The aim of this note is the critique of the style of some TRIZ research and publications on the example of the Tool, Object, Product Function Analysis (TOP) by Zinovy Royzen. I strongly believe that such examination would lead to improvement of TRIZniks works. The object of my inquiry is chosen not by reason of the most awful case (there is a lot of even worse publications), but because some TRIZ neophytes could uncritically accept TOP from the author who conducts a quite good TRIZ workshops. TOP as well as its protoplast the Altshuller's Su-Field Analysis deals with the existing technique and helps figure out ways of its improvements. Hence this part of TRIZ is quite important and, unfortunately, is not well-developed yet. Therefore, development of Su-Field Analysis should be interesting for many TRIZniks.

The paper under consideration was presented at the First Symposium on TRIZ Methodology and Application of Altshuller Institute for TRIZ Studies, March 7-9, 1999, Novi, Michigan, USA (Proceedings - pages 17-30) and "re-printed" at the Internet in the most popular TRIZ periodical in English www.triz-journal.com [1]. To the best of my knowledge this paper is the only one relatively full publication about TOP. It contains 3 parts - overview of TRIZ (about 45 % of the article), the TOP itself (only about 10 % of the paper), two case studies of TOP application (again about 45 % of the article body). The list of references concludes the paper.

Let me skip the first part of [1] that contains well-known repetitions about TRIZ including a survey of the methodology parts that are not important for the main theme of the publication and start the discussion with TOP directly.

First of all the author claims that "[T]ool-Object-Product (TOP) Analysis, the next generation of Substance-Field Analysis, was developed by Zinovy Royzen in 1989" but he refuses to show up copies of early TOP publications (A). The search of the most comprehensive library (Chelyabinsk Fond of TRIZ Literature) indicates that Z. Royzen published only a few papers about Altshuller's Standards and ARIZ at that time. In personal communication the author admits that these papers have nothing to do with TOP, so year 1989 is irrelevant (B). To the best of my knowledge, TOP is mentioned at the first time by Z. Royzen in the case study [2] published in 1998. This date is at least 10 years later than the description of similar analysis in terms of the functions through object-tool-product link (FOTP) in software engineering [3]. The results of Ref. 2 are included in [1] and duplicated in [4] (both papers published in 1999). Unfortunately, by the above mentioned reason, I am not familiar with some Royzen's papers (the references 5 and 7 in [1]) that were published in 1995-1997, so TOP may appear a little bit early.

Z. Royzen describes the essence of his model in the following three sentences as

"[T]he simplest useful function has four components. It has the tool of the function (or the function provider), the object of the function (or recipient of the action of the tool), the action of the tool at the object, and one more component - the product of the function. The useful function of the tool is to obtain the product of the function from the object."

The reader can conclude from the first and last phrases that TOP is only about useful functions. Fortunately, later on the author includes into consideration the idea about harmful part of technological activities. But this deficient statement is only "a light inaccuracy of a genius" to compare with mess with so-called four components of a function in the middle sentence.

Z. Royzen understands a function as something separate, unjustly existing itself and detached basically from the tool [1, 2, 4]. Moreover, according to TOP a function can create a tool, object and product, i.e., three different technical systems as well as a technological process. Probably I was not so lucky, but I have never seen during forty years of my life experience that the function "to cut wood" produces itself a tree (object), or an axe (tool) or a firewood (product). I have also never seen how three kinds of substances: tree (alive object), axe (metal) and firewood (insulator) select themselves at least one kind of field (mechanical) that is needed for the function "to cut wood" performance (or for "[t]he action of the tool" in the term of [1]) through many possible behaviors of these objects (C).

Any person who at least attends any introductory philosophical seminar must be confused by the cited statement. Let me explain why. Although there are numerous philosophical approaches all of them are based on two contradicting paradigms about essence of the world, i.e., materialism and idealism that were known even to ancient Greeks. The first paradigm states that a matter causes all phenomena, while the second one states that all that exist are the product of minds or ideas. It is necessary to stress that TRIZ (at least its so-called "hard" part) follows mostly the first concept (e.g. Contradictions, Evolution, etc.), but includes some elements (e.g., Ideality, when physical objects or processes named in TRIZ as the Ideal Technique have no existence outside of the mind that is conscious of them) of the second one [5,6]. Often TRIZ is named as applied dialectical materialism (D). Most of skilled TRIZniks has no doubts that any technique is a matter created due to ideas of engineers and designers (who are also a kind of matter). They know that mental and physical phenomena are distinct in the world of technique. Majority of seasoned TRIZniks has no doubts also that any function is an idea or more precisely "an abstract or manifestation of behavior" of a matter (see the detailed discussion in the paragraph 3.5.3 of Ref. [6]). According to Z. Royzen we must put things upside-down: while function (idea) is primary, the matter objects as a tool as well as a raw object and product (named as "[t]he object of the function (or recipient of the action of the tool)" and "the product of the function" in [1]) are secondary. They appear only to satisfy the existence of function even in the existing technique (the major field of application for Su-Fields and TOP). Of course any (even quite "smart") technique does not "know" which function from its possible set should be performed. For example, a simple knife (tool) can as easily cut (action #1) as drill holes (action #2) in a soft material (object) in order to perform the function "to change shape" and to make two different products.

