3.1.6 ALGORITHM of the INVENTION

RELATED TO SOLUTION OF TECHNOLOGICAL OR SCIENTIFIC PROBLEMS

 

M.D.Kats, A.M.Davidenko

 

The problem of increasing the efficiency of inventive activity was becoming more and more aggravated in process of increasing the number of invention and fields of invention.

 

At first the inventions were developed by exhaustion method. For example, patent No. 223898 of Jan. 27,1880 was granted to Thomas Alva Edison for incandescent lamp. To develop the invention Edison has examined 6000 filament materials. As Edison noted in the patent, "I have carbonized and used cotton and linen thread, wood splints, papers coiled in various ways, also lamp black, plumbago, and carbon in various forms, mixed with tar and rolled out into wires of various lengths and diameters". The best results were obtained when applying a bamboo filaments produced from a case for Japanese palm fan. This titanic work takes about two years [1].

 

Later the methods of brainstorm, morphological analysis, synectics etc. began to be used for developing inventions. A long practical application of these methods has shown that any method is applicable for solving a simple problem and inefficient when solving a complex problem.

 

A theory of inventive problem solving (TRIZ) developed by G.S. Altshuller was the most essential achievement when solving the problem. G.S. Altshuller has shown by analyzing the large collection of the patent information that there are 40 most efficient receptions for eliminating about one and half thousand technical contradictions met most frequently. Later G.S. Altshuller has developed a special table contained the characteristics of technical systems subjected to improvement in each column, and the characteristics of technical systems, which become excessively worse thereby, in each row. Receptions, which are capable to eliminate a certain technical contradiction with the greatest probability, were pointed out within a cross-cell of a row and column.

 

At present there are actively done works on further improvement of TRIZ related mainly to development of programs capable to help an inventor to analyze a pertinent art and to search the required physical phenomena, typical or standard solutions of inventive problems within interactive environment.

 

As far back as in 1979, G.S. Altshuller wrote, “TRIZ doesn’t yet solve some classes of problems (for example, a production of new substances, or an discovery of optimal conditions for a production process). These problems will be also solved in due time” [2]. However, the prophecy was not fated to be realized. Such is not the case that G.S. Altshuller himself and many his followers have given insufficient attention to research of formalized approaches to solution of inventive problems. The point is that there is a principal distinction between a problem solved by discovering the technical contradictions and searching one of the known ways of eliminating the contradictions, and a problem solved only if a certain knowledge is available. The later (technological) problems include problems of a new substance production and optimization of a new production process or a current production process.

 

Besides, there is a class of problems that could be named scientific problems. Results obtained when correct solution of such problems possess an essential scientific novelty, a high practical utility and ability to be patented. Such problems include, for example, problems of developing:

- differential diagnostics of almost indistinguishable diseases;

- preliminary prognostication of outcomes and after-effects for diseases;

- selection of an optimal strategy for a treatment of a certain disease, which would take into consideration the individuality of a patient;

- identification of microorganisms (development of a special complex test for every microorganism of the same family (strain);

- mathematical synthesis of the chemical formula for a new compound (for example, a bioactivity substance or a synthetic dyestuff) that would have the specified properties;

- psychological testing (development of a specific complex test for every gradation of an index tested).

 

There is required a large volume of specific knowledge for solution of the technological and scientific problems at inventive level. For example:

- a knowledge of dependence of an output index (or a set of output indices) from input parameters of a production process is required when optimizing a current or newly developed production process in metallurgical, chemical, biotechnological or petroleum refining industries;

- a knowledge of dependence of an output index (or a set of output indices) from a mixture formulation and conditions for conversion of the mixture into material when obtaining a new composite material (for example, a metal, alloy, rubber, plastic, catalyst, commercial form of a dye or pigment, or mix form of a medicinal preparation);

- a knowledge of a combination of symptoms (a symptom-complex, a differential syndrom) to characterize each disease aimed to be differentiated when diagnosing the almost indistinguishable diseases;

- a knowledge of dependence of treatment results from a disease symptoms, a treatment procedure and the individuality of a patient when selecting the optimal strategy for treatment of a certain disease at a certain patient.

- a knowledge of dependence of utilization qualities from a mutual arrangement of structural elements within a molecule when directed syntheses of chemical compounds of a certain class with the specified combination of utilization qualities.

 

Most of the technological and scientific problems belong by their information characteristics to the class of “large-scale” systems. A behavior of such system is determined with a great amount of input parameters (more than 10) and output indices (more than 1), as well with essential dependence of the output indices upon interaction of the input parameters.

 

A great deficiency of a priori information (a deficiency of expert knowledge in the appropriate subject field) about systematic relationships between input parameters and output indices of the “large-scale” system as investigated is the principal difficulty when solution of such inventive problems. The deficiency is explained first at all with a psycho-physiological limitation of a man ability to recognize abstract images (no expert can estimate dependence of an output index upon more than 2 input parameters interacting between each other).

