CC BY
3ДОИ
Kulakov A., Trantin A., Rubin M. Increase of Cause-Effect Chains Analysis Accuracy through Automation
Abstract. The design experience of the authors shows that cause-effect chains analysis (CECA) is one of the main analytical tools of TRIZ-projects at industrial facilities. However, with all its intuitive comprehensibility and popularity, the cause-and-effect analysis procedure is a complicated and painstaking process. Its results depend heavily on correct building of cause-and-effect relations between events, where should be no paralogisms and cognitive biases of the CECA actor. It is the authors' opinion that exactly elimination of these harmful effects is the main value in automation of cause-effect chains analysis.
This paper gives results of an investigation conducted on a sample of cause-and-effect chains of TRIZ-projects for mistakes made by TRIZ-specialists during cause-effect chains analysis. Types of such mistakes and their consequences are also given here.
The paper briefly describes results of an overview conducted on references dedicated to increase of cause-effect chains analysis accuracy including that through automation.
The authors also provide insight into functions of the Compinno-TRIZ module, where CECA is automated in two modes: without logic check and with logic check. The paper gives comparison of cause-and-effect chains made without logic check and with logic check in the Compinno-TRIZ CECA module, as well as provides practical examples of this module application in various modes.
Key words: cause-effect chains analysis, cognitive biases, human factor in TRIZ, automation of TRIZ, five whys, Ishikawa diagram, Compinno-TRIZ.
INTRODUCTION
Cause-effect chains analysis is a method for analysis of a system designed for detection of key causes resulting in occurrence of target undesirable effects in the system and based on building of cause-and-effect chains of the existing system disadvantages [6]. Apparently, one cannot agree with an opinion that CECA is something unique and inherent exceptionally to the 'Theory of Inventive Problem Solving' (TRIZ). Cause and effect analysis is wide spread in one form or another in many existing methods, such as lean production and TPS, where it is known as '5 Whys' [7] or its version '5 Whys + 1 How', quality management and TQM with their Ishikawa diagrams [8], current reality trees in TOC [9], etc. Whereas, the method roots can be found in works of Socrates, an Ancient Greek philosopher, who was skilled in the art of evidences based on building of logically sound cause and effect chains.
Such interest to this tool and its ubiquity are easily explainable. First, CECA is intuitively comprehensible for novice users, since it maximally involves the existing thinking mechanisms for building of cause-and-effect relations between occurrences and events. Second, cause and effect analysis elements as indicated above are present in many other popular management methods, which are applied in various production companies.
However, apparent simplicity of the tool is quite deceptive. The existing mechanisms of human thinking are non-ideal and they often contain cognitive biases. In practice, it is not unusual, when paralogisms in built cause and effect relations are committed even by experienced users of the tool. For example, there is a fairly common mistake in building of a cause-and-effect relation between successive events, while this relation can be absent at all. Generally, one can suggest a whole layer of mistakes of the 'Correlation does not imply a cause-and-effect relation' type [10]. The main actor during CECA is a person and it should be certainly taken into account that his/her cognitive biases can reduce efficiency of design work with all good practices of the CECA method. It is the authors' opinion that exactly elimination of these harmful effects is the main value in automation of cause-and-effect analysis.
Investigation of cause-and-effect chains of TRIZ-projects
The objective of our investigation was to identify the existing mistakes in cause-and-effect chains, define consolidated mistake types, and estimate possible consequences of these mistake types in projects. For investigation of typical mistakes in cause-and-effect chains (CEC), a sample of 12 production TRIZ-projects in different areas was taken. Cause and effect analysis was made in these projects by different TRIZ-specialists with international certification from Level 1 to Level 4. Two CEC from the sample are given below for example (Fig. 1):
Figure 1
38 mistakes were found in 12 chains. The following distribution was obtained after grouping of the mistakes (Fig. 2):
25
22
11
Fuzzy formulation of Establishment of false undesirable effects cause and effect rela
tions
Figure 2
70% 60% 50% 40% 30% 20% 10% 0%
It should be noted that this distribution is typical particularly for this sample and it may be significantly different in another sample. The main value is represented here by the mistakes themselves and their types regardless of their occurrence frequency.
The most common mistake type in the presented CEC sample was 'Omission of intermediate undesirable effects' - more than 60%. Below, an example of such mistake from the same sample will be given and its possible consequences will be demonstrated. Let's partially review one of 12 CEC (Fig. 3).
'Increase of bauxite fraction composition in ball mill' is a target undesirable effect. Milling in a ball mill is achieved through impaction of milling bodies against a milled material. Again, the impact energy depends on weight of a milling body itself and its movement speed, which is defined in its turn by lifting height of the milling body in the mill. I.e., the presented chain should contain at least another undesirable effect 'Weak impaction of milling bodies against milled material' between 'Increase of bauxite fraction composition in the ball mill' and 'Low lifting height of milling bodies in the mill'.
The target undesirable effect particularly in this project was eliminated at the level of key disadvantages. However, in cases, when a key disadvantage cannot be eliminated, for example, due to project constraints, a negation operator is used instead. Just then, omission of intermediate undesirable effects results in loss of additional problems. In its turn, this means loss of project flexibility.
The next frequent mistake type in our sample was 'Fuzzy formulation of undesirable effects' accounting for about 30%. Some examples of such formulations are given below: 'Welding procedure', 'Production culture', 'Untimely correction', 'Designed structure', 'Production process'. The objective of cause-and-effect analysis is to formulate key problems, solution of which will allow eliminating the target undesirable effect. The reader can make sure by himself/herself that it is impossible to state certain problems for elimination of the above undesirable effects, hence it will be impossible to solve them.
Despite sparsity of the remaining mistake types, it is deemed advisable to highlight them as well, since these mistake types may turn out to be more numerous for other CEC samples.
'Establishment of false cause and effect relations' accounted for only about 5% in the chains of our sample, although a somewhat higher value was expected due to common confusion between correlation and causation. Anyway, this mistake type takes place. An example from our sample is given below (Fig. 4).
It follows from the given CEC part that the cause of 'Premature attrition of the armour plate corrugations' is 'Quick wear of heads in bolts securing the armour plates' to the mill framework. However, the situation is actually different. Attrition of the armour plate corrugations takes place regardless of head condition in bolts securing the armour plates and consequently the latter cannot be the cause of the former.
Definitely, the reader can imagine the consequences of such mistakes. Building of false cause and effect relations results in formulation and solution of false problems. Finally, time will be lost and labour resources will be spent for solution of problems not allowing achievement of the project objectives. The last mistake type is 'Uncontrolled U-turn of analysis direction'. The CECA method has three distinct analysis directions - inward, outward and in plane. This mistake type means that a user begins an analysis in one direction, does not reach the key undesirable effects, and turns the analysis to another direction. An example of such chain is given in Fig. 5.
Decrease of the contact cross section' formulation quite defines an operational zone - busbar-rod contact space. It is followed by an undesirable effect in the same space with a clarification - 'Formation of blisters in the busbar and anode rod'. However, the next undesirable effect enlarges the operational zone and the analysis is generally switched to another integral process -anode beam racking and anode effect. I.e., the operational zone gets fuzzy. This will induce solution of more complicated problems. One should state a problem for elimination of 'Anode effect during anode beam racking' for such chain. Instead of the narrow busbar-rod contact zone, solution of this problem will affect the integral production process and pot as a whole, since anode effects occur in a pot as a whole. I.e., the project team will come to solution of a more fundamental problem. The solution of such problem will be more difficult to find and much more troublesome to implement.
During the investigation, 5 mistake types were revealed, each of which could result in negative consequences in a TRIZ-project. All these five mistake types occur, because, despite the existing CECA procedure rules, CECA is made yet by a person, who is subjected to cognitive biases to a greater or lesser degree. In view of CECA relevance in TRIZ-project activities, some developments should be focused on reduction of impact from cognitive biases. It is the authors' opinion that one of such directions can be automation of cause-and-effect analysis.
OVERVIEW OF AVAILABLE REFERENCES ON INCREASE OF CECA ACCURACY
During work to increase CECA accuracy, the authors analyse many various references concerning not only CECA under TRIZ-projects. For purposes of this paper, the authors focused on the methods primarily implying automation of the CECA process, thus aimed at increase of CECA accuracy and minimization of impact from actors' cognitive biases on final conclusions after cause-and-effect analysis. Below, the authors will selectively review the references, which in the authors' opinion affect the above-mentioned objective of this paper.
