How effective is TRIZ ?

I got interested in TRIZ as a freshman at university in 1973. This happened not because I was seeking to improve my problem solving abilities but because I wanted to study the thinking process. In the course of time I got involved in the work on producing new TRIZ heuristics by imitating Altshuller's techniques. He did it by amassing arrays of interesting inventions that seemed to have a common idea behind them. And then he tried to formulate this idea in the form of a heuristic rule of making all similar inventions.

It was a breathtaking exercise but I never really applied my heuristics to real problem solving. For some reason the problems which I faced never fell into the domain of the applicability of the heuristics. Neither Altshuller's heuristics did. The entire TRIZ seemed to be inapplicable to problems which I had to solve. That is why I was always interested to know how effective TRIZ really is.

Unfortunately, all "evidence" falls into two categories:

  1. Somebody tells that TRIZ helps him or her (without elaborating on how specifically it helps) or that he/she solved so-and-so problem with the help of TRIZ (without proving that).
  2. Somebody in a company claims that the company has a tremendous economic effect from utilizing TRIZ and that this many inventions were accomplished with the help of TRIZ.

Needless to say that such "evidence" cannot be taken seriously. That is why I always ask people for more details. But never hear anything in response.

Here is an example. Valeri Souchkov claimed in "Business Week" that people, who took his TRIZ training sessions, were able to successfully solve problems by TRIZ, which they could not solve before. As an example he cited the invention of the surgical tool, which is able to adjust to the shape of blood vessels. The solution was to make the tool to consist of many micro-segments. He did not tell whether these segments were connected, forming a chain, or were disconnected. But this is a very important detail. If they form a chain, then why the tool could not be made of a flexible material ? This would be definitely better than a chain. But if the segments are disconnected, then how are they controlled ? The solution seemed to be either trivial or impractical (which I suspect is often the case with most solutions allegedly obtained by TRIZ). I wrote Mr. Souchkov an e-mail asking for more explanation of the problem and proposed solution and whether any patent was obtained and received nothing from him in response.

All TRIZ practitioners behave like this. They never answer questions aimed at investigation of what a role TRIZ really played in obtaining a solution. Instead they prefer just empty declarations: