RU2746687C1 - Smart enterprise management system - Google Patents

Abstract

FIELD: enterprise management system.

SUBSTANCE: system contains a hardware and software complex that ensures the functioning of a mathematical model based on the Cobb-Douglas production function in the formalism of Petri nets, connected by data transmission channels with a set of "man-machine" systems corresponding to a given functional purpose, ensuring the functioning of a dynamic mathematical model of an enterprise in a single enterprise cloud computing environment with a scalable architecture of network flows, clustering and unified software, structured on the basis of an artificial neural network of adaptive resonance, controlled by the administrator's hardware and software complex, which provides training of the artificial neural network and the allocation of enterprise resources, and connected by data transmission channels with multiple "man-machine" systems and equipment that records the state of objects and processes at the enterprise, while the neural network of adaptive resonance provides tracking and adjustment of the state of objects and processes at the enterprise.

EFFECT: technical result consists in automating the management of the state of objects and processes at the enterprise.

1 cl, 4 dwg, 3 tbl

Description

The proposed invention relates to the field of computing with elements of artificial intelligence, namely, information and analytical control systems designed to automate enterprise management processes based on an artificial neural network, simulation, computer network administration and project management.

From the patent for the invention RU 2679541, Russian space systems, publication in 2019, a project management system is known that provides the construction of a model for the implementation of satellite monitoring of a construction project and control of the current state of its execution from the standpoint of solving economic problems, aimed primarily at analyzing planned and actual indicators of the use of materials, labor costs and other accounting indicators. Analysis of the data obtained during the control process does not apply to a number of material resource components, incl. personnel, infrastructural, natural and climatic, as well as - for all intangible resource components, namely: technological, informational, moral, psychological and organizational resources, the state of which is important in the implementation of various projects.

The prototype of the invention is a system for modeling situations related to conflicts and / or competition in the field of satellite navigation systems proposed in the patent for invention RU2665045, Russian Space Systems, publication 2017. RU2665045 provides for an automated analysis of the potentials of conflicting and / or competing parties based on collection and processing of qualitative and quantitative indicators related to the parties to the conflict and / or competitive situation and to the simulated scenarios, including information about the material and non-material resources used: personnel, material and technical, natural-geographical, as well as information, psychological, technological resources, respectively , information about which is expressed only in terms of quantitative and qualitative characteristics of objects. Models are formed - information arrays that characterize each side of the conflict and / or competitive situation, embedded in the geographic information system, for simulation of the conflict and / or competitive situation.

The proposed invention is based on the conceptual provisions of the scientific and methodological apparatus for the analysis and presentation of information interaction objects presented in RU 2665045. At the same time, the proposed engineering and technical implementation of the developed mathematical models provides adequate dynamic modeling of both production processes and changes in the states of the scientific and technical products at all stages of the life cycle. Along with an increase in the level of automation of monitoring and management of the daily production and / or scientific activities of the enterprise, the proposed intelligent enterprise management system provides a solution to the problems of network planning and control, taking into account the quantitative and qualitative indicators of scientific and technical products, identifying potentially vulnerable weaknesses in the production process, including , first of all, the provision of resources, as well as predictive modeling of options for managing the processes of improving production and / or further support of scientific and technical products at the stages of the life cycle (operation, repair, etc.), which makes it possible to predict risks when developing and adopting strategies development of production. Based on the analysis of the most pressing problems of the development of modern high-tech production aimed at minimizing the human factor, the proposed invention implements scientific and technical solutions that provide automation of life cycle management processes for scientific and technical products with the function of monitoring and accounting for the influence of various factors on the state of not only material, but also intangible assets at all stages of production and life cycle of products, including forecasting by modeling scenarios for the development of various situations.

That is, unlike the prototype describing a system for modeling situations related to conflicts and / or competition, in which the potentials of competing structures in the satellite navigation services market are compared, reflecting vulnerabilities, weaknesses and / or strengths, the proposed invention provides dynamic enterprise modeling. and produced scientific and technical products at the stages of the life cycle, including tracking the implementation of the network plan for the implementation of the project for creating scientific and technical products in dynamics using intelligent control methods, based on an artificial neural network capable of self-learning in the process of functioning, identifying patterns and complex dependencies between input and output data characterizing, against the background of the network plan time scale, changes in qualitative and quantitative indicators of the state of resource components involved in production, as well as scientific and technical products at all stages life cycle. Particular emphasis in the proposed invention is placed on the accounting in the modeling of intangible, primarily financial, moral-psychological and technological production resource characteristics. When modeling the work of the intangible resource components of an enterprise, methods of network planning and production management are used at the stages of the life cycle of scientific and technical products, as well as the enterprise itself.

An intelligent enterprise management system is proposed, including a hardware and software complex that ensures the functioning of a mathematical model based on the Cobb - Douglas production function in the formalism of Petri nets, built according to demographic, material and technical, infrastructural, geographic, technological, organizational, informational, moral and psychological characteristics. (resources) connected by data transmission channels with a plurality of "man-machine" systems corresponding to a given functional purpose. Information interaction and data exchange between hardware and software, which represent "man-machine" systems, is provided by channel-forming means of receiving and transmitting information, on the basis of which a single integrated information and control space is formed, with a variety of sources and consumers of information. Unlike the prototype, the mentioned hardware and software complex ensures the functioning of a dynamic mathematical model of an enterprise in a cloud computing environment, information interaction with which is provided by a single integrated information and control space, structured on the basis of an artificial neural network of adaptive resonance (Grossberg-Carpenter network) and related data transmission channels with a variety of man-machine systems and / or physical objects corresponding to their functional purpose within the enterprise. To build a mathematical model of an enterprise, in addition to the listed geographic, technological, organizational, informational, moral and psychological characteristics, its financial characteristics (resources) are used, taking the moral and psychological characteristics (resource) as the main one.

The hardware and software implementation of the proposed enterprise management system provides not only the collection, automatic processing and analysis of large data arrays with the prompt formation of the current values of production indicators on the means of displaying information in graphical, textual and visual form, but also the functional of intelligent object management based on an artificial neural network adaptive resonance, in which, at a higher level of generalization, the analysis of dependencies between top-down (given) and observable (real) data characterizing the state of an object is carried out, the resource components of which, combining several homogeneous components of material and non-material resources, are considered as clusters of an artificial neural network. The mathematical model of the enterprise with its hardware and software implementation is built and functions as an artificial neural network of adaptive resonance, when the given (planned) production indicators of clustered resources are compared with the current controlled indicators of the corresponding clusters, on the basis of which the actual state of parallel production processes is classified and / or scientifically - technical products at the considered stage of the life cycle. Management decisions are made on the basis of an analysis of integrated indicators that come to the input of the neural network control cluster that characterize the state of resource components. In this case, the vector characteristics of the decisions made are formed at the network output. That is, based on the emerging discrepancies in the observed resource components of the current values of quantitative and qualitative indicators that differ from the specified ones, signals are generated at the output of the cluster in the artificial neural network with the corresponding control vectors that form control commands for performing corrective actions to maintain and / or restore compliance. the current values of the indicators specified.

Under the concept of an enterprise in the proposed system is meant a property complex used to carry out entrepreneurial activities, considered as a set of "man - machine" systems as part of a complex socio-technical system that has the properties of an object of information interaction. The model of such a socio-technical system is represented by a set of production elements and is characterized by the properties of an information agent, including the functional ability of receiving, processing, perceiving with subsequent actions, storing and disseminating information, reflection in accordance with the dynamics of changes in the situation under the influence of intrasystem processes and outside influence. Based on the synthesis of the classical concepts of the production function, which in mathematical economics means the functional dependence "output - costs", and an abstract dynamic system from the theory of systems, a mathematical model of the dynamic production function has been developed, which describes the totality of production processes, degradation and recovery that affect production and the state of scientific and technical products at all stages of the life cycle. When describing the functioning of the models, tangible, intangible and financial characteristics are used. With the help of a dynamic production function, the production of a product is described, which at the output consists of "a useful product" to obtain which production activity is directed, and "production waste", the volume of which should, as a rule, be reduced.

The hardware and software implementation of the proposed system provides for the creation, processing, storage and distribution of information resources between departments - participants in the production process using application software from the generalized information resource of the enterprise, such as automation tools for design and technological design, software for managing product data, means of automated planning and enterprise and production management, software and methodological tools for analyzing logistics support and maintaining databases based on the results of such analysis, software for managing workflows, software for modeling and analyzing business processes, etc. The use in the development of an automated production management system not only of the paradigm of the traditional program-targeted approach, which does not sufficiently take into account the dynamics of rapidly changing conditions of the external environment and its impact on the internal production environment, but also of the conceptual principles of project management systems, allows not only to quickly analyze the current state of resources and production in general, but also to respond flexibly to changes associated with internal processes and influences from the external environment. An important difference between the proposed invention and the prototype is the application of the basic principles of information support for scientific and technical products at all stages of the life cycle based on IPI / CALS technologies, which ensures the integration of management processes, including: project management; production configuration management; management of a single integrated information and control space; quality control; work flow management; management of changes in production structures, etc.

The practical application of the proposed intelligent enterprise management system can be described as follows.

The mathematical model of automated production management of scientific and technical products, developed on the basis of the method of network planning and management, provides modeling of the network structure of production, displaying the planned order and the current state of parallel operations that affect each other and the process of transition of products from the initial state to the final one, which characterizes the completion of the production task and / or its part. In order to optimize the computational space of the subject area by the approximation method, the network structure of production processes and / or changes in the state of scientific and technical products at all stages of the life cycle in time is set using event numbering (figure 1). In the mathematical model of project management for the creation of scientific and technical products, the production process is presented in the form of a network structure of ordered work, for the implementation of which certain resources (material, intangible and financial) are used. The fundamental model of such a network planning and management system includes a set of phases and stages (Table 1) of the development of components, the integrated state of which - d _ij characterizes the general state of readiness / operability of the product at each stage of the life cycle, which include: 1-2 - preliminary design; 2-3 - engineering and design study; 3-4 - design; 4-5 - production of a prototype; 5-6 - factory (autonomous) tests; 5-7 - acceptance (range) tests; 6-7 - revision of the sample; 7-8 - serial production; 8-9 - serial production and operation; 6-11 - operation; 9-10 - development of statistical data for modernization; 9-11 - repair, modernization and post-repair application; 10-11 - decommissioning, write-off; 11-12 - disposal.

Table 1

Phase number, name Stage Stage name I - R&D phase 1-2 Avanproekt 2-3 Engineering and design study 3-4 Design II - OCD phase 4-5 Prototype manufacturing 5-6 Factory (autonomous) tests 5-7 Acceptance (range) tests 6-7 Refinement of the sample III - phase of serial production 7-8 Mass production 8-9 Trial operation 6-11 Exploitation IV - phase of operation, repair and modernization 9-10 Generation of statistical data for repair and modernization 9-11 Repair, modernization and post-repair application 10-11 Decommissioning, decommissioning V - recycling phase 11-12 Recycling

In accordance with the methodological principles of building a project management system, a network model for the development of scientific and technical products at the stages of the life cycle has been formed. Data on the state of quantitative and qualitative indicators, one or another characteristic of the resource component at a given point in time (t) are received at the tops of the network mining model (indicated by circles). In accordance with the data received in the storage devices, the state of resource components corresponds to one of the values m ⁱ _r (t), where i = 1, 5. Scientific and technical products consist of a significant number of resource components, each of which is modeled as a production element and works according to the corresponding production function. The generalized network model for the release of scientific and technical products provides a solution to the problem of parallel functioning with the task of achieving a given state (reaching the corresponding peak), if not fulfilled, the state is classified as an unreached level with the formation of a control action aimed at achieving a given state, including by changing the involved resources per unit of time, as well as the application of the "critical path" method, described by the expression d _ij = {t _ij , s _ij }, where t _ij is the operation time - (i, j); with _ij - the set of resources spent on the operation. In turn, the set of production elements integrated into the generalized model form a network model for the development of scientific and technical products at the stages of the life cycle.

