the respective effect.
Therefore for an AND operator, the impact value propagates unchanged to all input branches
since the elimination of a single contributing cause is capable of eliminating the effect. For an
OR operator, in turn, the impact is split among the branches in shares proportional to the weights
indicated in the causality matrix for particular inputs.
The results depicted in Fig. 6 show the normalized relative influence of the incident causes. For
example, five causes in total were identified for effect 7. The direct causes 9, 10, and 15 bring
33% each due to the equal weights and OR operator involved, while the impact of the next level
causes 8 and 14 is also 33%, as they are connected in linear chains with 10 and 15, respectively.
Another set of results is shown in Fig. 6, with impact splitting rules changed so that disad-
vantages shown in gray are excluded if they are not considered blockers (see explanation be-
low). Data row 4 describing the target disadvantage has all columns populated in Fig. 7, without
column 4 reflecting the same disadvantage. The plots in Fig. 8 use data from this row.
Logical expressions shown in Fig. 9 describe particular rows in the direct form (reflecting con-
tributions of specific inputs to the given disadvantage) and in the inverted form (indicating how
the inputs contribute to eliminating the given disadvantage). As explained in section 3, the OR
operators combining solely disadvantages categorized as scaling factors (shown as orange in
Fig. 9 and 10) are not converted into ANDs during the generation of inverted functions. These
functions are used for calculating the expanded expressions in direct and inverted form, for the
target disadvantages or for all disadvantages, possibly excluding key disadvantages. If the key
disadvantages fully influence the target disadvantage, the resulting expanded expression should
only refer to the key disadvantages, and the reduced expression should be empty.
Depending on the parametrization, the expansion is performed down to primary causes or key
disadvantages, usually found earlier in the chains. The logical expressions may also be gener-
ated in reduced variants, skipping key disadvantages (assigned rank K in Fig. 5, 6, 7, and 9),
insignificant disadvantages (rank I), or even hard to eliminate disadvantages (rank H) in some
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
36
configurations. The minimal reduction only removes the key disadvantages so that all H- and
I-rank disadvantages remain in the expression unless they are masked by key disadvantages
found earlier. On the second level, the reduction also removes all I-rank disadvantages, and on
the third level, the H-rank disadvantages are also removed if they are located behind OR oper-
ators combining solely scaling factors. The justification for the latter approach is that such ORs
are not converted to ANDs upon switching from combining key disadvantages to combining
partial solutions so that removing ALL the input disadvantages is not required in such situa-
tions. In other words, an irremovable disadvantage located behind a "quantitative OR" operator
does not constitute a blocker, as in the case of a regular OR, because it may be made negligible
by eliminating any other scaling factor contributing to a given effect.
5 Summary and further research
Several concepts and activities contribute to the Advanced CECA method in its current form:
decomposing model into context-dependent contents and context-independent structure,
analyzing model structure using Boolean algebra and techniques from the digital design,
solving the contradiction regarding direct and indirect removal of disadvantages,
introducing a hazard-vulnerability approach to the identification of causes in linear chains,
introducing quantitative CECA extensions based on the risk management approach,
indicating methods for merging quantitative measures with logical functions,
introducing structural impact factors and the algorithm for calculating them,
introducing structural correctness criteria,
distinguishing causation and influence relations in CECA diagrams,
solving the contradiction regarding logical operators and influence-related disadvantages,
expanding logical expressions distinguishing hazards, vulnerabilities, and scaling factors,
expanding expressions distinguishing hard to eliminate and insignificant disadvantages,
supporting selection of key disadvantages with expanded logical expressions,
supporting selection of key disadvantages with impact and profitability ranking,
introducing causality matrix data structure as a representation of a CECA model,
developing algorithms for numerical processing of impact factors in causality matrix,
developing a prototype application supporting Advanced CECA.
Currently, the causality matrix uses a single row to reflect one disadvantage with an optional
logical operator on input. This representation is convenient for the algorithms developed so far,
but it brings a constraint of a single connection between the operator and disadvantage. Conse-
quently, operator-operator connections and multiple output connections from an operator can-
not be represented in the matrix directly, and these configurations require dummy intermediate
disadvantages to be added, which appear on the final diagrams (like f16 in Fig. 10). Such boxes
may be misleading for the analysts since they do not have counterparts in the original model,
and therefore the future development of the application should eliminate this side effect.