As many CECA investigators, the authors come to the conclusion that many of the offered software solutions, such as [11], [12], are focused exceptionally on various methods of data visualization. However, even the existing standard software, which is not designed for building of cause-and-effect chains, can cope with this task in many respects. So, the authors of references [1], [2], [3] indicate that cause and effect chains built as diagrams allow achieving the visualization objectives. For that, various editors are applied, such as MS PowerPoint or other more specialized diagram editors. However, the author also notices that the editors suitable for creation, edition, and visualization of cause-and-effect chains are not able to provide CECA support through automation, particularly for big chains. The author offers to apply a cause-and-effect matrix (Fig. 6). The matrix contains rows
Figws 5
with causes, columns with effects, and 1 at crossings with an available relation between a cause and an effect. Logical operators (AND/OR) are also taken into account. The author offers an algorithm for processing of such matrix, which allows establishing the shortest ways for elimination of the target undesirable effect.
In opinion of this paper authors, this vector of CECA automation is focused on just secondary problems occurring during analysis. When there is a competently built CEC, selection of an optimal way for elimination of the target undesirable effect takes incommensurably less time than for correct formulation of undesirable effects and building of cause-and-effect relations between them. Furthermore, mistakes are made during placement of logical operators, rather than during analysis of already placed operators.
The author of reference [13] raises a question of complication in building of cause-and-effect chains in tangled situations. The offered method for increase of CECA accuracy (and search for both root causes and solutions as well) supposes application of 'introspection', 'which consists in more detailed CECA of any relation or block'. Such transition to the microlevel can really assist in better understanding of the situation, but it is difficult to suggest reduction of impact from the actor's initial cognitive biases on new cause and effect relations.
OJtM OfMfMOft effect« 1 2 2 1 3 2 ' 2 1 1 a 0
»Utti » («ertiv i] x2 »3 «4 iS x6 k7 ■8 x9 vl V2
0 ia
' o «2
0 x) ty
0 e4
2 ON KS 1 1
2 OR e6 1 1
2 k7 1 1
2 AND «S 1 X
2 on 19 1 1
2 ANO Vl 1 1
2 on v2 1 1
Figure 6
It is worth particularly mentioning the solutions, which implement automation elements of the CEC building process for increase of analysis accuracy through a set of check questions preset in the system including AI-based and ready lists [14, 15]. For example, when a user finds a cause of any undesirable effect, he/she can study it in more details using the questions offered by the programmed. Successive cutting of 'unnecessary' possible causes can quite greatly narrow the list of possible candidates for the 'villain' of the whole cause and effect analysis. However, relations between possible causes and undesirable effects are still defined by the user and consequently risks of CECA accuracy decrease still exist.
The obtained results of the authors' investigations can be reduced to a simple thesis - CECA developments are generally focused on creation of a graphic editor, which would allow facilitating the specialist's mechanical efforts, but the developers keep the user still responsible for checking the logical validity in building of cause-and-effect chains. There are also some works aimed to automate selection of the most optimal way for elimination of the target undesirable effect or to reduce search for possible causes using the method of check questions or its variations.
Automation of cause-and-effect analysis in Compinno-TRIZ
CECA is quite formalized analytical tool allowing finding the key undesirable effects. Despite availability of rules, typical mistakes can be made during CECA. These mistakes are caused by cognitive biases [4], to which any person is subjected to a greater or lesser extent. In order to eliminate impact from human cognitive biases to the maximum, the Compinno-TRIZ module has logic check functions. An example from the authors' catalogue is given below to demonstrate operation of this module. It should be noted that this comparison is meant to demonstrate the capabilities of quality improvement in elaboration of cause-and-effect chains using the Compinno-TRIZ module. Each individual TRIZ-specialist can choose to use these capabilities or not.
