© Genrikh Altshuller
ON THE THEORY OF SOLVING INVENTIVE PROBLEMS

(Design methods and theories,Volume 24,Number 2, 1990. Translated by w.Gasparski.
This paper has been published for the first time in Polish language in the special Issue of the journal Prakseologia. 1977, 1/2, pp. 485-495.)

Inventiveness is the oldest occupation known to man. It was the invention of tools which began the process of humanization of our earliest ancestors.[…]

At the end of the 19th century Edison improved the trial and error method. Hundreds of people were employed in Edison’s laboratory and it was therefore possible to divide one technical problem into several tasks and check the many available variants for each task. In fact, Edison set up a complete scientific research institute - and it was this that was probably his greatest invention.

It is accepted that a thousand navvies will dig ditches which are different in quality those that would be dug by one navvy. But the method of digging itself does not change. The contemporary "invention Industry" is still working on Edison’s principle: the more difficult task, the more people are made use of to solve it.[…]

Today it is widely believed that great inventions are made not by individuals but by teams. But like all other aphorisms, this tells only some of the truth. There are various individuals and various teams - and the most important thing is the level or work organization. The "individual" excavator operator works much more effectively than the "team" of navvies. […]

A scientific approach to research into the work of invention begins with awareness of the fact that there are various degrees of inventiveness. Basically, inventive problems can be divided into five ranks. The first group are easy to solve after consideration of only a few variants. The solution of problems of the second rank requires consideration of several dozen variants. Problems of both the first and second rank can be solved by any engineers or technicians. Things get more complicated with problems of the third rank, which require hundreds of trial and errors. And it is at this stage that the fundamental drawbacks of the trial and error method become palpable: consideration of the variants becomes drawn out and a great deal now hangs on the individual characteristics of the inventor - his powers of endurance and ability to examine the variants doggedly, month after month, year after year. Problems of the fourth and fifth rank, which require consideration of thousands, or tens or hundreds of thousands of variants, are usually solved by the "relay” system: that is, one generation of inventors considers a part of the barren variants, thus narrowing the field of research, and then later generations of Inventors begin to review the variants etc.

The ranking of the problem is defined not only by differing number of variants which are necessary for solution. The variants are also differentiated in quality. At the lowest level, they concern the same field as the problem itself. […] At higher levels, the solutions are found to a wide extent outside the boundaries of the field in which the problem lies. […]At the lower levels, it is possible to consider the variants with the aid of knowledge from one’s specialization. But at higher levels, "further" information becomes increasingly necessary. […]

An effective tactical scheme for solving inventive problems can be worked out only on the basis of the objective laws of development of technological systems.[…] One of the objective laws of development of technological systems is that systems with an uncoordinated rhythm are replaced by more effective systems with a coordinated rhythm. […]

Work on creating a scientific theory of solving inventive problems was begun in 1946 by the author of this study. From the very beginning the research was directed towards discovery of the objective laws of the development of technological systems, and not towards examination of the secondary and outward symptoms of creativity: "revelation", "Inspiration" etc.

The laws discovered have been formulated in a system which algorithm of inventive problem solving - ARIZ (Altshuller, 1973).

ARIZ is a logical programme for solving inventive problems. Objective laws in the development of technological systems are assumed in the very structure of the programme, or occur in the shape of concrete operators of the programmе. Apart from this, ARIZ is equipped with specially concentrated information, obtained from analysis of thousands of patents.

When he uses ARIZ, the inventor is primarily analyzing the problem, step by step, according to precise rules. The analysis begins with revelation of the Ideal final result (IFR), towards which the technological system under examination should be directed. Next, factors hampering the achievement of the IFR are defined. This makes it possible to formulate the technological contradications which lie at the heart of the problem.

Technological contradiction is one of the fundamental concepts of the theory of Inventive problem solving. Technological contradiction lies in the fact that improvement of one element of a system (or one of its characteristics), through methods previously known, leads to a deterioration in other elements {or other characteristics). […]

Systems analysis is of particular significance in ARIZ. The passage from a given system to a super-system and sub-systems takes place according to particular rules. This makes It possible to arrive at indirect solutions, which are basically better than simple solutions.

With its operators and concentrated information resources, ARIZ is a model of artificial intellect, and more important, of a creative intellect. On the whole, ARIZ is a model of more-than-talented inventive thinking. It is "more-than-talented" because no real Inventor could be so versatile, nor could he conduct such logical and paradoxical thinking operations, nor have command of so much information on means of overcoming contradictions.

ARIZ gives shape to many statements on the theory of inventive problem solving, but the theory of course does not lead only to ARIZ. Recently, for example, field-matter analysis has been developing swiftly. This should be discussed separately.

Any geometrical figure can be reduced to a sum of triangles. This is why trigonometry is so significant in mathematics, as it dealt with the smallest geometrical figure, the triangle. Technological systems also include a minimal system, composed of two cooperating objects. A system of this kind, which contains two elements of matter and two energy fields is called a field-matter system. The concepts of matter and field are in this context understood in the broadest meaning: matter may mean machines, machine parts, natural objects; fields may be of any kind, including mechanical and heat fields etc. For example, if an Ice-breaker is sailing through tee, we are dealing with a field-matter system: the ice-breaker is one element of matter and the ice another; the mechanical actions are a force field.

Field-matter systems are governed by straightforward rules:

• a field-matter system is more effective than a non-field-matter system,
• a steered field-matter system 1s better than an unsteered one.
• in order for a field-matter system to be steered, it is sufficient for at least one element of the system to be steered (to be easily transposable), etc.[…]

ARIZ and field-matter analysis have made it possible to work out "standard solutions of Inventive problems". Each standard is a rigid rule for solving a particular class of inventive problem where the standard - and this is of fundamental importance -guarantees a high level of solution of the problems of its own class.

The standards - where it is possible to apply them - make creativity impossible, but they give a result which has been linked to date with creativity.

This paradox can be easily explained. The concept of creativity is constantly changing in meaning. For example, in the middle ages, tournaments were organized in which the rivals set each other to solve equations of the third rank. The elements of these equations were found by the method of trial and error and formed a typical example of creativity. The appearance of the Cardan formula ended this kind of creativity and every student was able to solve quadratic equations.

At the moment an analogous process is taking place in technology (of course, on a considerably greater scale). Over a millenium we have not found a way other than creativity for the development of technological systems. This is an extremely imperfect method, but we have had no other and creativity appeared to be an eternal and ultimate category.

The development of technological systems can be, and ought to be, carried our according to plan and according to principles - and not under the influence of inspiration. Technological creativity will inevitably be superseded by the science of the development of technological systems.

Moreover, we are deeply convinced that creativity as a method of innovation will also be superseded in other areas of human activity.

Thirty years ago (1946) the idea that it was possible to construct an algorithm for solving Inventive problems appeared to be a heresy and quite unreal. At the moment, classes on ARIZ research are held in dozens of towns in the USSR, and syllabuses are being worked out, text-books written and lecturers trained. Data on the practical application of ARIZ are being quickly accumulated. At the all-union "Heuristics" conference in 1974 it was emphasized that in colleges and Institutes alone in 1971-1974 hundreds of Inventions were made with the aid of ARIZ.

The contemporary scientific and technological revolution is usually associated with the development of atomic energy, the conquest of space etc. But the basis of the scientific and technological revolution is the development of new methods of producing knowledge. This is why we are convinced that the reconstruction of inventive methods will have enormous influence on the whole of the scientific and technological revolution.