Business systems

At any time, a business system can be described by some model, which will be its temporary slice. The analysis of this slice, which we will call the structural-functional analysis, should answer the question about the static functioning of the BS, as if its dynamic qualitative changes did not occur.

In addition, this analysis (for example, several structural and functional sections) can predict the nature of BS development and, with timely and adequate measures, mitigate the process of reinstalling the business system and product diversification. The task involves a common solution and applications that describe the most typical business systems and invariant elements for a given level of entrepreneurship development.

The basis for modeling business systems is based on the metaphor of the interconnectedness (recursiveness) of management, organization and execution and can be illustrated by the following scheme:

Simulation models

Unfortunately, the historical development of simulation modeling has developed a stable immunity among potential customers of such systems. This is caused, first of all, by the fact that no model is similar to its phenomenal prototype. However, we believe that simulation modeling is useful at least in order to give a reason to look at the Object with a new look from a different point of view.

Associative solver

A very interesting example of using the approach to generate statements based on the source text, which we will discuss in more detail.

The associative solver is designed to solve problems of constructing logical, formal models, verification and examination of models, as well as obtaining conclusions, results and model solutions. It can be one of the main modules of active information systems in terms of their algorithmic processing. An associative solver is a system with two source spaces: a field of facts (or a model) and a field of inference rules. It differs significantly from the existing expert systems and the so-called syllogistors in terms of flexibility, versatility and ability for self-modification.

Based on the initial fields, the associative solver forms a statement that is complementary to them. This is essentially all his work.

It may happen that one of the elements of the field of facts will be associated with others more than once (ring, cycle), then we have a wonderful opportunity to check the statements without going beyond the initial field of statements. If the relations between objects obtained by different rounds of the ring have no intersection (which is the solution of the relation), then in the collection of statements there is at least one that contradicts the other. In this case, we sometimes can even find out which one (without additional conditions).

Detection of incorrectly set conditions long before the model is solved will give an inevitable result and even an economic effect. (Especially for the models of the Companies Development Programs, which affect the existence of their own existence.)

The ability to test the model for correctness of the statement can provide interesting applications for a psychological lie detector. When we go through the model several times, changing the direction of the relationship and using synonyms to initiate objects.

It can be assumed that the development of this direction is a good alternative (addition) for building systems of anthropic (and other?) Perception.

Based on the rules field, the output model can be optimized. It looks something like this. We get some output statement, fix it and begin to vary the parcels. You can search for a minimum of parcels or thematic optimization, for example, by specifying a minimum of relationships, etc. Almost the same can be optimized and the very rules of the output rules. For one model, there are several correct conclusions (both universal and unique-individual). You can search for specific conclusions for models of a particular class, etc. In addition, several output fields can be supported over the model. The same matrix can have an effect on several models. In addition to all the obvious consequences of this fact, you can still see the possibility of "pairing" models and combining them into one. There are not so many such correct ways, you can even talk about some uniqueness.

The addition to the model of probabilistic relationships creates an emulation of probabilistic logic on the Associative solver. This is a very important application, because very often we will not have an exact solution, but will have to limit ourselves to a set of allowed solutions. In this case, it is possible and convenient to maintain a probabilistic logic that is very simply compatible with the exact solutions.

But apart from this, the logic of contradictions is possible - a monstrous brainchild of this approach. The fact is that we have the ability to support obviously contradictory statements in one model and draw conclusions on their basis that may turn out to be correct. The logical inconsistency and the absence of an absolute value of “true” do not frighten us at all, because the “truth” for the Associative Solver is nothing more than a context-compatible triple: model - rule - solution. There may be contradictory models for some matrices and correct in others and vice versa. Yes, and in life it is all that way.

And the strangest thing is that all this is very similar to the device of the language and the actual statement = record + protocol.

Since the periphery of the AIS should be able to not lose data, precisely execute and work in an algorithmic "manner", we can assume that these processes should be handled by the Associative Solver as the most compact and convenient processor for these occupations.

Solving technological problems

Conditionally splitting any technological task into three components: Technological map of operations (description of the sequence of operations performed); Specification of input-output products (at the entrance - abstract raw materials, at the output - an abstract result, of course, in each case - these abstractions are filled with specific content); Terms of resource coverage (description of the composition and value of the resource spent on the implementation of the technological cycle or operation); we get a unified pattern of passing tasks that arise before the Object.

The most important user property here is not so much the possibility of taking into account the passage of the task at each stage of the technological cycle, but the possibility of modifying the technological cycle based on the specific conditions of life and the development of new technological solutions.

(Remark. Of course, new solutions can arise only within the framework of the associative space elements used and specified in AIS. This is not a system, which produces optimal solutions that are not limited by the “Man-AIS” interaction properties.)

The most important property of AIS is changing its state according to the changed context. An example of this is that if one of the technological problems is not solved in time, this may entail a change in the environment of functioning as a whole, without reprogramming AIS models.

Solving communication problems

The problem of communication at the moment (if you follow Western periodicals) is fundamental to understanding the essence of Objects and their functioning. The corresponding hardware base made it possible to transfer this task to the category of purely practical. Office systems operating in one of the two main areas - GroupWare and WorkFlow - are gradually spreading in Russia.

To this topic, AIS can add its contribution to the concept of a “context menu” when the processing conditions of a focused system object change depending on the state of the space of the entire system. And also the principles of working with events can be significantly expanded.

Perception Model

We will focus on the model of perception as the closest and most interesting task of AIS in the future. No matter how we relate to computer information systems, they are nothing more than an executable algorithm, determined by both the types of data being processed and the processes of their processing. The most that we can “squeeze out” of the programming of the machine is contextuality, which is very convenient to use, modifiability, which is very convenient during operation, versatility, which removes the routine and reduces the cost of problem solving.

Only direct interaction with the user, with the person, makes the computer "intellectual" when it becomes a solver and acquires new, strange properties. This is how associativity arises when ordered data in a special way acquires a new sound, helping to analyze the material. This is how finite automata manifest themselves, which, based on the properties of the described objects and the laws of the environment, "create a cartoon" that allows to understand the essence of the phenomenon. This is how knowledge bases work that do not allow logical errors in complex transformations of the sentence system and provide us with a semi-digested product in the form of a final output. So we are absorbed by a toy that provides so many multi-component scenarios that we are not able to reveal its strategies for a long time, during which we are not bored.

AIS is a mirror of the Object, looking at which, the Object changes itself and hopes that these changes are positive to maintain its existence in the environment. AIS can also become a mirror for each specific user. By adapting the system to himself, he will create an individual structure, the analysis of which can tell a lot about its owner. If the mirror in which you look every day has the ability to preserve your features, would it not be of value in itself (carrying a terrible secret - sincerity).