 

“Absence of ability to simulate a phenomenon (and to solve a problem), which is multifactorial by its nature, is the main limiting factor for human cognitive and creative abilities” [3].

 

“Knowledge is a set of models for world around us” [4].

 

Therefore, there is the only way to cognize a “large-scale” system, namely, to develop a formal mathematical model for the system. However, an inventor meets practically insuperable methodical and computing difficulties in this way too. The point is that: “Developing a mathematical model is not formalized. The process includes always some assumptions. Every calculation and comparison with stored information is carried out on the basis of the assumptions [5].

 

The assumptions used when simulating a “large-scale” system are, as a rule, unconstructive. The assumptions consist mainly in an inadmissible simplification of a real problem (because of absence of actual knowledge). There are made, for example, the following assumptions: “The dependence is linear and additive. A hypothesis of compactness is observed for input and output variables. The system is stationary. The distribution of variables is normal”. So “the development of a mathematical model is still a matter of skill” [6].

 

A “simulator” uses all his skill to overcome many methodical problems of “large-scale” system identification unsolved up to now. However, no skill and intuition of a “simulator” allows principally to compensate the absence of knowledge about relationships determining a behavior of a “large-scale” system as investigated, or the absence of the appropriate formal methods for developing a mathematical model of a system.

 

So, any known method of mathematical simulation (regressive analysis or any modification of this method) is incapable, for example, to solve the following problems:

- structural identification, or representation of a general appearance of model;

At present the structure of a model is represented by an expert, who is principally incapable to imagine what is the “true” structure of the system due to psycho-physiological limitation mentioned above.

- parametrical identification, or estimation of coefficients for structural elements of the model on the basis of experimental data;

Application of the known methods of parametrical identification, namely, the least squares method and any modification of this method, is correct only by imposing a number of restrictions, namely, the input parameters are measured without errors or with negligible errors, all the errors are entered into an output index, there is only one output index, the output index is measured by a continuous scale and has the normal distribution etc., which are never fulfilled for real data.

- convolution of output indices vector into a generalized criterion measured by a continuous scale.

No correct formal method of the convolution is known. Nothing else is left, but to use a subjective imagine of an expert.

 

A fitness of the known simulation methods for real technological problems is fully expressed with the known aphorism: “If a problem includes less than 3 variables, then there is no problem, but if a problem includes more than 8 variables, then such problem is insoluble”, namely, insoluble due to essential deficiency of a priori information about the model structure and absence of a correct formal method for the identification of a “large-scale” system [7]. When analyzing the applicability of the known method for investigation of the “large-scale” systems an unexpected conclusion was drawn: “Every “large-scale” system created by man is nonoptimal and could be essentially improved”. This conclusion leads to paradoxical consequence: “Every patent granted for a solution of technological or scientific problem can be circumvented in principle by another, less assailable patent”.

 

However, information about a “large-scale” system required for development of the less assailable invention can be obtained only after the main problem of an artificial intelligence would have been solve, namely, a discovery of new systematic knowledge, which was known to no expert, about relationships between input and output variables on the basis of experimental data with help of formal procedures (i.e. without participation of an expert).

 

The problem can be solved when applying a principally new method of mathematical simulation (a mosaic portrait method) [9, 10] developed as a part of Intelligent Technology of Complex System Study (ITCOSS) [8]. The method allows developing an adequate mathematical model of the system as investigated on the basis of the initial experimental data and realizing an algorithm of the invention for that technological problem.

 

The initial data can be represented as a table of experimental material, wherein each row contains the values of input parameters and output indices obtained in the same realization of an object as investigated.

 

As a matter of fact, the mosaic portrait method consists in the following operations:

 

- Going from measurement of input parameters and output indices of an object as investigated by continuous scales to measurement by discrete scales.

 

Going to measurement of input parameters (X_{i) by discrete scales is carried out with the help of the following procedure formalized completely. A value variation range for every input parameter is divided into 3 sub-ranges such that every sub-range contains the approximately equal number of the table rows of experimental data (or about equal number, if the amount of rows is not divisible by 3). Each sub-range is coded so that the sub-range number grows as the value of the corresponding input parameter is increasing.}

 

Going to measurement of output indices by discrete scales is carried out with the help of the following procedure formalized completely. An output index obtains binary code (Y = [Y] =1), if the index satisfies to restriction [Y], or binary code (Y ≠ [Y] =0), if the index doesn’t satisfy to restriction [Y]. If there are a few output indices, the generalized criterion is formed as follow. Y = [Y] =1, if all output indices (Yj) satisfy to their restrictions [Y], or Y ≠ [Y] =0, if even one of output indices does not satisfy to the corresponding restriction.