The problem giver offered the following problem statement. An additional tank for naphthalene collection does not fit in the shop. In order to find the key undesirable effects, they made a cause-and-effect analysis inward and obtained the following cause and effect chain (Fig. 7):
Additional process equipment does not fit in the shop
l
Duplication of the units is necessary t
The units are frequently taken out of service for repair
I
The operating unit is quickly clogged with the product
i
The product sticks to the unit wall
: i
The product is melted on the unit wall a
I
The product abruptly decelerates colliding with the unit
wall
Figure 7
Based on this chain, the following problems can be stated for elimination of the key undesirable effect and breaking of the cause-and-effect chain using the negation operator:
1. How gradual deceleration of the product can be achieved?
2. How deceleration of the product without melting can be achieved?
3. How melting of the product without sticking to the wall can be achieved?
4. How sticking of the product to the wall without quick clogging of the unit can be achieved?
5. How quick clogging of the unit without frequent taking out of service for repair can be achieved?
6. How frequent taking out of service for repair without duplication of the units can be achieved?
Figure 8
Based on this chain, problems can be formulated and contradicting demands can be immediately found as well. The list of problems and contradicting demands is given below:
1. Contradicting demands: 'The shop area cannot be changed' and 'Two units are necessary'.
2. Contradicting demands: 'Continuous naphthalene production is necessary' and 'The unit shutdown for cleaning is necessary'.
3. How the situation can be achieved, when the naphthalene stalagmite blocks the naphthalene supply tube, but the unit shutdown for cleaning is not necessary?
4. How the naphthalene supply tube can be made movable?
5. How the unit wall can be made movable?
6. How the situation can be achieved, when the naphthalene supply tube is fixed, but the naphthalene stalagmite does not block the tube?
7. How the situation can be achieved, when the unit wall is fixed, but the naphthalene stalagmite does not block the tube?
8. How crystallization of the liquid naphthalene layer can be prevented?
9. How holding of the liquid naphthalene layer by the solid naphthalene layer can be prevented?
10. How adhesion properties of the wall can be reduced?
11. How the situation can be achieved, when the wall has adhesion properties, but it does not hold the liquid naphthalene layer?
12. How heating of naphthalene particles without melting can be achieved?
13. How collision of naphthalene particles with the unit wall without heating can be achieved?
14. How collision of naphthalene particles with the unit wall can be prevented?
After comparison of the resulting two lists, the authors made the following conclusions:
• The cause-and-effect chain with logic check results in statement of almost two times as many problems;
• Accuracy of the formulations and chain building logic allows almost immediately formulating the contradictions of demands and proceeding to their resolution using the Compinno-TRIZ module ('Contradiction' and 'Principles' interfaces);
• The problems in the second list are free of deficiencies and negations, which are difficult to make, if models are used during formulation of undesirable effects in Compinno-TRIZ.
Operation principles of the CECA module in Compinno-TRIZ
The previous section demonstrated the results of using the CECA module in Compinno-TRIZ. As noted above, the cause-and-effect chains made in Compinno-TRIZ with logic check are characterized by accurate formulations of the undesirable effects. Such accuracy is achieved by means of the 'Phrase editor' and 'Component passport' interfaces in Compinno-TRIZ. Their detailed description is given below.
A 'phrase' in this context means the smallest separate speech unit with a complete sense. The following phrases can be given for example:
• The shop is a part of the plant;
• The town is a supersystem for the plant;
• The chair was a tree in the past;
• Gulls are often present by the seashore;
• The husband's mother-in-law is his wife's mother;
• It is raining;
• The kettle heats water; normal temperature of a human body should not exceed 37.6°С and should not be below 35°C.
The following word combinations are not phrases: a blue pencil, hot water, a sad person, primary aluminium, a financial crisis.
'Phrase editor' in Compinno-TRIZ is a formalised interface comprising function models, relationships, operations, and constrains of the component, which can be used for description of undesirable effects.
The process of using the 'Phrase editor' interface is based on the inventive thinking cycle [5]. The module user passes all stages of inventive thinking:
• Analysis - break a free-formulated undesirable effect (phrase) to elements and a structure of relations between these elements in the phrase;
• Synthesis - select a model (function, operation, relationship, constrain of the component), which is most suitable for the distinguished elements and structure of relations, and synthesise a new phrase;
• Estimation - compare senses of two phrases: free-formulated and synthesised according to the selected model. If the phrase senses disagree, repeat the cycle.
Let's consider examples of using the 'Phrase editor' interface for each of the phrase models taking the undesirable effects from the above example.
Example 1. Free-formulated undesirable effect (phrase) - 'Additional equipment does not fit in the shop'. The following elements are distinguished during analysis of this phrase: 'Additional equipment' and 'Shop', i.e., this phrase contains two different elements. This immediately limits the possible suitable models to two models: a function or a relationship. Indeed, an operation model refers to the same element with any changed parameter and a component constrain model comprises only one component. A function model supposes that the object parameter is changed by the function carrier. However, our case does not suppose changes of any parameter of the phrase elements. Therefore, only the relationship model remains. It should be also noted that the free-formulated phrase contains a negation, which is not permissible according to the CECA rules.
During filling of the components in the selected model, the 'Additional equipment' element is specified to 'Two units'. The new synthesized phrase ('Two units occupy an area more than the shop') actually has the same sense as the free-formulated phrase ('Additional equipment does not fit in the shop'). At that, the new phrase is more accurate and it does not contain the negation. The figure below shows the interface with the filled cells.
Figure 9
Example 2. The next free-formulated phrase for review is 'Duplication of the units is necessary'. One element can be distinguished during analysis of this phrase: 'Units'. In this case, the range of suitable models is narrowed to an operation model and a component constrain model. The phrase has no direct or indirect indication on change of the 'Units' element parameters, but it contains the 'Necessary' word, therefore the component constrain model is the most suitable after all.
During filling of the component constrain model, the 'Units' element is specified to 'Two units'. The free-formulated phrase ('Duplication of the units is necessary') and synthesized phrase ('Two units are necessary') have a close sense, therefore the cycle can be stopped. The figure below shows the interface with the filled phrase.
Figure 10
Example 3. Free-formulated undesirable effect (phrase) - 'The operating unit is quickly clogged with the product'. Two elements can be distinguished during analysis of this phrase: 'Operating unit' and 'Product'. To completely cover the sense of this phrase, a model is necessary with two components - a function and a relationship. In addition to the elements, the phrase contains a verb indicating the action of one element on another. I.e., the function model is the most suitable in this case. The synthesized phrase is 'The product clogs the operating unit' and it has the same sense as the free-formulated phrase, therefore the cycle can be stopped.
Figure 11
Example 4. Free-formulated undesirable effect (phrase) - 'The liquid naphthalene layer crystallizes'. One element can be distinguished during analysis of this phrase: 'Liquid naphthalene layer'. There are no other elements in this phrase and they are not implied. As in the previous examples, the range of applicable models for synthesis is narrowed to an operation model and a component constrain model. The 'Crystallizes' word indicates a process, rather than a constrain. Therefore, the operation model is selected. The synthesized phrase is 'The liquid naphthalene layer crystallizes to the solid naphthalene layer'. The free-formulated phrase and synthesized phrase have the same sense; therefore, the cycle can be stopped. It should be noted that the synthesized phrase is more accurate - it contains the naphthalene layer parameter changed during the operation.
Figure 12
The examples given above by the authors show how the 'Analysis - Synthesis - Estimation' cycle is implemented in the 'Phrase editor' interface of Compinno-TRIZ. Apparent advantages of using this editor:
• More accurate phrases of undesirable effects and thus more accurate formulation of problems from the cause-and-effect chain;
• Specified objects of the problem situation;
• Difficult to use negations in undesirable effects.
When the cause-and-effect chain with the synthesised undesirable effects is made, it should be
logically checked. The CECA logic checks and specification interface represents a table (Fig. 13).