The mathematical model of production activity is described by an equation of state that takes into account the processes of changing quantitative and qualitative characteristics used in the production of material and intangible resources: R (t) = <R ^m (t), R ⁿ (t)>, where R ^m (t) - vector of material, and R ⁿ (t) is a vector of intangible resources, which, respectively, are described by expressions of the form R ^m (t) = <X (t), Y (t), Z (t), G (t)> and R ⁿ (t) = <Т (t), Н (t), I (t), Ψ (t)>, where:

X (t) - human resource (X-resource) is characterized by the staff involved in the execution of the n-th production task (n = 1, N), and is described by an expression of the form X (t) = <M ^xn (t), K ^xn _b (t)> (n = 1, N; b = 1, B), where М ^хn (t), K ⁿ _b (t) - quantitative and qualitative indicators of the X-resource;

Y (t) - material and technical resource (Y-resource) is characterized by the presence, efficiency and composition of the means of production involved in solving problems for the production of scientific and technical products, and is described by an expression of the form Y (t) = <M ^yn (t), K ^yn _b (t)> (n = 1, N; b = 1, B), where M ^yn (t), K ^yn _b (t) are the quantitative and qualitative indicators of the Y-resource;

Z (t) - infrastructure resource (Z-resource), characterized by the state of infrastructure and logistics, described by an expression of the form Z (t) = <M ^zn (t), K ^zn _b (t)> (n = 1, N; b = 1, B), where M ^zn (t), K ^zn _b (t) - quantitative and qualitative indicators of the Z-resource;

G (t) - a geographic resource (G-resource) is characterized by indicators of the geographical space in which production activities are carried out, taking into account natural-climatic, landscape, soil and other features, is described by an expression of the form G (t) = <M ^gn (t) , K ^gn _b (t)> (n = 1, N; b = 1, B), where M ^gn (t), K ^gn _b (t) are the quantitative and qualitative indicators of the G-resource;

T (t) - technological production resource (T-resource) is characterized by a set of effective technological methods that are used in production activities, and is described by a vector of indicators, of the form T (t) = <K ^тn _b (t)> (n = 1, N ), where K ^тn _b (t) is the vector of indicators of the qualitative characteristics of the b-th effective technological method (b = 1, B); B is the total number of effective technological methods that can be used to implement the n-th task, N is the total number of jobs (operations);

H (t) - organizational resource (H-resource) is characterized by the degree of coordination of the actions of various departments and shows what part of the actions taken contributes to the release of "useful products", and what is spent on "waste", is described by a vector of indicators, of the form H (t) = <M ^Нn (t), K ^Нn _b (t)> (n = 1, N; b = 1, B), where М ^Нn (t), K ^Нn _b (t) are the quantitative and qualitative components of the organizational Н-resource , respectively, while M ^Нn (t) characterizes the general level of organization and production management in the implementation of the n-th task (n = 1, N), and K ^Нn _b (t) displays which part of the organizational resource can be used to implement the b-th (b = 1, B) an efficient technological method;

I (t) - information resource (I-resource) in a generalized form characterizes the amount of knowledge, experience, skills, abilities and software that are used in production, is described by a vector of indicators, of the form I (t) = <М ^In (t), K ^In _b (t)> (n = 1, N; b = 1, B);

Ψ (t) - a moral and psychological resource (Ψ-resource) is characterized by the forces of an incentive and incentive that form the desire to accurately and on time to complete the production task, is described by a vector of indicators, of the form Ψ (t) = <M ^Ψn (t), K ^Ψn _b (t)> (n = 1, N; b = 1, B), where M ^Ψn (t), K ^Ψn _b (t) are quantitative and qualitative indicators characterizing the state of the Ψ-resource;

S (t) - financial resource (S-resource) is characterized by the state of the financial sector, described by an expression of the form S (t) = <M ^Sn (t), K ^Sn _b (t)> (n = 1, N; b = 1, B), where М ^Sn (t), K ^Sn _b (t) - quantitative and qualitative indicators of the S-resource.

Each type of material, intangible and financial resources is combined into appropriate clusters, which, in turn, include components characterized by their quantitative and qualitative indicators. Over time, in the production process, the quantitative and qualitative indicators of resource components change under the influence of processes within the socio-technical system and external factors. Since the functioning of the socio-technical system is associated with processes of an economic nature, in particular, the implementation of the results of its production activities, the acquisition and sale of material and intangible resources, etc., the generalized model of the production element provides for one that does not directly participate in production, but has an important value and having certain specific properties, a financial resource - S (t), also characterized by its quantitative and qualitative indicators, for example, the exchange rate and purchasing power. The mathematical model of a financial resource is described by an expression of the form S (t) = <M ^S (t), K ^S (t)>, where M ^S (t) and K ^S (t) are quantitative and qualitative indicators. Thus, the generalized mathematical model of a production element is described by an equation of state of the form R (t) = <R ^m (t), R ⁿ (t), S (t)> = <X (t), Y (t), Z (t ), G (t), Т (t), Н (t), I (t), Ψ (t), S (t)>.