 

- Realizing a procedure formalized completely for searching such combinations of codes for input parameters, which meet only in rows of initial table with [Y]=1 and don’t meet in rows, wherein such dependence doesn’t satisfy.

 

Searching such combinations of codes that are meet in rows with Y=0 is carried out by the same way.

 

In that case, if an output index is measured by a discrete scale (for example, by name of a disease when differential diagnostics), the mosaic model is developed separately for each disease (the disease simulated is coded by 1, and all other diseases are coded by 0).

 

Each combination of codes for subsets of the input parameters included into the mosaic model is interpreted as a formal hypothesis that is consistent for on the given experimental data and describes systematic relationships of an object as investigated (a process, system, phenomenon). Each of these hypotheses that includes more than 2 input parameters is a new nontrivial hypothesis, which was known earlier to no expert. Substantial interpreting the hypotheses obtained by terms of an appropriate subject field will allow experts to “break through” into a field of systematic relationships of the “large-scale” system as investigated, which were hidden till now for experts, and to obtain a new knowledge required for developing an investigation.

 

Examples of substantial interpretation of formal hypotheses obtained with the mosaic model are given below. Depending on a problem as solved the formal hypothesis can represent:

- a specific symptom-complex (a differential syndrome) for one of diseases when differential diagnostics;

- a combination of structural elements observed only at compounds with the specified combination of utilization qualities when forecasting utilization qualities of a new chemical compound;

- a test-complex specified for a certain bacterium when identifying microorganisms etc.

 

It should be noted that the discrete mathematics contains no general method for solving a problem of searching a “useful” combination of any elements but exhaustion method. In case of a “large-scale” system the exhaustion results in practically infinite expenses of time for realization. The development of the mosaic model method allowed to solve the problem with a polynomial dependence of working hours upon the dimensionality of a problem. Therefore, it is found a method available for realization when mathematical description of a “large-scale” system with practically any dimensionality of input and output variable vectors.

 

The mosaic model contains enough large volume of systematic information required for correct solution of problems of diagnostics and forecast of behavior for a “large-scale” system as investigated.

 

An optimization of a system as investigated is carried out by logical programming method [11]. The logical programming method (LPM) is based on the known axioms of logical algebra about a truth of statements:

- a compound statement is truth in case, if and only if the statement contains true simple statements and contains no false simple statement;

- a compound statement is false in case, if the statement contains at least one false simple statement.

 

If a statement of the mosaic model related to “good” class of an output index (Y=1) is considered to be a true simple statement, and a statement related to “bad” class of an output index (Y=0) is considered to be a false simple statement, then a compound statement of n range (where n is the amount of input parameters of a system as investigated), which contains the statements of “good” class and contained no statement of “bad” class, will be a “true” compound statement.

Depending on a problem as solved the substantial interpretation of a true compound statement obtained with the help of the formal procedure can represent, for example:

- an area of input parameters space determined by a corresponding subset for each parameter which mapping into a space of output indices will show an area of output indices space, wherein a value of any output index corresponds the specified value, when optimizing a production process;

- an area of input parameters space determined by a corresponding subset for a content of each component and for a value of each parameter for conversion of the mixture into material with the specified combination of utilization qualities when developing a new composite material;

- a combination of treatment procedures allowed to ensure a maximum efficiency of cure when optimizing a treatment of a certain disease;

- chemical formulas of compounds capable to have a specified combination of physical, chemical, biological or other properties when formal synthesis of chemical formulas for compounds of a certain class.

 

A PROBLEM FORMULATION, WHEN STUDYING AND OPTIMIZING A “LARGE-SCALE” SYSTEM,

AND METHODS OF THE PROBLEM SOLUTION

 

According to existing paradigm it’s necessary to be an expert in an appropriate subject field in order to formulate a problem, and to know ways of the problem solution by known mathematical methods. There are a practically infinite number of problem formulations in each subject field due to that the knowledge of experts about systematic relationships of “large-scale” systems is limited, and a general correct method of solution for a complex technological or scientific problem is unknown. It is easy to make sure of the fact by visiting a scientific conference related to medicine, metallurgy, or chemistry, wherein almost every lecture contains a new problem formulation.

 

The mosaic portrait method allows solving a problem of a “large-scale” system study by the following general formulation: “It’s necessary to develop a mathematical model enabled to describe dependence of an output index (a combination of output indices) upon input parameters for an object as investigated (a process, system, phenomenon) on the basis of available initial data about the input parameter and output indices values”. The model can be used to forecast a behavior and to diagnose the “large-scale” system of any physical nature. The problem of “large-scale” system optimization is solved by the logic programming method or situational control method developed by modifying the logic programming method [12].

Therefore, ITCOSS enables to formulate and solve a minimal amount of important problems within every subject field. Thereby the results obtained when solving a problem serve as basis for developing an application for an invention. Below you find the examples of the general formulations for technological and scientific problems.