The proposed quality criteria indicate allowable combinations of the numbers of inputs and
outputs for different categories of disadvantages and logical operators, possibly extended by
matching topological features with a disadvantage type declared by the analyst using colors.
These criteria are objective and easy to assess by humans and software, but they only address
the structure of the model without analyzing the contents reflected by descriptions of the boxes.
Therefore, the approach is similar to a spellchecker, which can detect orthographic errors but
cannot assess if the story makes sense. The extended categorization (hazard vs. vulnerability
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
37
vs. scaling factor) allows strengthening quality assessment criteria by checking the type of an
effect and the types of its contributing causes. Only a few of the 180 possible configurations
were decided so far to be correct or incorrect, and most of this work is still to be done.
Moreover, some specific words (like insufficient or excessive) or phrases (too high, too low,
and the like) may be easily recognized in box descriptions and used to suggest the categorization
of specific disadvantages as scaling factors to extend automation. Similar rules may supposedly
be devised for other keywords. For instance, reference to a system or supersystem component
could imply recommended categorization as vulnerability or hazard, respectively. The recom-
mendation rules could also cover logical operators. The scaling factors would presumably feed
an OR operator aggregating their impact, while AND operator seems appropriate for combining
a hazard with one or more vulnerabilities so that triggering the unwanted effect requires all
contributing disadvantages to be acting jointly.
This research direction seems especially attractive for several reasons. First, it corresponds with
the extended quality criteria assessed over a building block spanning an effect with the input
operator and contributing causes. Second, contrary to the basic quality assessment, operating
on separate objects, this approach addresses overlapping model fragments, implying more reli-
able verification. Third, employing the box descriptions in processing crosses a bridge between
the model structure and the model contents, which appeared beyond reach before. Finally, this
approach may use Natural Language Processing techniques to develop far more sophisticated
verification scenarios, like checking if the neighboring disadvantages refer to the components
interacting in the function model of a system or building a resource list by extracting nouns
from the box descriptions. Such extensions would greatly support and automate the integration
between ACECA and other TRIZ tools.
Acknowledgments
The author gratefully acknowledges TRIZ Master Dr. Oleg Abramov for his patient supervision
and support in this work, including valuable discussions, questions, and comments.
References
1. Ponomarenko, A.I.: Selecting Tasks Using the Operator of Negation of Unwanted Action
(in Russian). Journal of TRIZ, 1995, No. 1, pp. 51-53.
2. Litvin, S.S., Akselrod, B.M.: Cause-Effects Chains of Undesired Effects, Methodical Theses
(in Russian). Saint Petersburg (1996).
3. Abramov, О., Kislov, A.: Cause-Effect Analysis of Engineering System's Disadvantages,
Handbook on Methodology (in Russian). Algorithm, Ltd., 2000.
4. Pavlov, V.V.: Method for Building a Map of Key Problems (in Russian), 2001. [online]
www.triz.natm.ru/articles/pavlov/pavlov01.htm.
5. Souchkov, V.: Root Conflict Analysis (RCA+): Structuring and Visualization of Contradictions.
Proceedings of the ETRIA World Conference TRIZ Future 2005. Graz, Austria. pp. 474-483.
Leykam Buchverlag 2005.
6. Pinyayev, А.: A Method for Inventive Problem Analysis and Solution Based On Why Why
Analysis and Functional Clues. TRIZ Master Thesis, 2007.
7. Axelrod, B.: Systems approach: modeling engineering systems using Interactions Causality
Scheme. In: Grundlach, K. (ed.) Proceedings of TRIZ Future 2007 Conference. Frankfurt,
Germany, pp. 131-138. MATRIZ 2007.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
38
8. Khomenko, N., De Guio, R.: OTSM Network of Problems for representing and analysing problem
situations with computer support. In: León-Rovira, N., (ed.) CAI 2007. IFIP AICT, vol. 250,
Trends in Computer Aided Innovation, pp. 77-88. Springer 2007.
9. Efimov, A.: Identification of Key Disadvantages and Key Problems using Cause-Effect Chains of
Undesired Effects (in Russian), 2011. [online] www.metodolog.ru/node/993.
10. Howladar, A., Cavallucci, D.: Analysing complex engineering situations through problem graph.
Procedia Eng. 9, pp. 1829. Elsevier Ltd. 2011.
