• Letters are always welcome. Write to . Please mention if you do not want your letters published.
• Readers are asked to respond directly to the author when appropriate.

Re: Assessing The Accuracy Of The Contradiction Matrix For Recent Mechanical Inventions by Darrel Mann, TRIZ-Journal, Feb. 2002.

The Study

I pay tribute to Mr. Mann in conducting such a useful study for verifying TRIZ applicability. The outcome is far from a modest result. I think we need more studies like this and as pointed there is a need to renew the Matrix with additional Product Characteristics as well as additional new Principles.

Especially there is a need to expand the Matrix to serve not only the eminent electrical and electronics industry but also others e.g. SW development work. There is probably a huge market for that.

Who might be the owner of this development work? Who should fund such a project? Maybe the European Commission could provide the assets?

Additional Study Proposed

The nature of patents is to protect one’s inventions against unauthorized users. But often people even apply patent as a countermove to somebody else’s patent. These are often very superfluous, made primarily for product advertisement purposis to compete known or recently patented competing products.

So the question arises whether the remaining ‘not matching the Principles or Matrix’ patents were made just to compete some other patents. Maybe they are not at all significant ones, but made in desperate situation. Or perhaps they should have actually used the Principles given in the Matrix? Could they be better if the ‘right’ Principles were applied? Maybe the competing patents match the Matrix?

This could be very exciting to find out and should be the topics of the next studies.

(which will be published later this year) are that the 48% figure is, if anything, on the high side. The thing I find most odd about Richard's letter is the comment about 'imaginative problem solvers'. My experience in this regard is that the imaginative problem solvers are the ones who will almost definitely NOT use the Matrix - many exhibiting an endearing stubborn-ness that refuse to let them stop until they have developed a better solution than anything the Matrix might have recommended.

Nevertheless, I believe the Matrix is a very useful tool; it encourages people to think hard about what the contradictions in a system are, and it can give good results. The reason CREAX is committing so much energy to updating it is partly our belief in the concept and partly our belief that the implementation could be very much better than it currently is.

From: Richard Kaplan

I have enjoyed reading Darrell Mann’s many contributions to the TRIZ Journal. However, I am concerned that some readers may jump to the conclusion that the Contradiction Matrix (CM) is only 50% effective, based on its efficiency for predicting useful principles for mechanical invention patents cited in his February 2002 TRIZ-Journal “Assessing …” article. A reader of Mann’s article may infer that the selected patent is the ultimate solution (and, hence, best comparison), that the principle used in the patent is the best for solving the problem, and that the CM was not useful in generating solutions to about half the problems solved by the patents, all of which may not be the case.

I believe that the 50% number may well be underestimating the benefits of the Matrix, especially to an imaginative problem solver. It appears to me that to assess this, three levels of solutions to the selected problems “solved by the patents” need to be considered. These are problem solutions that are inferior to those patented, as good as those patented, and (perhaps) even better than those patented. CM - suggested principles NOT used in the patent may yield problem solutions that are inferior, as good, or (possibly) even better than the “wow” patent solutions used in the analysis! Boris Zlotin, Ideation International, Incorporated’s Chief Scientist, told me that he had solved his first 80 problems after learning about TRIZ using just the CM. However, a number of CM - derived solutions would not be implemented, due to impracticality, technical difficulty, unsatisfactory economics, or availability of better solutions.

Therefore, while Mann’s analysis is useful and interesting, it should only be viewed within its stated confines, i.e. showing the CM’s efficiency in predicting the principle used in the cited patent. However, use of the CM may be helpful in a significantly larger percentage of problems.

From: Darrell Mann:

The specific title of the piece was intended to set the theme that, of course, the exercise was looking at a very small slice of a big cake. (Having said that, if anyone can point me towards a bigger dataset, I would be very interested in seeing it.) As far as possible the analysis exercise conducted for the article was intended to mimic the conditions for the original research - i.e. select a 'good' patent and see what contradictions the inventor tackled and what Principles they used. The results showed that - for the small sample size - if we were devising a new Matrix today, it would have a different content than the classic version we have.

The motivation for the work was the result of considerable feedback from users of the Matrix, many of whom took great delight in informing me that they solved a contradiction using an Inventive Principle NOT recommended by the Matrix, and many more who tell me they fail because they could not connect the suggestions made by the Matrix to their problem.

CREAX is analysing many thousands of patents from all disciplines in the course of 'really' updating the Matrix. The results of that analysis