PROBLEM OF STUDY AND OPTIMIZATION FOR CURRENT PRODUCTION PROCESSES

 

The following problem formulation is suitable for production processes of petroleum refinery, or chemical, metallurgical, biotechnological and other industries: “It’s necessary to find conditions, under which a production process provides an essential increase of efficiency in relation to a combination of specified indices”. The problem can be formulated more strictly as follows: “It’s necessary to find conditions, under which a production process enables to take a product with specified quality and minimum cost”. Since the expenses of row materials and power are the main components of overall product cost, and row materials and power consumption calculated for unit of the product are determined fully with a yield of the product calculated for the row material feed, the problem can be formulated yet as follows: “It’s necessary to find conditions, under which a production process enables to take a product with maximum yield and specified quality”. Such formulation allows also to solve an ecological problem due decrease of waste amount. Material balance of production “row materials = product + waste” shows that an increase of product yield leads to the decrease of the waste amount.

 

In those rare cases, when the demand outruns the production, it is more advisable to formulate the problem as follows: “It’s necessary to find conditions, under which a production process enables to take a product with maximum profit and specified quality”.

 

The mosaic portrait method (MPM) helps to develop a mathematical model of a process as investigated, and the logical programming method (LPM) allows to determine conditions, under which the production process ensures the solution of a problem formulated, on the basis of the mathematical model of the process. If there is a patent suitable as a prototype, it’s necessary only to compare the conditions obtained with the conditions of the patent in order to find an essential distinction and to prepare a patent application. The invention claim will have the following appearance: “A process for manufacturing product A with (improved output indices), said process being carried out under (common parameters), wherein (distinctive parameters)”.

 

For example, when optimizing a process for manufacturing sulfamic acid, there were found the optimal conditions that are essentially distinct from the conditions comprised in a patent used as a prototype. The application of these production conditions enabled to improve essentially the following output indices regulated:

- a mass content of the base substance was increased from 83.0 % up to 95.62 % (raised by 15.2 %);

- a content of sulphuric acid was decreased from 6.0 % down to 2.58 % (reduced by 57.0 %);

- a productive capacity was increased from 3000 t per year up to 3000 t per year (raised by 10 %).

As a result there was obtained SU Inventor’s Certification No. 1060565 for “A method for producing sulfamic acid”.

 

A PROBLEM FORMULATION WHEN DEVELOPING AND OPTIMIZING A NEW PRODUCTION PROCESS,

OR A NEW COMPOSITE MATERIAL

 

When developing a new production process or a new composite material, the initial information required for the development can be obtained by the way of active experiment in keeping with special schedule that enables to develop a corresponding mathematical model to describe a dependence of an output index on every input parameter of a system as investigated with the minimum number of experiments [13-15].

 

Since the optimization of an object as investigated is carried out by discrete scales, the recommendations with respect to values of input parameters and component proportions are given as appropriate optimal sub-ranges for the parameter values.

 

When developing a new production process, the conditions are discovered, under which the production process allows manufacturing the product with the specified output index (a combination of output indices). The development results in a patent application.

 

If there is a patent suitable as a prototype, the invention claim will look like as follows: “A process for manufacturing product A with (improved output indices), said process being carried out under (common parameters), wherein (improved sub-ranges for parameter values)”.

 

When developing a new composite material, the optimal component proportions and condition, under which the production process allows converting the component mixture into material with the specified combination of utilization qualities, are discovered. The development results in a patent application.

 

If there is no patent suitable as a prototype, the invention claim will look like as follows: “A process for manufacturing material A with (specified output indices and their values), said process being carried out by converting a mixture of components into material, wherein (optimal sub-ranges for component proportions) and (optimal sub-ranges for conditions of conversion)”.

 

If there is a patent suitable as a prototype, the invention claim will look like as follows: “A process for manufacturing material A with (improved output indices), said process being carried out under (common parameters), wherein (improved component proportions of the mixture and improved sub-ranges of parameter values)”.

 

PROBLEM OF SYNTHESIS

FOR A NEW CHEMICAL COMPOUND WITH SPECIFIED UTILIZATION QUALITIES

 

A problem of directed synthesis for a chemical compound with specified properties is one of the most difficult scientific problems having a great practical importance. It is known, for example, that it’s necessary to test 800 -1000

New compounds synthesized for discovering a dyestuff suitable for insertion into assortment, or to screen about 100000 compounds for developing a new medicine [16].

 

One of last attempts to use ARIZ for solving the problem was a development of information system that allows “to drop out some compounds capable to become a medicine and to select only those compounds, which appear to a qualified chemist to be capable to meet with success after their synthesis. First at all there are left those compounds, which have a structure born a strong resemblance to the structure of the medicines published. Here is a stage, wherein a chemist begins to play an active role. Computers and algorithms help to narrow an immense virtual world up to that space, which can be comprehended with a chemist in order to take a qualified decision” [16].

 

Now there is a great experience in developing various man-machine complexes, expert systems etc. As a rule, such systems are not effective when applying them for a “large-scale system” (in practice every class of chemical compounds belongs to “large-scale systems” by information characteristics, namely, the number of substituents in various positions of a molecule, which interactions determine the physico-chemical, biological and other properties, or utilization qualities of a certain compound). But … “we learn from history that man can never learn anything from history (George Bernard Shaw).

 

The problem is formulated with the help of ITCOSS as follows: “It’s necessary to develop a systematic model for dependence a combination of the specified physico-chemical, biological and other properties on the chemical structure of a certain class of chemical compounds, and to carry out the directed synthesis of chemical formulas for new compounds with the combination of the specified physico-chemical, biological and other properties on the basis of the model obtained”.

 

An algorithm for solving the problem comprises:

- developing a mathematical model for the dependence of a property on the structure with the help of MPM;

- synthesizing chemical formulas for new compounds capable to have a specified combination of utilization qualities on the basis of the model with the help of LPM;

- executing the application for “new chemical compound with the specified properties…” after the chemical synthesis and appropriate test of the new compounds.

 

For the first time there is developed a systematic mathematical model for dependence between a structure of a certain class of chemical compounds and their properties, which is suitable for the directed synthesis of new chemical products [17]. Chemical formulas of 273 disperse monoazodyes included into home assortment and assortments of known foreign firms or synthesized in Institute of Chemical Technology (Rubezhnoye, Ukraine) were used as initial data when carrying out a research. All the dyestuffs were tested for light fastness, resistance to sublimation and dye-receptivity when 1% dyeing. When developing the model, an appropriate serial number (a code) was given to each substituent located at a certain position in order to describe a structure of a dyestuff as a corresponding combination of the codes.

 

The model obtained consisted of 2 subsets for combinations of dye structural elements (the one subset comprises the combinations characterizing the dyes, which meet the specified requirements for all the output indices, the other subset comprises the combinations characterizing the dyes, which don’t meet even one of the specified requirements).

 

The model allowed to solve the following problems:

 

- Forecasting properties of a new dyestuff, which is not yet synthesized, on the basis of the chemical formulas.

 

Chemists have proposed formulas of 73 new compounds suitable for synthesis. Their properties were predicted on the basis of the mathematical model. There was shown that only 10 formulas could correspond to dyestuffs with acceptable utilization qualities.

 

After the forecast all 73 dyestuffs were synthesized, and experimental test of their properties was carried out. It was found that 9 of 10 dyestuffs, which were expected to be “good”, meet to the specified requirements, and tenth one was inferior in light fastness and resistance to sublimation by 0.5 point. (By the way, an accuracy of estimation for these indices is also 0.5 point in accordance with standard GOST 97-33-61.) There wasn’t enough information in the mathematical model for the appropriate forecast of 12 dyestuffs (16.5 %), results of the forecast and experimental test have completely coincided for 58 dyestuffs (79.5%), and the forecast has appeared erroneous only for 3 dyestuffs.

 

- Formal synthesis of the chemical formulas for dyestuffs, capable to have a combination of the specified properties.

 

16 chemical formulas of yet unknown monoazodyes were synthesized with the help of formal procedures on the basis of the model “structure - combination of properties”. It was found after the chemical synthesis and coloristic tests that 14 dyestuffs (87.5%) meets completely to the specified restrictions (light fastness ≥ 6 points, resistance to sublimation ≥ 6 points, dye-receptivity when 1% dyeing meets to the requirement). The dye-receptivity of 2 dyestuffs was inferior to the specified value by 0.5 point, which doesn’t exceed the error of estimation for the index in accordance with standard GOST 97-33-61.

 

PROBLEM OF DIAGNOSTICS, PROGNOSTICATION AND OPTIMIZATION OF TREATMENT IN MEDICINE

 

The problem of differential diagnostics for almost indistinguishable diseases is formulated as follows: “It’s necessary to discover the symptom complex (a differential syndrome) for each of diseases as differentiated on the basis of experimental data table, wherein each row contains information about the disease signs and verified diagnosis for a certain patient”.

 

A systematic mathematical model is developed with the help of MPM to discover a symptom complex (a differential syndrome) for each of diseases as differentiated. The developing symptom complexes allow to carry out a reliable recognition of these diseases. The model obtained serves as a basis for complication of patent application: “In a method for differential diagnostics within a group of almost indistinguishable diseases, for example, diseases À, Â, Ñ and D, … wherein the improvement comprises (symptom complexes of the diseases À, Â, Ñ and D), whereby the accuracy of differential diagnostics is increased”.

