Editor | On 25, Jul 2001

Tz-Chin Wei
Flotrend Corporation
3F, 72 Sungteh Rd.
Taipei, Taiwan 110 R.O.C.
peterwei@ms9.hinet.net

Semyon D. Savransky, Ph.D.*
The TRIZ Experts
6015 Pepper Tree Court, Newark, CA 94560, USA
TRIZ_SDS@hotmail.com

1. Introduction

Many attempts to compile rules of thumb or heuristics that were used by inventors have been carried out purposing to guide designers toward the most promising solutions [1-6] . The research methods are varied with the subjects of study. In general, these methods can be classified by

– studying the thinking process of creative persons in erection of their works,

– studying the inventive works regardless of their creators thinking features.

While the first approach, popular in Western countries, leads to development of cognitive science and Artificial Intelligence, the second approach (recently extended from the former USSR to many countries) leads to a quite different methodology for problem solving, known as TRIZ [3,1] .

Based on numerous patents fund (one kind of creative works that contains much deep technical knowledge) analysis, G. S. Altshuller and his co-workers identified the Inventive Principles for resolution of pair (technical) contradictions (or GSA Principles for short) that work as generic rules of thumbs for many engineering fields [3,1] . Moreover, the patents fund, as the primary sources of study, proved that the technology-based TRIZ heuristics have stronger advantages then those heuristics that were developed using psychological or cognitive approaches.

GSA Principles contain many sub-Principles that provide detailed recommendations that help inventors use these heuristics more effectively. However, many of GSA sub-Principles are still too general for use (e.g. Principle 1A “Divide an object into independent parts”), and it would be more beneficial to add specific interpretations that bring GSA Principles closer to problem solver’s fields. It is known that not every issued patent is implemented. Perhaps this is the reason why GSA Principles lack specific sub-Principles.

Moreover, G. S. Altshuller constructed the Matrix for the resolution of [3] (technical) contradictions resolving [3] that has been extended by one of the authors [1] . The statistics of GSA Principles in the extended Contradiction Matrix [1] is shown in the following figure. On average, each principle is used about 106 times in all cells of the extended Contradiction Matrix **.

On the another hand, A. I. Polovinkin and his co-workers [4,5] took a similar technology-based approach for heuristics of systems transformations, but they restricted the knowledge base of their research to implemented designs created by highly experienced engineers in the former USSR. They have identified about two hundred heuristics that allow problem solver’s to resolve design and technical problems [4,5] . The most culturally independent Polovinkin’s heuristics are denoted as 129H and have been classified into nine groups [1, 2] .

In this article, we will make comparisons between 129H and GSA Principles for the following purposes:

– to find what is common between the two independently developed sets of heuristics that used dissimilar sources of creative works and conducted the research in different periods of time.

– to find whether 129H and GSA Principles can enrich each other.

– to find new Inventive Principles and sub-Principles.

– to extend Contradiction Matrix for resolving contradictions.

We use numeration of reference [1] for 129H and GSA Principles here for a reader convenience.

2. Comparison Results

2.1 Altshuller’s Principles perspective

Some GSA Principles have all their sub-Principles directly related to 129H (“fully directly related”), while some Principles have none of their sub-Principles directly related to 129H (“not directly related”). Remaining Principles have some of their sub-Principles directly related to 129H, and are further classified into “strongly directly related” and “weakly directly related” categories. Table 1 summarizes these results. Detailed correlations between 129H and GSA Principles will be presented in our forthcoming book [6] .

Table 1

Relations with 129H	Principles
Fully directly related (11 items)	#5, #6, #7, #12, #14, #16, #21, #27, #31, #33, #40 Average usage in the extended Contradiction Matrix is 64.9
Strongly directly related (9 items)	#1, #4, #8, #13, #15, #19, #20, #26, #39 Average usage in the extended Contradiction Matrix is 117.5
Weakly directly related (8 items)	#3, #10, #17, #25, #28, #29, #30, #35 Average usage in the extended Contradiction Matrix is 172.6
Not directly related (12 items)	#2, #9, #11, #18, #22, #23, #24, #32, #34, #36, #37, #38 Average usage in the extended Contradiction Matrix is 81

Principles can further relate to 129H in the following “indirect” ways:

– incorporate 129H as specific interpretation of sub-Principles

– incorporate 129H as specific interpretation of Principles

– fit into 129H that have more general interpretation

Therefore, “Weakly directly related” Principles may become “Firmly indirectly related” Principles if they gain more associations with 129H through the above indirect ways. Table 2 shows the result of such transitions.