The TOP description in the paper [1] gives also an impression that the author perceives usefulness (harm), as some own property of a product and/or function, regardless to human opinion. The product (technical system or process) or function (as well as a behavior or action) alone cannot be harmful or useful - these attributes come into the game with the human's perceptions. The division of the technique's functions into good or bad, primary and secondary is not absolute. Actually any function does not exist itself, it is only an abstraction. Additionally any label of a function (as well as of an element of a technique) is arbitrary (E). The exercises mentioned in the footnote (C) of this work should help TRIZ students handle these two abstraction properly.

Case studies [1] deal with two quite different techniques - computer hard drive and space shuttle class main engine (F). Although these case studies should help understand TOP better but they have two weaknesses. The representations of the both problems seem incomplete because a reader gets the pieces of information that is absolutely necessary for solution (about presence of water in the engine and coil in the drive's arm) only at the very end of the each case studies descriptions. The first task also presented in the way that the result that we should reach is obviously known. Hence, all reasonings on a point of issue have the purpose to receive this result. It reminds me the pupils who are trying to adjust the course of the decision to the beforehand known answer in their homework. Let me also note that the nice color photography of a hard drive (fig.7) and a little bit more pertinent engine schema (fig. 10) that were designed mostly without application of TOP seem superfluous. These case studies are also duplicated in other Royzen's presentations at various conferences (see [2, 4] and references 6, 8 in the paper [1]), so a scrupulous researcher of Z. Royzen activities could wrongly conclude that TOP can be applied only to the above mentioned techniques.

It is easy to demonstrate that the problems represented in these case studies can as easily be solved with the familiar Altshuller's Su-Field Analysis [5, 6] as with TOP (G).

The eleven references in the paper [1] end are not cited in the body of article at all. Probably a reader should himself/herself assign these references to parts of the article. Note, that the list contains self-citation of eight (i.e., 73%!) Royzen's works mostly at various conferences during 1993-1999, and absolutely no citation of very important results in the frameworks of FAST, FOTP, SADT, IDEF and other models that are similar to TOP (H). The experienced TRIZnik would also be surprised that names of several colleagues (except G. S. Altshuller) whose ideas Z. Royzen used in the paper [1] as well as in other his publications are not mentioned. Unfortunately, it was the bad habit of the methodology creator to ignore results of others TRIZniks, and many TRIZ masters uncritically accepted such manner of publications that also push our methodology outside scientific community.

Let me leave the paper [1] and consider TOP shortly outside this publication.

Z. Royzen argues in recent private discussion (Fall 2001) that TOP in contrast to Su-Field can describe the work of a simple projector that consists of a light bulb and a mirror. Anybody who knows Su-Field Analysis and physics easily conclude that the projector can be represented graphically as

Here F1 is a photon (that in this case can be presented as a wave of electromagnetic field rather as an elementary particle) emitted by the light bulb (due to its high temperature) with speed vector directed from the bulb to the mirror S, and F2 is a photon reflected by the mirror (due to intrinsic properties of metal) with speed vector directed from the mirror to any object in space that should be illuminated. Let me note that this picture is shown also at page 192 of Ref. 6 (upper right), where the Su-Field Analysis is reviewed.

During the discussion Z. Royzen also claims that TOP is good because it is "used in many cases for long time". As the proof of this argument he is able to refer us only to the single paper about a task in PERL programming published just recently by Michael Schlueter [7]. Although TOP is mentioned in this paper, the author (a former Royzen's student) incidentally uses in fact the software tool FOTP [3, 8]. In general I believe that the argument like "my work is perfect because other scholars use it" cannot be valid. Just recall history from the Ptolemeus vision about the universe to the Marx-Lenin approach for development of human society.