И AND
» y JarrpiTbe shop area с ¿шло! be changed D у öS 3 3
Оояа ¿Tivo ' гг.': are necessary О ■M 680Î
» И AND
4- HioiTbe unit shutdown for cleaning is necessary □ esc«
i Ha^Tbe naphthalene stalagmite block: the naphthalene supply tubs □ 6807
» U AND
PaThe niphthilene stalagmite size is equal to distance from the unit wall ши to the tute О 6S13
▼ HAND
HI TThe solid naphthalene laver holds the liquid naphthalene [aver i" О 6S19
т О CThe unit wall holds the Liquid naphthalene layer □ 6ms
T BAND
« .Solid iiaphtlialene particles are melted о ✓ 1 SS OB
w ^^BNaphtlialene particles are belted D . / «BIO
Naphtlialens particles collide with tire unit will ra:oii О Mi 6821
** The unit wall has adhesion properties D у 6SI0
KB ilThe liquid naphthalene layer crystallises □ 681S
— TpThe mphthdene supplv tube ls fixed □ 6815
C iThe uni: wali is fixed □ 6S14
OdsfiContinuous naphthalene production is necessary □ 6812
Figure 13
At the very beginning of using the module, symbol is present at all undesirable effects. This means that system relations should be put between components of the undesirable effect phrases for the CECA logic check, fej The 'Phrase editor' interface contains the 'Component passport' for each entered component, where the system relations are put.
The Compinno-TRIZ module uses the following concepts of system relations [6]:
• Supersystem is a system, which includes the considered system as an integral part.
• Subsystem is a system object, which can be represented as an independent system comprising elements and with certain integrity. The subsystem elements form a subset of the system elements set. Each system may have many various subsystems.
• Neighboring system is a system (or a system element) of the same order with the considered system, which has established or potential capabilities to establish direct interaction with the considered system through one or another field of interaction (physical, chemical, biological, social-cultural, economic, legal, etc.). A neighboring system cannot be a supersystem or a subsystem for the considered system.
Let's consider some phrases for better understanding of the logic check operation.
4 Two units are necessity □ Ld íso;
▼ AND
4. The unit shutdown. for cleaning is necessary o Ld 6S06
ta^Tt16 naphthalene stalagmite blocks the naphthalene supply tube tata 6807
Figure 14
The synthesized phrase ('Two units are necessary') is a component constrain by its structure. Its component is 'Two units'. The next phrase ('The unit shutdown for cleaning is necessary') is also a component constrain and the component is 'Unit'.
The unit is a part of two units. Therefore, the system relation is put in the 'Unit' component passport that 'Two units' is a supersystem for 'Unit'.
Component passport
Xame .Unit Component parameters
Source Project ontology, phniet Relations The unit shutdown for cleaning is neceîsaiy □JFifild rlPrecess rIFlow ¡¡Canonical (basic) ivord form
Name System relations
Two uiiits Supeis-vitera
Figure 15
When this system relation between the components of two phrases is put, fc^i symbol changes to * . This means that the analysis is made 'inward' the system.
After that, the specialist, who makes the analysis, compares this direction with the intended analysis direction. If the actual and intended directions agree, a transition to the next phrase is made. Otherwise, the phrase itself and its components should be specified or the system relations should be reviewed.
Let's consider the next phrase in our example. 'The naphthalene stalagmite blocks the naphthalene supply tube' has two components - 'Naphthalene stalagmite' and 'Naphthalene supply tube'. The naphthalene supply tube is an integral part of the unit, therefore the system relations in the 'Unit' component passport are as follows:
Figure 16
symbol changes to
The analysis direction remains
Similar to the previous phrase, unchanged.
If a logical fault occurs and the analysis direction changes, the algorithm alerts about this with fed symbol at the incorrect phrase and a tooltip. In this case, the phrase, its components, or system relations should be specified.
The following symbols can be encountered in the CECA module:
- Mistake alert: no system relations put between components of effect and cause phrases, change of analysis direction, logical skips;
+ - Analysis direction 'inward', from system to subsystem; t - Analysis direction 'outward', from system to supersystem;
. ~ - - Analysis direction 'in place', specification of undesirable effect phrase;
m - Analysis direction 'in plane', from system to neighbouring system. When all mistakes in the cause-and-effect chain are eliminated, one can proceed to formulation of problems. Each of undesirable effects in the chain can be selected for formulation of a problem for its elimination and a problem through a negation operator:
II
ESO
CcapaEecdœdf
.TlWiicp'üffiClClDt-tS'Chnt.^i' / mil
» t TwOHElli-lfSïSûliy'' ^¿j HI!