It is interesting to note that Principles #9, #23, #25, #30, and #34 have the least weak relations with 129H, while non-Altshuller’s Principles A, D, and E have no relation with 129H. We need to remark here that 129H was developed for technical (sub)-system transformations only, while GSA Principles can be selected for resolving any type of pair (technical) contradictions [1] .

Table 2

Relations with 129H	Principles
Firmly indirectly related (from previous weakly directly related ones)	#3, #10, #17, #28, #29, #35 Average usage in the extended Contradiction Matrix is 211
Firmly indirectly related (from previous not directly related ones)	#2, #11, #18, #22, #24, #32, #36, #37, #38 Average usage in the extended Contradiction Matrix is 104.7
Lightly indirectly related (from previous weakly directly related ones)	#25, #30 Average usage in the extended Contradiction Matrix is 57.5
Lightly indirectly related (from previous not directly related ones)	#9, #23, #34 Average usage in the extended Contradiction Matrix is 57
Not directly related: i) non-Altshuller’s Principles ii) GSA sub-Principles	#A, #D, #E 9A, 17D, 17E, 23A, 23B, 25A, 25B, 30B, 34B

2.2 Polovinkin’s Heuristics perspective

The relations between 129H and GSA Principles or/and sub-Principles are summarized as the following table 3.

Table 3

Relations with GSA Principles/sub-Principles	# of heuristics	%
Relate directly with GSA sub-Principles	31	24 %
Become specific interpretation of GSA sub-Principle(s)	39	30 %
Become specific interpretation of GSA Principle (also candidates for possible new GSA sub-Principles)	33	26 %
Example of heuristic related directly with GSA sub-principles (heuristics are considered more general than GSA sub-Principles)	5	4 %
Become general interpretation for several GSA sub-Principles (also candidates for possible new Principles)	18	14 %
No relations	3	2 %

About 80% of the 129H can be inserted in the framework of GSA Principles. 18% of the 129H (heuristics 1.9, 2.6, 2.8, 3.2, 3.6, 3.15, 4.1, 4.6, 5.4, 5.5, 5.10, 6.6, 6.9, 6.12, 6.16, 6.21, 6.22, 6.23, 7.7, 8.7, 8.13, 9.1, 9.2) are more general than GSA sub-Principles and can have sub-Principles as their specific interpretation. For example, Heuristic 4.1 can include sub-Principles 9A, 9B, 10A, and 11A, while Heuristic 6.22 can include sub-Principles 18D, 28D, 29D, 31A, 32C, 34A, 39B and 40A, and then Heuristic 8.13 may include sub-Principles 8B, 26A, 28B, 29C, 32A, 36A and 37A.

Finally, there are three heuristics 3.16, 8.8 and 8.14 that seem too general to relate with GSA Principles. Note, that the Heuristic 8.8 relates more to the Ideality concept, while the Heuristic 8.14 relates directly to the so-called Tend of Uneven Development of System.

129H adds several specific interpretations to twenty-three GSA sub-principles (1A, 2A, 2B, 3A, 3B, 4B, 5A, 6A, 13A, 14B, 15A, 15C, 17A, 17B, 19A, 20B, 28A, 28B, 29A, 31A, 33A, 35A, 35B) of seventeen Principles, as well as to Principle C “Use of Pause” and Principle F “Concentration-Dispersion” (see [1] ). 129H also adds possible additional sub-Principles for a dozen of GSA Principles (#1, #3, #5, #6, #10, #13, #16, #17, #19, #26, #33, #35).

2.3 Cross-reference of 129H and GSA Principles

129H are grouped into nine classes of typical transformations of system [1,2] . Table 4 summarizes the correlations between nine classes of 129H and GSA Principles.

As it is noted in Ref. [1] some non-Altshuller’s Principles can be considered as GSA sub-principles, for example, B is sub-principle to the Principle 34, C is sub-principle to the Principle 19 and F is sub-principle to the Principle 35.

Empty cells can be used to generate new heuristics. For example, the intersection of “Preliminary anti-action” (Principle #9) and “Structure Transformation” (Heuristic class #2) might spark a useful hint like “TAKE A BLOW UPON ONESELF”, or “Preliminary anti-action” (Principle #9) and “Space Transformation” (Heuristic class #2) together might spark a hint “HIDE IT INTO A SAC”, or “Preliminary action” (Principle #10) plus “Expedients of differentiation” (Heuristic class #7) might spark a useful concept like “Pre-selection” (e.g. “SELECTIVE ASSEMBLY”). Such new heuristics should be verified with the patent fund as it is described in [1] and implemented in TRIZ.

Table 4: GSA PRINCIPLES AND 129H CORRELATIONS

Principles –	129H classes– see Ref [1,2,6]
see Ref [3,1]	Shape	Structure	Space	Time	Motion	Material	Difference	Quantity	Evolution
1			Y	Y			Y	Y	Y
2		Y	Y		Y	Y	Y
3	Y	Y				Y	Y	Y	Y
4	Y							Y
5		Y	Y	Y		Y	Y	Y	Y
6	Y	Y						Y
7			Y
8					Y			Y
9	Y			Y
10	Y	Y	Y	Y		Y
11				Y
12					Y
13			Y	Y	Y	Y
14	Y				Y
15	Y	Y			Y	Y		Y	Y
16								Y
17			Y		Y
18					Y	Y		Y	Y
19				Y	Y			Y	Y
20				Y	Y
21					Y
22						Y		Y
23							Y
24			Y						Y
25						Y
26	Y	Y				Y		Y
27		Y				Y
28		Y			Y	Y		Y	Y
29		Y				Y		Y
30						Y
31	Y					Y
32		Y				Y		Y
33	Y	Y				Y
34						Y	Y
35						Y	Y	Y	Y
36						Y		Y
37						Y		Y
38						Y
39			Y			Y
40						Y
A
B							Y
C				Y					Y
D
E
F		Y					Y	Y

(Mark “Y” is used for those GSA Principles that have correlation with 129H.)

3. Conclusion

Only a small portion of 129H (~24%) and GSA Principles (~30%) are directly related. The remaining part of 129H and GSA Principles enrich each other significantly. Some GSA sub-Principles find their specific interpretations in 129H, while some of 129H further add additional sub-Principles for several GSA Principles. Some of 129H find their specific interpretation in GSA sub-Principles and therefore open doors to finding other new sub-Principles.

The simple statistical analysis shows that most of 129H and GSA Principles firmly indirectly related or weakly directly related in terms of Principles usage in the extended Contradiction Matrix. Therefore, some 129H can be included in the modern Contradiction Matrix.

By correlating nine classes of 129H and GSA Principles in the table presented in the previous section, we can expand the number of heuristics for technical problem solving even more. We hope to “discover” such new heuristics in the future.

The authors would like to thank Marco A. de Carvalho (Brazil) and Andrey P. Khvostov (Russia) for stimulation discussions.

References:

1. Savransky, S. D., Engineering of Creativity: Introduction to TRIZ Methodology of Inventive Problem Solving, CRC Press, 2000.
2. de Carvalho, M. A., Wei, T. C., Savransky, S. D., “Validation of Heuristics for Systems Transformations”, in: Proceedings of TRIZCON2001, Altshuller Institute, Woodland Hills, CA, 2001.
3. Altshuller, G. S., 40 Principles: TRIZ Keys to Technical Innovation, Translated by Lev Shulyak, Technical Innovation Center, Worcester, MA. 1998.
4. Polovinkin, A. I., Theory of New Technique Design: Laws of Technical Systems and their Applications, Informelektro, Moscow, 1991 (in Russian).
5. Polovinkin, A. I., The ABC of Engineering Creativity, Mashinostroenie, Moscow, 1988 (in Russian).
6. Belousov, V., Doncean, G., Plahteanu, B., Salamatov, Yu.P., Savransky, S. D., Wei, T. C., de Carvalho, M. A., et. al. Guide for Inventors, RO-INI (to be published in 2001) ***.

Footnotes:

* Corresponding Author
** Thus it can be enough to study only 15 “high-used” Principles (from #34 and above in the figure 13.1 from Ref. 1) in short (less than 2 weeks) TRIZ courses.
*** The book will be available soon, please see http://www.jps.net/triz/PR_GFI_Book.htm