11. Falkov, D., Misyuchenko, I.: Analysis of typical errors made when choosing logical functions
(in Russian), 2013. [online] www.metodolog.ru/node/1643.
12. Falkov, D., Misyuchenko, I.: Characteristics of building fragments of Cause-Effect Chains with a
serial connection of disadvantages (in Russian). 2013. [online] www.metodolog.ru/node/1654.
13. Medvedev, A.: Algorithm for Automated Building of Cause-Effect Chains of Disadvantages
(in Russian). TRIZ Master Thesis, 2013.
14. Yoon, H.: Occasion Axis and Parameter-Function Pair Nexus for Effective Building of Cause
Effect Chains. In: Souchkov, V., Kässi, T. (eds.) Proceedings of the TRIZfest-2014 International
Conference, Prague, Czech Republic, pp. 184-194. MATRIZ 2014.
15. Kisela, T.: RCA+ Evaluation Based on Vague Data. In: Souchkov V., Kässi T. (eds.) Proceedings
of the TRIZfest-2014 International Conference, Prague, Czech Republic, pp. 216-230. MATRIZ
2014.
16. Dobrusskin, C.: On the identification of contradictions using Cause Effect Chain Analysis.
Proceedings of ETRIA World Conference TRIZ Future 2015, Berlin, Germany, 2015,
pp. 221-224.
17. Kislov, A.: Cause-Effect Modeling and Analysis of Disadvantages of a Technical System
(in Russian), 2016. [online] http://ratriz.ru/wp-content/uploads/2016/08/Kislov-A.V._Prichinno-
sledstvennoe-modelirovanie-i-analiz-nedostatkov-TS.pdf.
18. Chrząszcz, J., Salata, P.: Cause-Effect Chains Analysis using Boolean algebra. 16th International
TRIZ Future Conference TFC 2016. In: Koziołek, S., Chechurin, L., Collan, M. (eds.) Advances
and Impacts of the Theory of Inventive Problem Solving. The TRIZ Methodology, Tools and Case
Studies, pp. 121-134. Springer 2018.
19. Chrząszcz, J.: Quantitative approach to Cause-Effect Chains Analysis. In: Souchkov, V. (ed.)
Proceedings of the TRIZfest-2017 International Conference. Krakow, Poland, pp. 341-352.
MATRIZ 2017.
20. Chrząszcz, J.: From Problem to Objective – Complementing TRIZ with NLP. In: Souchkov V.
(ed.) Proceedings of the TRIZfest-2017 International Conference. Krakow, Poland, pp. 168-177.
MATRIZ 2017.
21. Lok, A.: A Simple Way to Perform CECA And Generate Ideas in Practice. Proceedings of the
TRIZfest-2017 International Conference, Krakow, Poland, pp. 23-30. MATRIZ 2017.
22. Sun, Y.: Initial Disadvantages Identification. In: Souchkov, V. (ed.) Proceedings of the
TRIZfest-2017 International Conference. Krakow, Poland, pp. 217-221. MATRIZ 2017.
23. Abramov, О.Y., Savelli, S.: Identifying Key Problems and Conceptual Directions: Using the
Analytical Tools of Modern TRIZ. In: Mayer O. (ed.) Proceedings of the TRIZfest-2018
International Conference, Lisbon, Portugal, pp. 55-68. MATRIZ 2018.
24. Chrząszcz, J.: Indicating system vulnerabilities within CECA model. In: Mayer, O. (ed.)
Proceedings of the TRIZfest-2018 International Conference, Lisbon, Portugal, pp. 31-37.
MATRIZ 2018.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
39
25. Chrząszcz, J.: Modeling CECA diagram as a state machine. In: Cavallucci, D., DeGuio, R.,
Koziołek, S. (eds.) Automated Invention for Smart Industries, 18th International TRIZ Future
Conference TFC 2018 Proceedings, IFIP AICT, vol. 541, pp. 302-314. Springer 2018.
26. Lok, A.: Selecting and Validating Key Problems in TRIZ projects. In: Mayer, O. (ed.) Proceedings
of the TRIZfest-2018 International Conference, Lisbon, Portugal, pp. 151-169. MATRIZ 2018.