To my great surprise Z. Royzen did not recognize probably the only one positive TOP aspect - the hidden time arrow that exists in any model for process, i.e., a transformation of one state (e.g., raw material) into another state (e.g., product). Here we again use in TRIZ very general philosophical and physical concept of materialism that anything exists and evolves in time and space [5, 6]. This concept is represented relatively good in the so-called evolutionary part of TRIZ and poorly furnished in the Su-Field Analysis. We should unhide the time arrow for description of any technological process in TRIZ.

Unfortunately now most TRIZniks only passively use the results of others (sometimes without a mention of the authors) and do not attempt to improve our methodology. Probably they afraid to get trivial or even wrong results. There is the famous proverb "Only a non-working person does not make mistakes". It is known also that more than 60% of publications in the most famous and influential physical journal "Physical Review" are incorrect. I think that it is better to conduct research, to make mistakes, to recognize and admit these mistakes than not to develop TRIZ at all.

In conclusion, I would like to stress out three points:

a) TOP is not the best (as well as not the first) attempt to "improve" Altshuller's Su-Field Analysis. Actually TOP is weaker than other similar analyses.

b) I believe that TRIZ experts must conduct active and intensive research in order to develop this methodology. Any TRIZnik must take into account other related ideas outside our methodology in order to make TRIZ better.

c) Results of TRIZ research (even if they carry a new paradigm) should be reported in the framework of regular scientific traditions. Otherwise TRIZ will never be accepted seriously in academia.
 

Footnotes

(A) In contrast to the whole scientific community many TRIZniks often ignore requests for their publications copies. These TRIZniks do not understand that their rejection on a request creates an impression of publication’s poor-quality in the eyes of the author him/her-self and the person outside of common scientific culture. In addition an author cannot expect that others would cite his/her work if he/she creates barriers (including language) for their distribution. In opposition I have always received research papers in physics, material and computer sciences, economy and finance, engineering, philosophy and other disciplines of my current curiosity from scientists around the globe. So the question "Is TRIZ a science?" appears.

(B) Unfortunately such situation is not unusual now in our field there a commercialization of TRIZ is going on. Some TRIZniks declare that they have a long history of research activity that cannot be proven.

(C) There are three popular TRIZ exercises for students: 1) to mention as many as possible potential functions of a given technique, 2) to name as many as possible different technical systems (tools) that can perform the same function, and 3) to separate all functions collected during the first quiz into useful, harmful and neutral and to explain the reasons of such division.  Probably even some seasoned TRIZniks need to workout these exercises occasionally.

(D) A good experienced TRIZnik should be at least aware about philosophy and a TRIZ expert should navigate perfectly in thoughts of this science in the ultimate case.

(E) Probably this fact alone makes impossible the full computerization of TRIZ.

(F) This is really nice because TRIZ itself is not limited by one or a few engineering disciplines and a good TRIZnik should be able to solve problems in various fields.

(G)  I think it would be a good training for the TRIZ amateurs to analyze these case studies in the common Su-Field Analysis terms.

(H) I do not provide references here because I think it would be a good exercise to find them for whose colleagues who often ignore publications outside TRIZ. Hint here is to use Internet search engines.
 

References:

1. Z. Royzen, Tool, Object, Product (TOP) Function Analysis. www.triz-journal.com. September 1999 (http://www.triz-journal.com/archives/1999/09/d/index.htm)

2. Z. Royzen, TRIZ Solves a Reliability Problem Concerning a Space Shuttle Class Main Engine. The ASI 4th Annual International Total Product Development Symposium, November 18-20, 1998, pages 43-50, City of Industry, California, USA.

3. H. Rex Hartson and Deborah Hix, Toward Empirically Derived Methodologies and Tools for Human-Computer Interface Development. Virginia Polytechnic Inst. and State University 1988.

4. Z. Royzen, The Ideality of Products, Substance-Field Analysis, and Standard Solutions. The First European TRIZ Congress, Vienna, January 21, 1999, pages 1-13, Vienna, WIFI, Austria.

5. G. S. Altshuller, To Catch an Idea. Introduction in the Theory of Inventive Problem Solving. Nauka, 1986 (in Russian).

6. S. D. Savransky, Engineering of Creativity: Introduction to TRIZ Methodology of Inventive Problem Solving. CRC Press, 2000.

7. M. Schlueter, TRIZ for Perl-Programming. www.triz-journal.com. May 2001 (http://www.triz-journal.com/archives/2001/05/a/index.htm).

8. D. Garlan and M. Shaw, An Introduction to Software Architecture. In V. Ambriola and G. Tortora (ed.), Advances in Software Engineering and Knowledge Engineering, Series on Software Engineering and Knowledge Engineering, Vol 2, pages 1-39, World Scientific Publishing Company, Singapore, 1993.