- ANDH
1 Tins ut'ihËilm^ûîaiiÉgiïmôHiyll □ MM K»
* Hiî mhtialtci ilalzpJe hbisliï'iizptiâHlHie-njiplY hitir
IgM t
Fh" rnt; -l-c -rë.in''; ;~Vl -Tk en npli tluU f.nla;mira.Mnrt\Tli ;■!! :plit1ialaii npp.lv
tnli'r ' □ H
'TkinifiktbaJeLÉ-iislBjpiuiÊ'blKb-feiâpfcislEiii iupply'ube', jjfurjcT dis-irraurah' a-aflfcrt -TliiniiilThiiriinra f«r rljanin;-ii n imiin'1'
Figure 17
Any of the automatically formulated problems can be taken on to an individual cycle or a solution idea can be formulated and saved to the idea registry.
SUMMARY
In view of specific character of production TRIZ-projects, CECA is one of the most demanded analytical tools. At the same time, the authors demonstrate the typical mistakes made by TRIZ-spe-cialists of various levels and experience due to cognitive biases inherent to each person: omission of intermediate undesirable effects, fuzzy formulation of undesirable effects, establishment of false cause and effect relations, uncontrolled U-turn of analysis direction.
The authors' overview of references dedicated to increase of cause-and-effect analysis accuracy including that through automation shows that most of them are focused on creation of graphic editors, which certainly facilitate efforts of TRIZ-specialists in drawing of analysis results, but do not protect against typical logical mistakes in building of cause-and-effect chains.
The Compinno-TRIZ CECA module with logic check allows obtaining cause and effect chains with more accurate formulations of undesirable effects without negations and fuzzy formulations. At that, contradicting demands can be immediately found. The authors also demonstrate that intermediate undesirable effects are not omitted thanks to more accurate formulations, therefore more problems are stated as well. Furthermore, using the CECA module with logic check, phrase editor, and component passport is training of the user's inventive thinking.
REFERENCES
1. Jerzy Chrz^szcz, AN OVERVIEW OF ADVANCED CECA METHOD, Proceedings of MATRIZ TRIZfest-2022 International Conference. August 31- September 1-3, 2022
2. Jerzy Chrz^szcz, TOWARDS AUTOMATION OF CAUSE-EFFECT ANALYSIS, TRIZ Review: Journal of the International TRIZ Association - MATRIZ. Vol 2/1. April 2020, TRIZ ReviewJournal (matriz.org)
3. Chrz^szcz J., Salata P. Cause-Effect Chains Analysis using Boolean algebra. In: Advances and Impacts of the Theory of Inventive Problem Solving. The TRIZ Methodology, Tools and Case Studies, 2018, ISBN 978-3-319-96531-4, pp. 121-134.
4. Daniel Kahneman. Thinking, Fast and Slow: AST, Moscow, 2014
ISBN 978-5-17-080053-7
5. N. Rubina. Inventive Thinking: Formation & Diagnostics // Concept. - 2015 - #02 (February) Inventive Thinking: Formation & Diagnostics (cyberleninka.ru)
6. Collective of authors: Ramez Kassu, O. Kraev, A. Kulakov, I. Misyuchenko, N. Rubina, M. Rubin, A. Trantin. Generally edited by M. Rubin. TRIZ-glossary 2023, International Public Institution 'Summit of TRIZ Developers'.
7. T. Ohno. Toyota Production System. Beyond Large-Scale Production. Institute for Complex Strategic Studies, 2008 ISBN 978-5-902677-04-1.
8. K. Ishikawa. What is Quality Control? N.Y., 1985.
9. W. Dettmer. Goldratt's Theory of Constraints: A Systems Approach to Continuous Improvement. Alpina Publisher; 2012 ISBN 978-5-9614-1952-8
10. https://habr.com/ru/companies/ods/articles/544208/
11. https://easyrca.com/
12. https://www.sologic.com/en-us/rca-software/overview
13. A. Olivkevich. Experience in Application & Development of Cause & Effect Analysis (CEA) Tools. IX International Conference. TRIZ. Practical Application & Development of Methodical Tools, 2017.
14. https://taproot.com/
15. https://clickup.com/