In the mathematical model of an enterprise, not only intangible resources are described in most detail, but also their interaction with material components, as well as mutual influence on each other, incl. based on statistical data, which makes it more adequate and consistent with reality in dynamics. Among the intangible resource components, the moral and psychological resource - Ψ (t) is considered as one of the most important components both for the successful implementation of projects and production programs, and during the use of scientific and technical products in the process of performing the "Operation, repair and modernization" phase ... The basic psychophysical law of Weber - Fechner, which is generally described by the expression ε (t) = π [S (t), S _max (t), S _min ( t)], where ε (t) is the driving force of emotional experience; S (t) - the current state of the stimulus image; S _max (t) - the upper threshold (normative) value, identified with the desired “level of good” (level of claims) of the employee, which is used in the artificial neural network in the class related to “real indicators”; S _min (t) - lower threshold (normative) value, identified with the state of "unacceptable level of damage caused", used by an artificial neural network in the class of "real / planned indicators"; π [*] - logarithmic function. This mathematical model takes into account two classes of sets: a motivational component and a stimulating component. The motivational component encourages the employee to act and is described by the expression ΔS _m (t) = [S _max (t) -S (t)] / [S _max (t) -S _min (t)]. The stimulating component, affects and irritates the employee, is described by the expression ΔS _with (t) = [S (t) -S _min (t)] / [S _max (t) -S _min (t)]. The resulting emotional response to a stimulus is described by an expression of the form ε (t) = {μ _m (t) logΔS _m (t) + μ _s (t) logΔS _s (t)}, where μ _m (t) and μ _s (t) μ _m (t) + μ _с (t) = 1 are the coefficients of motivation and incentives, respectively, showing the significance of the production element for modeling the moral and psychological state of the employee, the experience associated with the desire to approach a state that brings him joy, and the experience associated with a desire to avoid a state that causes him dissatisfaction.

The mathematical model of the resulting emotional reaction for the totality of the constituent sets: psychophysical, psychosocial and spiritual, the sum of the coefficients of which ρ _f (t) + ρ _p (t) + ρ _d (t) = 1, is described by the expression E (t) = [ε _f (t)] ρ _f (t) + [ε _p (t)] ρ _p (t) + [ε _d (t)] ρ _d (t). In turn, the mathematical model of the target functional of the Ψ-resource is described by an expression of the form Ж _i [U _{* i (t)} , T _и ] = max _t0 ∫ ^Т Ξ _i [t, U _{* i (t)} ] dt (Т∝ →) , where Ξ _i [t, U _{* i (t)} ] = ∑ _{jωij (t)} ϒ _ij [t, U * i (t)], ϒ _ij [*] is the evaluation function (reflection) of the i-th production element, characterizing the assessment of the level of satisfaction with the production element that simulates the employee, both his individual needs (j = i), and the satisfaction of his needs by other production elements that are part of the moral and psychological resource cluster; ω _{ij (t)} - expressivity coefficient (-1≤ω _{ij (t)} ≤ + 1) shows the significance for the i-th production element of the subjective state of the j-th production element, which takes a value on the interval [-1, + 1], at the same time, ∑j⏐ω _ij (t) ⏐ = 1 (i.e., at ω _ii (t) ≈0, there is a tendency to altruism, at ω _ii (t) ≈-1 - a depressive state of indifference, at ω _ii ( t) ≈1 - clearly expressed egoism), U * _ij (t) - the desired vector of control, including the vector of control of the sphere of material production, socio-political and spiritual-moral spheres, U * _пj (t) = <U * E _j (t), U * P _j (t), U * AND _j (t)>. In general, the resulting state of the Ψ-resource, which characterizes the level of moral and psychological readiness to perform a production task, is described by an expression of the form M _Ψ (t) = <m ¹ _Ψ (t), m ² _Ψ (t), m ³ _Ψ (t), m ⁴ _Ψ (t), m ⁵ _Ψ (t)>, where m ¹ _Ψ (t) corresponds to the complete moral and psychological readiness of personnel to solve production problems; m ² _Ψ (t) - shows a sufficiently high level of readiness at which the existing problems are solved using methods of moral encouragement; m ³ _Ψ (t) - corresponds to the average level of readiness, requiring additional moral and material incentives; m ⁴ _Ψ (t) is an indicator indicating moral and psychological dissatisfaction and serious social instability; m ⁵ _Ψ (t) is an indicator characterizing the absolute moral and psychological unwillingness and unwillingness to perform production tasks.

The assessment of the level of the technological resource, in accordance with which the technological readiness of the enterprise for the production of scientific and technical products is assessed, is carried out on the basis of the audit data and the calculation of the technological readiness index for each production operation in accordance with the technological documentation. The level of technological readiness is assessed according to one of five states: m ¹ _T (t) - "fully executed", when, according to the audit data, the technology used ensures the efficient execution of certain types of operations; m ² _T (t) - "performed to a significant extent" means, on the whole, good technological support with some drawbacks that do not significantly affect the performance of the operation with the specified indicators; m ³ _T (t) - "performed partially, fragmentarily" the state of the technological resource component in the n-th operation does not sufficiently ensure its performance with proper indicators; m ⁴ _T (t) - "almost not performed" means a low level of technological equipment, which cannot provide the specified indicators when performing the production task; m ⁵ _T (t) - "not applicable" means that the state of the technological resource for the operation of the n-th type excludes its execution. The generalized indicator of the technological resource, on the basis of which the technological readiness of the enterprise for the production of scientific and technical products of the n-th type is assessed, is described by an expression of the form М ⁿ _Т (t) = <m ¹ _Т (t), m ² _Т (t), m ³ _Т (t), m ⁴ _T (t), m ⁵ _T (t)>.

The engineering and technical implementation of the functional of the production element is based on the apparatus of the formalisms of Petri nets, adapted to the problems being solved by expanding the content of the concepts of "generalized accumulator" and "generalized transition" (figure 2). Generalized storage devices are indicated by rectangles and circles of elementary storage devices inserted into them, from which models of generalized production elements are formed as part of the corresponding receiving clusters at the input of an artificial neural network, including packets of elementary storage devices, in which signals with corresponding data and vectors of parameters characterizing their current state are generated , changing in accordance with the parameters of signals arriving from generalized transitions (represented by bars) that simulate vector functions. Thus, the structure of the generalized mathematical model of a production element in the formalisms of Petri nets forms a neural network of adaptive resonance of the Grossberg - Carpenter type, which contains clusters of resources in model blocks that ensure its functioning. The main component of the network is the production area of the production element, since its indicators determine the state of the resources involved in production and the degree of implementation of the production task according to the network plan in time.

The production area of the production element performs the main function of production control and management (figure 2). It receives data on the state of the products produced at the previous stages, monitors the conditions and output of products, including "useful products" to the control unit of manufactured products, which, when the quantitative and qualitative indicators correspond to the specified ones (a permitting sanction is activated), enters the production the sphere of the production element of the next stage, and waste that enters the recycling unit, through which the modeling of the use of waste as recyclable materials is provided) The production sphere interacts with the "Stocks" block, which controls the availability of resources and ensures their replenishment in cooperation with the "finance" module by generating a signal for the execution of financial transactions, as well as with the current management block, interacting with the managerial representation and performing a management function, including permission and / or termination of the work process. In the model of the production element, the b-th (b = 1, B) effective technological method is used, characterized by indicators that correspond to each technological operation and are introduced during the parametric adjustment of the production area of the production element (figure 2).

The output of useful products - π ^p (t) is implemented by a generalized transition - d ^f , the work of which is described by the functional dependence “output - costs” of the form π ^p (t) / dt = <π ^m (t), π ^k (t), S ( t)>, where π ^m (t) / dt = a ^fm _b E ^k _b (t) U _н (t) [П ^м _bп (t)] ^αмb [П ^н _bп (t)] ^αнb S (t)] and π ^k (t) / dt = a ^fk _b E ^k _b (t) U _n (t) [P ^m _bp (t)] ^γmb [P ⁿ _bp (t) S (t)] ^γнb indicators of the quantity and quality of output useful product; a ^fm _b and a ^fk _b - scale factors of the "output" efficiency of the b-th technological method; E ^m _b (t) and E ^k _b (t) - indicators of the influence of technological conditions affecting the rate of formation of quantitative and qualitative characteristics when using the b-th technological method; P ^m _bp (t) and P ⁿ _bp (t) are the production potentials of the resources used for the production of products by the b-th technological method; α ^m _b and α ⁿ _b - indicators of the contribution of production potentials to the formation of quantitative characteristics of the manufactured product ("useful" - ^m and waste - ⁿ ) when using the b-th technological method (α ^m _b + α ⁿ _b = 1); γ ^m _b and γ ⁿ _b are indicators of the contribution of production potentials to the formation of qualitative characteristics of the manufactured product ("useful" - ^m and waste - ⁿ ) when using the b-th (b = 1, B) technological method (γ ^m _b + γ ⁿ _b = 1); U _n (t) is an indicator of the control vector that sets the intensity of the use of material, intangible and financial resources. With the parametric tuning of the production element model, the performance indicators - a ^fm _b and a ^fk _b , the indicators of the degree of quantitative and qualitative changes - α ^m _b and α ^k _b , γ ^m _b and γ ^k _b are entered into the drive P ^Ut of the current control unit, from where through transitions d ^f and d ^g go to the storage devices of the production sphere of the production element, forming the influence of the b-th technological method on the output. Specified indicators of technological conditions - E ^m _b (t) and E ^k _b (t), production potentials - P ^m _bp (t) and P ⁿ _bp (t) and control vector - U _n (t), are relative and correspond to scalar value.

When comparing the set (planned) and real values of production indicators, the resulting data with a positive value indicates the effectiveness of the technology used in production and vice versa, a negative value is an indicator of its inefficiency. Indicators of production of useful products - π ^p (t) enter the product control system, where it is rolled up as a generalized indicator of the form r ^p (t) = π ^m (t) π ^k (t), and then integrated into an expression of the form π ^p (t ) = ₀ ∫ ^Т r ^p (t) dt (r ^p (t ₀ ) = 0). Further, the obtained value of π ^p (t) in the transition d ^to in the accumulator P _to (t) is compared with the given π ^{p *} (t) under the condition π ^p (t) ≥ π ^{p *} (t), in accordance with which it is authorized by the signal vector in case of a positive result in the production element, the execution of the next operation (stage) or in case of a negative result of the comparison is blocked with a display of an abnormal situation on the information display means with the corresponding color and sound signal. Generalized branch - d ^g tracks changes in the resources of the production item. The mathematical model of the operation of the generalized transition d ^g with a set of drives R ^m (t), R ⁿ (t), which control the characteristics of the resource components of the r-th type, which at the current time correspond to one of the five operability states (the number is given by the superscript), is described expression of the form M _r (t) = <m ¹ _r (t), m ² _r (t), m ³ _r (t), m ⁴ _r (t), m ⁵ _r (t)>, where m ¹ _r ( t) - indicator of the state of resource components that are in a fully operational state; m ² _r (t) - the indicator of the health of the resource components is at least 90% of the specified; m ³ _r (t) - the indicator of the health of resource components is at the level of 90-75% of the specified; m ⁴ _r (t) - the indicator of the health of resource components is at least 50%; m ⁵ _r (t) - indicator of the health of resource components less than 50%.

The change in characteristics in the model of the state of material and intangible resources, which determines the work of the generalized transition - d ^h in the production of useful product b (b = 1, B), depends on the processes of degradation and recovery, which together determine the current dynamics of change in the state of health production item. The model of the processes of changing the state of resource elements (figure 3) displays the change in the states of operability, the indicators of which depend on the processes of natural degradation caused by wear and tear of resource components; forced degradation caused by the action of both external and internal destructive influences of a purposeful and non-purposeful nature; compulsory recovery associated with targeted activities to restore the lost resource components of performance. A mathematical model of the processes affecting the change in the state of resource components using the b-th technological method, a mathematical model of the logistic function is used in the form of a differential equation Dq _r (t) / dt = w _r (t) [1 - q _r (t) / q * _r ] q _r (t), (0 <q _r (t ₀ ) <1; q * _r = 1), where: q _r (t) is an indicator of the amount of wear of resource elements of the r-th type at the time (t ). The process of changing the states of resource components is carried out through the transition bars, which set the intensity of the signal flow from drives with a lower number (v) to drives with a higher number (n), caused by natural and external processes.

Modeling of processes affecting changes in the states of resource components is implemented in the formalism of Petri nets (figure 3) using transitions d _r ^vn (v ≠ n; v, n = 1.5; v> n), which with intensity - λ _ρ ^vn ( t) (the lower control input of the transition d _r ^vn ) move the indicators of the state of the resource components of the r-th type from the storage unit (vertex) with number v to the storage unit (vertex) with the number n. In general, the intensity values - λ _r ^vn (t) do not remain constant, but change over time. The nature of their change depends on the technical (physical) characteristics of the resource components, the effective technological method used, internal and external conditions. The model reflecting the intensity of the forced degradation of resource elements of the r-th type (figure 3) is implemented in the formalisms of Petri nets using the transition bar - d _r ^vn (v ≠ n, v, n = 1.5; v> n), which with with intensity μ _ρ ^vn (t) (upper control input of the transition d _r ^vn ), the indicators of resource components of the ρ-type are moved from the accumulator (vertex) with number (v) to the accumulator (vertex) with number (n). The nature of the change depends on the influence of the external environment. In the presented mathematical model, the intensities of natural - λ _r ^vn (t) and forced - μ _r ^vn (t) degradation are independent and possess the property of additivity, that is, the transition d _r ^vn performs the procedure for adding the intensity of natural - λ _r ^vn (t) and forced - μ _r ^vn (t) degradation - ν _r ^vn (t) = λ _r ^vn (t) + μ _r ^vn (t).

The model reflecting the processes of forced recovery (Figure 3) is implemented in the formalisms of Petri nets, is implemented by transitions d _r ^kn with reverse numbering (the larger index precedes the smaller one) and is characterized by the recovery rate - β _r ^vn (t) (v ≠ n, v <n ), which is determined by the volume of resources involved in restoration and the technological conditions of restoration in the course of the production process. In this case, the value β _r ²¹ (t) corresponds to the intensity of the current repair, the value β _r ³¹ (t) - the intensity of the average repair, the value β _r ⁴¹ (t) - the intensity of the major repair. Recovery in the physical (technical) state m ⁵ _r (t) is impossible, and therefore it is considered an "absorbing" state. Intensities λ _r ^kn (t) (v ≠ n, v> n, k, n = 1.5), μ _r ^vn (t) (v ≠ n, v> n, v, n = 1.5) and β _r ^vn (t) (v ≠ n, v <n, v, n = 1,4) are formed using the corresponding model blocks. In the model of the production element, the intensity of purchases δ _r (t) and utilization γ _r (t) are formed by the model blocks "Procurement Management" and "Utilization Management", which are implemented by transitions d ^z _r and d ^y _r .

The mathematical model of the change in the state of resource elements of the r-th type is described by a system of differential equations of the form

dm ¹ _r (t) / dt = -m ¹ _r (t) [λ _r ¹² (t) + λ _r ¹³ (t) + λ _r ¹⁴ (t) + λ _r ¹⁵ (t)] + [μ _r ¹² (t) + μ _r ¹³ (t) + μ _r ¹⁴ (t) + μ _r ¹⁵ (t)]} +

+ m ² _r (t) ⋅ β _r ²¹ (t) + m ³ _r (t) ⋅ β _r ³¹ (t) + m ⁴ _r (t) ⋅ β _r ⁴¹ (t) + m _r ⁰ (t) ⋅ δ _r (t);

dm ² _r (t) / dt = m ¹ _r (t) ⋅ [λ _r ¹² (t) + μ _r ¹² (t)] - ² _r (t) ⋅ {[λ _r ²³ (t) + λ _r ²⁴ (t) + λ _r ²⁵ (t)] + [μ _r ²³ (t) + μ _r ²⁴ (t) +

+ μ _r ²⁵ (t)]} - m ² _r (t) ⋅β _r ²¹ (t) + m ³ _r (t) ⋅ β _r ³² (t) + m ⁴ _r (t) ⋅ β _r ² (t );

dm ³ _r (t) / dt = m ¹ _r (t) ⋅ [λ _r ¹³ (t) + μ _r ¹³ (t)] + m ² _r (t) ⋅ [λ _r ²³ (t) + μ _r ²³ (t)] - m ³ _r (t) ⋅ {[λ _r ³⁴ (t) + λ _r ³⁵ (t)] +