27. Lee, M.G., Chechurin, L., Lenyashin, V.: Introduction to cause-effect chain analysis plus with an
application in solving manufacturing problems, International Journal of Advanced Manufacturing
Technology (2018) 99: pp. 21592169.
28. Chrząszcz, J.: Deriving Quantitative Characteristics From CECA Model. Proceedings of the
TRIZfest-2019 International Conference, Heilbronn, Germany, pp. 89-99. MATRIZ 2019.
29. Chrząszcz, J.: Logical Negation in CECA Model. Proceedings of the TRIZfest-2019 International
Conference, Heilbronn, Germany, pp. 222-233. MATRIZ 2019.
30. Chrząszcz, J.: Exploring state machine CECA model. In: Benmoussa, R., De Guio, R., Dubois, S.,
Koziołek, S. (eds.) New Opportunities for Innovation Breakthroughs for Developing Countries
and Emerging Economies, 19th International TRIZ Future Conference TFC 2019 Proceedings,
IFIP AICT, vol. 572, pp. 388-399. Springer 2019.
31. Chrząszcz, J.: Towards Automation of Cause-Effect Analysis. In: TRIZ Review: Journal of the
International TRIZ Association MATRIZ. Vol 2/1, April 2020, pp. 100-105.
32. Sun, Y.: A New Operator In CECA. In: TRIZ Review: Journal of the International TRIZ
Association MATRIZ. Vol 2/1, April 2020, pp. 106-111.
33. Chrząszcz, J.: Eliminating Disadvantages by Changing Transitions in a State Machine Cause-
Effect Model. In: Systematic Complex Problem Solving in the Age of Digitalization and Open
Innovation. 20th International TRIZ Future Conference TFC 2020 Proceedings. IFIP AICT, vol.
597, pp. 268-279. Springer 2020.
34. Chrząszcz, J.: Causality matrix - a computer-friendly cause-effect model representation poster
presentation. MATRIZ Online Forum 2020.
35. Hanifi, M., Chibane, H., Houssin, R., Cavallucci, D.: A Method to Formulate Problem in Initial
Analysis of Inventive Design. In: Nyffenegger, F., Ríos, J., Rivest, L., Bouras, A. (eds.) Product
Lifecycle Management Enabling Smart X. PLM 2020. IFIP AICT, vol. 594, pp. 311-323.
Springer 2020.
36. Chrząszcz, J.: Indicating and assessing quality criteria for cause-effect models. In: Creative
Solutions for a Sustainable Development, 21st International TRIZ Future Conference TFC 2021
Proceedings, IFIP AICT, vol. 635, pp. 220-232. Springer 2021.
[ all online references last accessed on 10 April 2022 ]
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
40
TRIZfest-2022
APPLICATION OF TRIZ IN IPD
Jun Huang
ZheJiang Uniview Technologies Co.,Ltd, China
Abstract
IPD solves the problem of what to do and how to do, but IPD cannot comprehensively solve a series of
technical problems faced by enterprises in the process of R&D. TRIZ can not only solve the technical
problems in R&D, predict the development of products, but also solve the problems in the process of
technology transfer. This paper introduces how TRIZ integrates into IPD and how to solve problems in
each stage.
Keywords: IPD, TRIZ, Fuzzy front end, New product development, Commercialization, Forward prob-
lem flow, Backward problem flow, Self-solving problem flow
1 Introduction
IPD is a systematic approach to product development that achieves a timely collaboration of
necessary disciplines throughout the product life cycle to better satisfy customer needs.
- Software Engineering Institute
Organizational Innovation is the guarantee for efficient and continuous innovation of enter-
prises. IPD (Integrated Product Development) combined with TRIZ, it can accelerate organiza-
tional innovation to level 5.
Table 1. The Level of Organizational Innovation
Level
Explaination
Level 1 informal management
Personal experience/non-standard practices
Level 2 excellent function
Functions are clear and complete, but cross-func-
tion operation is difficult
Level 3 excellent projects
The project realizes effective operation across func-
tions from concept to market
Level 4 Excellent Product Portfolio
Realize leverage utilization of product platform, ex-
cellent combination management, project selection
and execution
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
41
Level
Explaination
Level 5 world-class R&D capabilities
Form a leading position in key technology and re-
lated product fields, significantly improved R&D ef-
ficiency
IPD provides a series of processes and tools to ensure product delivery, but does not provide
corresponding solutions to technical problems and difficulties. While TRIZ provides a variety
of problem solving and innovative tools to overcome resource constraints and eliminate tech-
nical bottlenecks, and can provide more and more innovative solutions for IPD projects. At the
same time, the unique technology and function prediction principle of TRIZ can help enterprises
better grasp the future of the project and effectively improve the efficiency of project imple-
mentation.