 

A simple and non-invasive method of differential diagnostics for “gastric ulcer - gastric cancer” based on application of symptom complexes for each of diseases as differentiated was developed in co-operation with Military Medical Academy (St. Petersburg). The method was practically applied, for example, in gastro-enterological compartment of Municipal Clinical Hospital No. 20 (St.-Petersburg). It was found that the diagnosis obtained with application of the proposed method has coincided with the verified clinical diagnosis (gastroscopy with aiming biopsy, operation) in 96.4 % of all cases.

 

A problem of prognosticating an after-effect (an outcome) for a disease can be formulated as follows: “It’s necessary to develop a method for the preliminary prognostication of after-effects (outcomes) for a certain disease (A) on the basis of the disease signs collected on initial stage of the disease”.

 

A method of prognosticating the after-effects at acute stage of the myocardial infarction (the cardiogenic shock, rupture of the myocardium, insufficiency of blood circulation, ventricular fibrillation, or no after-effect) observed on 3^{rd to 20th day of hospitalization is developed in co-operation with Military Medical Academy (St. Petersburg). The initial information about a patient was collected immediately after hospitalization, and the diagnosis of the after-effects was recorded after the realization of the method. A mosaic model developed on the basis of the initial data contains 5 subsets of symptom complexes, which are characteristic for every possible after-effect.}

 

When realizing the prognostication, the information about values of input parameters for a new patient was collected immediately after hospitalization and coded, as it is customary for developing a mosaic model. Then it is established which of symptom complexes is observed at a patient. If only one symptom complex is observed at the patient, then a preventive therapy is selected according to the prognostication. But if 2 or more symptom complexes are observed at a patient, it can be taken 2 solutions:

- to select a preventive therapy for a more probable after-effect, for which there were observed more symptom complexes;

- to exclude those medicines and treatments, which are contra-indicated for a less probable after-effect, from a preventive therapy for a more probable after-effect.

 

The mosaic model for prognosticating the disease after-effects serves as a basis for complication of patent application: “In a method for preliminary prognosticating after-effects for myocardial infarction (the cardiogenic shock, rupture of the myocardium, insufficiency of blood circulation, ventricular fibrillation, or no after-effect)… wherein the improvement comprises (specific symptom complexes for every after-effect of myocardial infarction), whereby the accuracy of prognostication is increased”.

 

Here are given, for example, 3 symptom complexes for each of the after-effects:

 

- SYMPTOM - COMPLEXES FOR THE CARDIOGENIC SHOCK:

1. A patient is a male; the anamnesis shows an insufficiency of blood circulation; a hypotension is absent (arterial pressure is less than 90 mm Hg).

2. There was never a hypertonic disease; a hypotension is present; there is no gallop rhythm.

3. There is a hypotension, no myocardial friction rub; an electrocardiogram shows no back infarction.

 

- SYMPTOM - COMPLEXES FOR RUPTURE OF THE MYOCARDIUM

1. There is a pericardial friction rub, no wet wheeze, and no group extrasystole.

2. There was never an insufficient blood circulation; there is a hypertension (arterial pressure is more than 160/90 mm Hg), no paroxysm, and an aneurysm of heart.

3. Radiation of pain is atypical; there is no wet wheeze, no bradycardia (heartbeat at a rate less than 50 beats per minute), and an atrioventricular block.

 

- SYMPTOM - COMPLEXES FOR INSUFFICIENCY OF BLOOD CIRCULATION

1. The myocardial infarction is not primary; the anamnesis shows a disease of arterial blood vessels; there is a gallop rhythm; an electric pulse therapy wasn’t applied.

2. The anamnesis shows a disease of arterial blood vessels and insufficiency of blood circulation; an increase of liver size is observed; the electrocardiography shows the infarction of all walls.

3. The anamnesis shows a hypertonic disease; there is a wet wheeze, an increase of liver size; the electrocardiography shows a front infarction.

 

- SYMPTOM - COMPLEXES FOR VENTRICULAR FIBRILLATION

1. There is no cardiac asthma, no pulmonary edema, no hypertension, and a polytopical extrasystole.

2. There is a paleness of skin covers, a hypertension, and a group extrasystole.

3. There is a clear consciousness, a group extrasystole; the electrocardiography shows no front infarction.

 

- SYMPTOM - COMPLEXES FOR NO AFTER-EFFECT

1. Pain is damped down with medicines; there is no tachycardia, no ventricular extrasystole, no atrioventricular block.

2. There is no paleness, no cyanosis, a clear consciousness, no tachycardia, and no ventricular extrasystolia.

3. Pain is damped down with medicines; there is no cardiac asthma, no pulmonary edema; cardiac glycosides and electric pulse therapy were not applied.