+ [μ _ρ ³⁴ (t)] + μ _ρ ³⁵ (t)]} - m ³ _ρ (t) ⋅ [β _ρ ³¹ (t) + β _ρ ³² (t)] + m ⁴ _ρ (t) ⋅ β _ρ ⁴³ (t);

dm ⁴ _r (t) / dt = m ¹ _r (t) ⋅ [λ _r ¹⁴ (t) + μ _r ¹⁴ (t)] + m ² _r (t) ⋅ [λ _r ²⁴ (t) + μ _r ²⁴ (t)] + m ³ _r (t) ⋅ [λ _r ³⁴ (t) + μ _r ³⁴ (t)] -

-m ⁴ _r (t) ⋅ [λ _r ⁴⁵ (t) + μ _r ⁴⁵ (t)] - m ⁴ _r (t) ⋅ [β _r ⁴¹ (t) + β _r ⁴² (t) + β _r ⁴³ ( t)];

dm ⁵ _r (t) / dt = m ¹ _r (t) ⋅ [λ _r ¹⁵ (t) + μ _r ¹⁵ (t)] + m ² _r (t) ⋅ [λ _r ²⁵ (t) + μ _r ²⁵ (t)] + m ³ _r (t) ⋅ [λ _r ³⁵ (t) + μ _r ³⁵ (t)] -

-m ⁴ _r (t) ⋅ [λ _r ⁴⁵ (t) + μ _r ⁴⁵ (t)] - m _r ⁵ (t) ⋅ γ _r (t), under the normalizing condition ∑ _n m ⁿ _r (t) = m ^Σ _r (t) (n = 1.5), where m ^Σ _r (t) is the total number of resource elements of the r-th type, for the given initial conditions M _r = <m _r ⁰ (t ₀ ), m ¹ _r ( t ₀ ), m ² _r (t ₀ ), m ³ _r (t ₀ ), m ⁴ _r (t ₀ ), m ⁵ _r (t ₀ )>, where m _r ⁰ (t) is the total number of resource elements r -th type, directed to replenish resource components that have lost their performance.

A feature of the structure of a modern enterprise is its internal network structure, formed by jointly functioning and interacting production elements with the above set of resource components that participate in the production process and influence each other. This interaction affects the "release" of the aggregate "useful" product and the "costs" of resource components of each production element involved in this process. In the generalized model of a production element, mutual influences are taken into account in the entire set of resource components, which are elements of a network system operating in a distributed network information space. The mutual influence of the resource components of a production element is taken into account using a macro-model - T ^G _j (t), which is used for parametric adjustment of functional dependencies that determine the indicators of degradation (wear) rates described by the Cobb-Douglas production function, of the form

w _xi (t) = [w ^o _xi D ^w _xi (t) u _Нi (t) h _i (t)] [Р _xi (t)] ^γxxi [Р _yi (t)] ^γxyi [Р _zi (t)] ^γxzi ,

w _yi (t) = [w ^o _yi D ^w _yi (t) u _Hi (t) h _i (t)] [Р _xi (t)] ^γyxi [Р _yi (t)] ^γyyi [Р _zi (t)] ^γyzi ,

w _zi (t) = [w ^o _zi D ^w _zi (t) u _Hi (t) h _i (t)] [Р _xi (t)] ^γzxi [Р _yi (t)] ^γzyi [Р _zi (t)] ^γzzi , etc., where

w ^o _xi , w ^o _Yi , w ^o _Zi , etc. are the levels of degradation rates of the resource components of the production element, which are statistical for the i-th production process; D _i (t) - indicator of the level of the technological method; Р _xi (t), Р _Yi (t), Р _Zi (t) - indicators of production potentials of resource components. At Р _xi (t) ≈1, Р _Yi (t) ≈1 and Р _zi (t) ≈1, i.e. when the factors are close to their optimal values), the wear rates - w _xi (t), w _yi (t), w _zi (t), etc. will be close to their specified values - w ^o _xi , w ^o _yi , w ^o _zi , etc., the exponents in characterize the degree of cross-effects of the production potentials of the resource components involved in the production process on the degradation rate of each other and satisfy the conditions γ ^x _xi + γ ^x _yi + γ ^x _zi = 1, γ ^y _xi + γ ^y _yi + γ ^y _zi = 1, γ ^z _xi + γ ^z _yi + γ ^z _zi = 1; h _i (t) - the coefficient of running-in (accumulation of experience) is calculated by a formula of the form dh _i (t) / dt = ω ⁰ _i [1-h _i (t) / h * _i ] h _i (t) (0 <h _i (t ₀ ) <1; h * _i = 1), where: ω ⁰ _i - coefficient characterizing the speed of ^{running-in (ω 0} _i ≥0); h * _i - the limit threshold of the running-in value.

In a generalized model of a complex socio-technical system, mutual influences are taken into account in the entire set of production elements, which in this case model the elements and nodes of a network system operating in a distributed network space. The values of the indicators of mutual influences on the quantitative (V ^m ) and qualitative (V ^k ) characteristics of production elements, calculated on the basis of statistical data, are used to parameterize the macro-model - T ^G _j (t). The corresponding control vectors are fixed in blocks that analyze the intensity of natural and / or forced degradation, as well as recovery. Since the resource components undergo two types of changes - quantitative and qualitative, the corresponding data are formed in two tables: a table with V ^m indicators, which are used to parameterize functional dependencies describing the processes of quantitative degradation of resource components (Table 2), and a table with V indicators ^k , which are used for parametric adjustment of functional dependencies describing the processes of qualitative changes in resource components (Table 3).

table 2

T ^G PE ₁ PE ₂ ... ... ... PE _N-1 PE _N PE ₁ one V ^m _1/2 ... ... ... V ^m _{1 / N-1} V ^m _{1 / N} PE ₂ V ^m _2/1 one ... ... ... V ^m _{2 / N-1} V ^m _{2 / N} ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... PE _N-1 V ^m _N-1/1 V ^m _N-1/2 ... ... ... one V ^m _{N-1 / N} PE _N V ^m _{N / 1} V ^m _{N / 2} ... ... ... V ^m _{N / N-1} one