Combined with TRIZ's IPD, product development and technology R&D can produce high-
quality solutions and inventions, form the key technology of the enterprise, enhance the leading
position of related products and technologies, and greatly improve the research and develop-
ment efficiency.
2 Origin of IPD
IPD is a set of product development models, concepts and methods. The idea of IPD comes
from the book <PACE——Product And Cycle-time Excellence> published by American
PRTM company, which describes in detail all aspects of this new Product development model.
IBM was the first to put IPD into practice.
Under the influence of IBM's successful practices, many high-tech companies have adopted the
integrated product development (IPD) mode, such as Boeing and Huawei, which have achieved
great success. Practices have proved that IPD is not only an advanced idea, but also an excellent
product development mode.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
42
3 Innovation throughout the R&D lifecycle
Fig. 1. Application of TRIZ in IPD
IPD solves the problem of what to do and how to do, but IPD cannot comprehensively solve a
series of technical problems faced by enterprises in the process of product R&D. TRIZ can not
only solve the technical problems in R&D, predict the development of products, but also solve
the problems in the process of technology transfer.
Fig. 2. Tools of TRIZ
TRIZ provides a variety of tools to solve difficult problems, which can greatly improve the
innovation ability of enterprises. Taking Samsung as an example, after the introduction of TRIZ
in 1998, Samsung's innovation ability gradually increased in the following years, and the
amount of invention patent authorization steadily increased. In 2003, Samsung made use of
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
43
TRIZ to guide R&D and contributed 1.5 billion US dollars, obtaining 1,600 patents, ranking
fifth in the United States that year. In 2010, Samsung was ranked second in the list of invention
patent authorization, becoming one of the world's most innovative enterprises after IBM. Sam-
sung topped the list of the world's most patented companies in 2021.
Fig. 3. Patents Activity in Samsung
Innovation runs through the entire R&D lifecycle of IPD. In the process of product development
and technology R&D, n problems may be encountered, they can be resolved in multiple ways
by TRIZ tools, and m inventions may be generated. Some of the inventions are further applied
to the process of product development and technology R&D, forming a benign development.
Combined with TRIZ's IPD, product development and technology R&D can produce high-
quality solutions and inventions, form the key technology of the enterprise, enhance the leading
position of related products and technologies, and greatly improve the R&D efficiency.
4 Innovation at different stages
Fig. 4. Three Stages of the Product Innovation Process
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
44
According to Koen's research, the product innovation process can be divided into three stages
or sub-processes: Fuzzy Front End (FFE), New Product Development (NPD) and Commercial-
ization.
Among them, the fuzzy front-end corresponds to the pre-order part of IPD, namely the analysis
work of product project initiation; The new product development corresponds to the product
development process of IPD, from product concept to product release stage; The commerciali-
zation corresponds to the product lifecycle management stage in IPD.
Fig. 5. Problem Flow in the Process of Innovation
Fig. 6. Problems & Inventions & Innovations
In the process of product development and innovation, various problems may be encountered.
The causes of these problems may come from the previous or later stages of a certain stage of
the main innovation process, or may be generated by this stage itself. All of the problems gen-
erated form a problem flow and are constructed into the problem flow model shown in figure
6. In the problem flow model shown in figure 6, there are three types of flows: forward problem
flow, backward problem flow, and self-solving problem flow.
(1) Forward problem flow: The cause of the problem occurs at any stage before a certain stage
of the main innovation process, but it must be solved at this stage. For example, the problem
must be solved by the conceptual design stage, but the cause of the problem occurs in the fuzzy
front-end stage.
2) Backward problem flow: The cause of the problem occurs at a certain stage of the main
innovation process, but it must be fed back to a certain stage before this stage to solve. If the
cause of the problem appears in the manufacturing stage, it must be fed back to the detail design
stage to solve.