 

Practical application of the method for prognosticating after-effects of myocardial infarction developed with the help of ITCOSS has enabled to reduce a lethality provoked with the macrofocal myocardial infarction by 36.8%, and the lethality provoked with microfocal myocardial infarction by 45,1%, for example, in cardiac department of Municipal Clinical Hospital No. 23 (Moscow).

Problems of early diagnostics (sometimes even at latent stage of a disease) for chronic diseases (a cancer disease, chronic renal insufficiency etc.), which are dangerous for life, are solved similarly.

 

Problem of selecting an optimal strategy for treatment of a certain disease is formulated as follows: “It’s necessary to synthesize the optimal strategy of treatment for a certain patient with the help of situational programming method on the basis of a disease mosaic model capable to fix a dependence of efficiency for treatment of a certain disease (A) on parameters characterizing: a disease and a patient individuality”.

 

MPM helps to develop a systematic mathematical model to fix a dependence of treatment efficiency of a certain disease on parameters characterizing: a patient individuality (a color of eyes, a group of blood, Rh factor etc.), a disease (anamnesis data, clinical observation data, data of laboratory and instrumental tests etc.), and parameters of a treatment used.

 

There is set a situation, namely, values of parameters to fix the patient individuality and disease, for every patient. The most efficient treatment is synthesized for the situation with the help of a situational programming method.

 

The situational programming method (SPM) is one of modifications of the logical programming method. SPM consists in discovering a situation, namely, a combination of codes for noncontrollable parameters characterizing the disease signs and individuality of a certain patient, and synthesizing the optimal treatment, namely, discovering the most efficient cure procedures (for that situation). An application package is developed to realize the discovery of the most efficient treatment of a certain disease for every situation (every patient).

 

The data obtained serves as a basis for complication of patent application: “In a method for optimizing a treatment of disease N with taking into account the individual features of a patient … wherein the improvement comprises synthesizing recommendations on the optimal treatment on the basis of disease mosaic model for every situation defined with parameters characterizing the disease and individuality of a patient”.

 

As a matter of fact, a comparison of the optimal area for values of input parameters obtained when optimizing a “large-scale system” by applying ITCOSS with the appropriate area for values of input parameters of the patent suitable as the prototype gives that essential distinction, which will then enter into invention claim: “ … wherein the improvement comprises …”.

 

Generally speaking, a comparison of a systematic model made of specific test complexes (the symptom complexes) for every state, which was obtained with the help of mosaic portrait method when prognosticating and diagnosing a “large-scale system” by applying ITCOSS, with the appropriate means of prognostication (diagnosis) of the patent suitable as the prototype gives that essential distinction, which will then enter into invention claim: “ … wherein the improvement comprises applying the specific test complexes (symptom complexes) specified for every state as prognosticated (diagnosed)”. The lists of test complexes (symptom complexes) for every state prognosticated (diagnosted) are inserted.

 

A present there is a significant discrepancy between working hours spent by scientists for collecting the initial data and importance of results obtained by analysis of that data. “Science world is swarming with assiduous scientists, which are capable to think quite logically and to work honestly, and nevertheless incapable to suggest new ideas. Unfortunately, new ideas are not prerogative of those who were engaged in research and development for a long time” [18]. The application of ITCOSS will enable to remove this contradiction out of scientific work. ITCOSS will allow a scientist not only to formulate correctly a problem as investigated, and to carry out the investigation expediently, but also to obtain new, yet unknown systematic relationships for an object as investigated.

 

It is difficult to imagine, how much new knowledge remained to be undiscovered out of results of previous researches! ITCOSS helps us to discover that knowledge without additional expenses for new experimental work.

 

At the beginning of 21 century the following technologies are considered to be the most perspective ones: new materials, semiconductors, improved semiconductors, digital processing of the images, storage of information on high-density media, high-speed computers, optoelectronics, artificial intelligence systems, super dataways, sensor technologies, biotechnologies, systems of medical diagnostics.

 

Application of ITCOSS formalized procedures for development of these technologies will allow to find an optimal solution for every of the problems and to protect the solution with appropriate patents.

 

CONCLUSIONS

 

1. The analysis of known methods of study, development and optimization for the "large-scale systems" resulted in unexpected conclusion:

- every "large-scale system" developed by man is nonoptimal and can be essentially improved due to the human psycho-physiological narrow-mindedness when recognition of abstract images (no expert can estimate a dependence of an output index on mutual influence of more than 2 input parameters), and the absence of correct mathematical methods for simulation of multifactorial systems (“If a problem includes more than 8 variables, then the problem is insoluble”);

2. Such conclusion leads to a much more paradoxical consequences that are nevertheless well-grounded logically:

- every patent granted for a problem solution of identification, forecast, diagnostics or optimization of a "large-scale system" can be circumvented by another, less assailable patent.

- a development of a less assailable patent can be carried out with the help of a formal procedure comprising part of Intelligent technology of complex system study (ITCOSS);

- ITCOSS can serve as a formalized algorithm of the invention in the field of identification, forecast, diagnostics and optimization of a “large-scale system".