(3) Self-solving problem flow: The problem occurs at a certain stage of the main innovation
process and is solved at this stage.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
45
Problems at any stage can bring creativity and invention, some of which are further imple-
mented in products. By using TRIZ, sometime, we can produce high-quality solutions and in-
ventions.
Fuzzy Front-end is the process of creativity, and R&D engineers are the main body of invention
or creativity, and the core process is completed by them. The core process is divided into five
stages, namely, opportunity identification, opportunity discovery, creation and improvement,
creation selection, concept definition or creation formation. The creativity formed in this stage
is the input of the follow-up new product development stage. Once confirmed by evaluation,
the follow-up R&D process can be started. However, creativity itself is only a feasible idea, and
it also needs to be implemented in the invention process of new product development.
For enterprises that already have technical strength, they can predict the future development
direction of technology and the related field of technology development through the analysis
of key technology S curve, so that not only can their R&D keep up with the development di-
rection of technology, but also can allocate resources to the research of related field.
According to their own strength, the enterprise adopts the patent layout strategy: sniper strat-
egy, blocking strategy, carpet strategy, isolation strategy, can prevent competitors from enter-
ing the existing market. And competitors can adopt the patent layout strategy: destroyer strat-
egy, encirclement strategy, and participate in market competition. Owning patent rights can
exchange the use of patent license with other companies, thus saving resources and R&D
costs.
4.1.1 Start from an innovation point
In the fuzzy front-end stage, we can start from a certain innovation point, use TRIZ to find multi
solutions. At the same time, we can pay attention to the depth of patent layout, and apply for a
patent horizontally.
Typical process:
Find out the related factors of this innovation point.
Find out other innovation points of each associated factor; one associated factor may contain
one or more innovation points.
A new technical solution is formed based on the summary of other innovation points. The
same innovation can be implemented in multiple ways, corresponding to one or more tech-
nical solutions or technical schemes. By using TRIZ, sometime, we can produce high-qual-
ity solutions and inventions.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
46
Some innovation
Related
factor
1
Related
factor
2
Related
factor
3
Innovation
aInnovation
bInnovation
cInnovation
dInnovation
e
Technical
scheme
1
Technical
scheme
2
Technical
scheme
3
Technical
scheme
4
Technical
scheme
5
Technical
scheme
6
Technical
scheme
7
Technical
scheme
8
Fig. 7. Example of Starting from an Innovation Point
Product development corresponds to the product development process of IPD, from product
concept to product release stage; Innovation can be carried out in many aspects at this stage.
With TRIZ tools, we can find multi solutions. The typical innovation process can start from the
project task and apply for patents horizontally and vertically.
4.2.1 Start from a project task
A project task can be a specific project, or a product or solution development.
Typical process:
Find the components of the task;
Analyze the technical elements of each component;
Resolve the problems in the technical elements, and find multi solutions;
Find out the innovation points of each technical element; One technical element may con-
tain one or more innovation points.
Output technical solutions based on innovation points. The same innovation can be imple-
mented in multiple ways, corresponding to one or more technical solutions or technical
schemes. By using TRIZ, sometime, we can produce high-quality solutions and inventions.
The adoption of the above process is conducive to the comprehensiveness of the patent layout,
and patent applications are carried out horizontally and vertically. If we further refer to the
technology development trend, we can make a patent layout for the future in terms of key tech-
nical elements.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
47
Component 1
Technical
element
1
Technical
element
2
Technical
element
3
Innovation
aInnovation
bInnovation
cInnovation
dInnovation
e
Technical
scheme
1
Technical
scheme
2
Technical
scheme
3
Technical
scheme
4
Technical
scheme
5
Technical
scheme
6
Technical
scheme
7
Technical
scheme
8
Component 2 Component 3
Project task
Technical
element
4
Technical
element
5
Fig. 8. Example of Starting from a Project Task
Commercialization corresponds to the product lifecycle management phase of IPD, which in-
volves the development of customized requirements and product maintenance. In the process
of implementing products in specific scenarios, various challenges, opportunities and problems
will be encountered, which can be the source of innovation.
4.3.1 Mining from problems encountered
Problems encountered in daily work, defects in existing products, inconvenience in use, or de-
ficiencies in existing methods and processes can all become the source of innovation.
Measures taken to solve technical problems and schemes taken to repair defects.