3. ITCOSS differs from ARIZ (Algorithm of Inventive Problem Solving developed on the basis of TRIZ) not only with an area of application (ARIZ is intended for solving inventive problems in engineering, but ITCOSS is intended for solving inventive problems in technology and science), but also with a level of formalization when formulating and solving an inventive problem. ITCOSS allows to formulate a problem correctly from the very beginning of research and to solve the problem with the help of a formalized procedure.

 

LITERATURE

 

1. Khurgin Ya.I. Da, net, mozhet byt’…(Yes, no, may be…). - Moscow: Nauka, 1977. - p. 208.

2. Altshuller G.S. Tvorchestvo kak tochnaya nauka (A creative work as a exact science). - Moscow: Sovjetskoye Radio, 1979. - p.175.

3. Lem S. Planeta Zemlya. Vek 21. (Earth. 21 Century.) - Komsomolskaya Pravda, Aug. 11, 1992.

4. Rasstrigin L.A. S komputerom nayedine (Tête-à-tête with a computer). - Moscow: Radio i svyaz’, 1990.

5. Chisla I mysl’. - In: Dorodnitsyn A.A. Matematika I opisatel’niye nauki (Mathematics and descriptive sciences). - Moscow: Znaniye, 1982, No. 5. - pp. 6 -15.

6. Dorodnitsyn A.A. Problema matematicheskogo modelirovaniya v opisatel’nykh naukakh (A problem of mathematical simulation in descriptive sciences). - Kibernetika, 1983, No.4, pp. 6-10.

7. Enon. Osnovniye zakonomernosti nauchnoy raboty (Basic relationships of a scientific work). In: Phiziki prodolzhayut shutit’ (Physicists continue to make fun). - Moscow: Mir, 1968.

8. Kats M.D. Optimal’nyi rezhym (Optimal conditions of operation). - MM Den’gi i tekhnologii, 2002, No. 10.

9. Kats M.D., Reznichenko V.V. New method of mathematical simulation for chemical production processes. - Avtomatizatsiya khimicheskikh proizvodstv, 1975, No. 2.

10. Kats M.D., Ardashev V.N., Yakovlev G.M. et al. Metod mozaichnogo portreta v prognozirovanii oslozhneniy infarkta myocarda (Application of mosaic portrait method for prognosticating the after-effects of the myocardial infarction). - Kardiologiya, 1981, No. 6.

11. Kats M.D. Metod optimizatsii khimiko-technologicheskikh processov po informatsii, poluchemoy v rezhyme normal’noy expluatatsii (Method of optimizing chemical engineering processes on the basis of data obtained under normal operating conditions). - Avtomatizatsiya khimicheskikh proizvodstv, 1980, No. 2, pp. 5-8.

12. Kats M.D. O postroenii situatsyonnoi modeli i situatsyonnom upravlenii khimiko-technologicheskimi processami (About development of a situational model, and a situational control of a chemical production process). - Avtomatizatsiya khimicheskikh proizvodstv, 1980, No. 5.

13. Kats M.D., Davidenko A.M. Metodologiya razrabotki novykh kompozitsionnykh materialov, obladayushchykh zadannym complexom physiko-chimicheskikh (potrebitel’skikh) svoistv (A methodology for developing a new composite material with the specified combination of physical-and-chemical properties (utilization qualities)). - In: Vestnik KhGPU, 2000, vol.104

14. Kats M.D., Nevmyvako V.N., Slatutskaya L.M. at all. Metodologiya issledovaniya pri razrabotke novykh khimiko-technologicheskikh processov (A methodology of investigation when developing a new chemical production process). - Khimicheskaya tekhnologiya, 1984, No.4.

15. Kats M., Kestelman V., Davidenko A. Methodology of developing a new composite material with specified combination of physical-and-chemical properties or utilization qualities. - SCIENTIFIC ISRAEL - TECHNOLOGICAL ADVANTAGES. Contents VOL. 5. 2003 No. 2 " Materials Engineering“.

16. Wife R., Hehenkampf Ju. SORT&gen - novaya informatsyonnaya technologiya otkrytiya lekarstv budushchego (SORT&gen - a new information technology for discovering the future medicines). - Rocciyskiy chimicheskiy zhurnal, 1998, N0.6.

17. Kats M.D., Mostoslavskaya E.I. Zavisimost’ mezhdu stroeniem dispersnykh momoazokrasiteley i ikh potrebitel’skimi svoystvami na lavsane (A dependence between the structure of disperse momoazodyes and their utilization qualities on Lavsan). - Zhurnal prikladnoy chimii, 1983, No.9.

18. Edward de Bono. Rozhdeniye novoy idei (Birth of a new idea). - Moscow: Progress, 1976. - p.141.