Summarize the technical problems to be solved, summarize the technical means adopted in
solving the technical problems and the expected results obtained, conclude a new technical
scheme, and try to carry out patent layout.
5 Conclusion
IPD solves the problem of what to do and how to do. TRIZ solves a series of technical problems
in the process of enterprise product development, deployment and application. The combination
of TRIZ's IPD can greatly improve the innovation ability of enterprises, not only can produce
high-quality schemes and inventions, form the key technology of enterprises, enhance the lead-
ing position of related products and technologies, but also can greatly improve the research and
development efficiency.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
48
6 Reference
1. "The demand of enterprise high-quality innovation for intellectual property services" He Ming-
ming
2. TRIZ-methods to overcome thinking inertia
3. Report of the 2018 TRIZ advanced seminar 20180519-Tan Runhua
4. 2018-Tianjin-TRIZCON-II-conference report-Tan Runhua
5. C- TRIZ and application invention process solution theory
6. Case Study on technology innovation of Samsung company in South Korea
7. Research and Application of DFSS process innovation based on TRIZ
8. Introduction of TRIZ at a company by using the distance TRIZ-trainer course and the Solving Mill
software (the example of SAMSUNG), Proceedings of the MATRIZ International Conference
TRIZfest-2021: September 15-18, 2021
9. PACE learning courseware-Baidu Library (baidu.com)
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
49
TRIZfest-2022
APPLYING TRIZ TO FORECAST THE
DEVELOPMENT OF CECA
Oleg Y. Abramov *, Jerzy Chrząszcz **
* Algorithm Ltd., St. Petersburg, Russia
** Institute of Computer Science, Warsaw University of Technology, Poland
** Pentacomp Systemy Informatyczne S.A., Warsaw, Poland
Abstract
Using TRIZ tools to analyze shortcomings and predict the future development of both TRIZ itself and
its individual tools is a logical step for TRIZ developers, and, indeed, various researchers have made a
few attempts in this direction. However, a clear algorithm for using TRIZ to forecast the evolution of
TRIZ tools has not yet been introduced and the applicability of TRIZ for this task remains questionable
considering that TRIZ methodological tools are not “material” objects for which TRIZ was originally
developed. This paper briefly presents an algorithm that involves Function and Main Parameters of
Value (MPV) analyses of the methodological tool whose future development is to be forecasted, as well
as using the S-curve analysis from the Trends of Engineering Systems Evolution (TESE). In this paper,
this algorithm is applied to Cause-Effect Chains Analysis (CECA), together with its extensions, and it
is shown how CECA can be decomposed and analyzed as an Engineering System. A Function Model of
CECA is proposed and discussed first, and then the Main Parameters of Value (MPVs) are identified.
Finally, the CECA’s level of development with respect to these MPVs is assessed and directions in
which CECA may evolve are forecasted using the results of the S-curve analysis. This approach seems
to create a foundation for further assessments using other Trends of Engineering Systems Evolution and
for predicting the future development of this method in more detail.
Keywords: TRIZ, Cause-Effect Chains Analysis, CECA, Trends of Engineering Systems Evolution,
TESE, Function Analysis, MPV Analysis, S-curve Analysis, prediction, forecasting
1 The challenge of using TRIZ on TRIZ
For decades, TRIZ, and in particular Trends in Engineering Systems Evolution (TESE), has
been successfully used to predict the evolution of various technical systems, helping to effi-
ciently develop new products and technologies that become evolutionary winners.
As TRIZ has proven to be applicable to various systems, some attempts have been made to
predict even the future development of TRIZ itself by means of TESE [1, 2], but these attempts
were related to TRIZ in general, and therefore provided only very general directions for the
future development of TRIZ, such as the increasing involvement of Artificial Intelligence.
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
50
On the other hand, TRIZ has a number of flaws [3] that need to be addressed, and for practical
purposes it would be valuable to forecast the future development of specific TRIZ tools to im-
prove them. However, this is a challenging task because:
1. TRIZ tools are not “material objects” for which TRIZ was originally developed,
2. the main functions of TRIZ tools are not well-defined, and
3. the components of TRIZ tools are not defined at all.
The purpose of this article is to determine whether it is possible to analyze TRIZ tools and
predict their further evolution using the TRIZ tools themselves. To answer this question, the
authors attempted to apply MPV and Function analyses, as well as S-curve analysis, to one of
the most important TRIZ analytical tools, Cause-Effect Chains Analysis (CECA), considering
it as a software-like technical system.
2 Introduction to Cause-Effect Chains Analysis
The concept of causality is used in various fields of human activities to investigate causes and
scenarios of past events (e.g., failures) as well as to predict and characterize future events (e.g.,
risks). Several systematic approaches have been developed over the decades for technical and
business purposes, and the following sections will briefly characterize them. The overall goal
of these methods is to provide a better understanding of the situation so as to eliminate or pre-
vent undesirable consequences. In all these methods a model of the system (or process) is built
that represents the relationships between the relevant causes and effects.
The Cause-Effects Chains Analysis (CECA) method was developed at GEN3 in the 1990s [4-6]
and is used in the TRIZ community worldwide as one of the problem analysis tools to identify
the key problems that need to be solved to achieve the project goals (see Fig. 1).
Fig. 1. Location of CECA method within a TRIZ project
System + Initial Problem to solve + Constraints
Solution ideas
Problem analysis to reveal Key Problems:
Function Analysis
Flow Analysis
TESE Analysis
CECA (the focus of this paper)
Key Problems solving:
Function Oriented Search (FOS)
ARIZ
Standard Solutions
Etc.
Information sources +
external subject matter experts
TRIZ project
Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
51
The CECA method begins by identifying the undesirable effects to be eliminated from the sys-
tem, which are referred to as target disadvantages. The causes of these effects are then identified
to build a chain of intermediate disadvantages until a cause beyond our control is reached, which
is considered the root cause and completes the analysis of this chain [7].
Although CECA may seem similar to the “5 Why’s” method used in the Root Cause Analysis
[8], Fault Tree Analysis [9], or the Current Reality Tree in Theory of Constraints [10], it has
important differences that make CECA superior [7, 11].
The outcome of CECA is usually depicted as a diagram consisting of linear chains of boxes
describing the disadvantages, connected by arrows indicating the flow of causality. The chains,
in turn, may connect at the inputs by common causes or at the outputs by logical operators.
When all the input causes are jointly required to trigger a given undesirable effect, the operator
is indicated as AND (logical conjunction). If the effect may be triggered by any of the input
causes acting alone, the operator is indicated as OR (logical disjunction), or the operator is not
used and each of the input causes is directly connected with the effect, to reflect its alternative
triggers. The model may contain loops of causes, representing feedback paths in the system.
Although the root causes cannot be directly eliminated, they, as a rule, indicate the main factors
influencing the system and define the boundaries of the model. The expected outcome of the
CECA process is a set of key disadvantages, which when eliminated will result in the removal
of the target disadvantages specified for the project. Such key disadvantages usually reflect
deep, primary causes, having large impact on the target disadvantages. After transforming key
disadvantages into key problems [12-15], they are expected to inspire strong solution ideas.
A general shortcoming of CECA, which seems to be common for all methods of cause-effect
analysis, is the uncertainty about the completeness and correctness of the outcome. Usually,
there is no way to prove with certainty that a given defect cannot be caused by another cause or
combination of causes not accounted for in the model. A typical way to address this problem is
to create a team of TRIZ experts who ask the relevant questions and subject matter experts who
provide competent answers. Thus, the probability of overlooking a known factor with signifi-
cant consequences may be reduced to an acceptable level. Hence, the nature of decisions in this
area is akin to those in risk management [16].
On the other hand, in practice, we only need CECA models, which are useful for identifying
key disadvantages that cause our target disadvantage (e.g., insufficient performance, high costs,
etc.), and are not necessarily “complete”. In fact, the “compression” achieved by ignoring un-
necessary information is the main advantage of any model. Therefore, a CECA model should
be as simple as possible, yet correct, complete, and expressive enough to support the problem-
identification process.
All causes in a CECA model are uniformly perceived as disadvantages, and this seems to be an
inherent limitation of this method since some of the phenomena contributing to the development
of the target disadvantages may be necessary for providing useful system functions. Eliminating
such “disadvantageous” phenomena may affect respective useful functions, inevitably incurring
some secondary problems in place of the eliminated one. The original CECA does not allow
information regarding useful consequences to be included in the model, and, therefore, cannot
use this information to select key disadvantages better [17, 18]. <