Table 3

T ^G PE ₁ PE ₂ ... ... ... PE _N-1 PE _N PE ₁ one V ^k _1/2 ... ... ... V ^k _{1 / N-1} V ^k _{1 / N} PE ₂ V ^k _2/1 one ... ... ... V ^k _{2 / N-1} V ^k _{2 / N} ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... PE _N-1 V ^k _N-1/1 V ^k _N-1/2 ... ... ... one V ^k _{N-1 / N} PE _N V ^k _{N / 1} V ^k _{N / 2} ... ... ... V ^k _{N / N-1} one

The vector function of the costs of resource components from mutual influence is given by an expression of the form

dR ^m _Pi (t) / dt = h _i [t, T ^G _j (t), T _i (t), G _i (t), {R _Pi (t)}, {V ^m _i } _T ^G _j , U (t)],

dR ^k _Pi (t) / dt = g _i [t, T ^G _j (t), T _i (t), G _i (t), {R _Pi (t)}, {V ^k _i } _T ^G _j , U _i (t)], where

T ^G _j (t) and T _i (t) - respectively, "external" macro-model and "internal" macro-model of the i-th production element from the network structure of the generalized model of the socio-technical system; {R _Рi (t)} is a vector of resource components functioning as part of a generalized model of a production element (i = 1, N); N is the total number of production elements (PE _{1 ... N} ) as part of the generalized model of the socio-technical system {V ^m _i } _T ^G _j and {V ^k _i } - the i-th column vectors in the tables of dependencies of resource components [V ^m ] _T ^G _j and [V ^k ] _T ^G _j , respectively. Thus, the obtained values of the values of the indicators of the "useful product" and industrial waste are analyzed in accordance with the specified criteria and the described states of the controlled system, which provides cluster analysis and processing of dependencies between the input and output data of an artificial neural network of adaptive resonance of the Grossberg-Carpenter type.

The criteria for the fulfillment of the production task are the conditions characterized by the correspondence of the actual (current) - R (T ₀ ) / K (T ₀ ) data to the specified (planned) - R _З (T ₀ ) / K _З (T ₀ ) quantitative and qualitative, respectively , indicators of manufactured scientific and technical products, namely:

- condition No. 1, characterizing the quantitative indicators of the release of a "useful product" R (T ₀ ) ≥R _З (T ₀ ), where R (T) = ₀ ∫ ^Т r (t) d (t), T ₀ is the production time tasks, R (T ₀ ) - integral production;

- condition No. 2, characterizing the quality indicators of manufactured products K (T ₀ ) = 1-Q (T ₀ ) ≥K _З (T ₀ ), where Q (T ₀ ) is an integrated indicator of product quality decline.

When the specified conditions are met within the specified deviations, the graphical user interface corresponds to the normal state of operability of the modeled social and technical system and compliance with the specified characteristics of the "useful product" being produced. In the event of an abnormal situation in the production process and / or a change in the values of quantitative and / or qualitative indicators that differ from the set values above the set values, a signal is generated and the necessary information is displayed on the information display means. The control vectors through the storage units m ⁱ _r (t), where i = 1, 5, are sent to the transitions d _r ^З (procurement management) to indicate the need to promptly resolve issues of replenishing the corresponding resource components and / or to the d _r ^У block to signal operational solving the problem of withdrawing from the production process and disposing of components, the state of operability of which does not correspond to the conditions necessary to fulfill the set production task. On the basis of the methods of forming an artificial neural network of adaptive resonance, software and hardware solutions of the intelligent component of the network model of an automated enterprise management system are implemented, which provides recognition of the state of resource components based on the results of the analysis of specified (planned) values, which are compared with the actually observed indicators and monitoring data of the state of controlled objects. and processes received from information sources, which include sensor, sensor and other equipment that records the state of objects and processes. Comparative analysis of the set values and current indicators of the state of monitored resources ensures that they belong to one of five categories that characterize the state of health and / or readiness of resource components to fulfill planned targets. When the values of the difference in the analyzed data in the interval between the planned and current values are within the limits not exceeding the specified values, the monitored resource component automatically belongs to one of five states.

The proposed intelligent enterprise management system based on an artificial learning model with a "teacher", the function of which is performed by the system administrator (figure 4), who enters the target values and carries out the parametric tuning of the models. Technical training of an artificial neural network consists in identifying internal relationships and mutual influences inherent in information exchange between clusters of resource components and is carried out, in particular, with the help of examples and statistical data given in tables 2 and 3 of functional dependencies. An important component of an intelligent enterprise management system is a modeling and management unit (figure 4), which provides an analysis of indicators of the main characteristics at each stage of the production process, such as: execution time (from the beginning to the end of each operation), the planned level of resource provision and actual costs resources at each current moment of time. Special control is exercised over the execution of critical operations that require the allocation of additional slack and a reasonable reservation of resources. The algorithm for processing data and assessing the indicators of the network model assumes parallel work on the analysis of the actual data coming from the sensor equipment and technological control means for each operation using methods of parallel analysis of real monitoring data and indicators of network planning and management with the subsequent generalization and formation of dependencies that provide in the event of signs of emergency situations, the use of the "critical path" method.

The software and hardware implementation of the simulation cluster is made in the formalisms of the language of Petri nets (figure 4). In the model of the j-th production program, the i-th production elements have a two-level architecture consisting of: the upper level - the information and psychological area (Q _i ), in which the processes associated with the functioning of the intangible production component are modeled, including the formation of predictive models, the control vector, and etc.; the lower level - the subject area (F _i ), which provides modeling and analytical processing of the production process data. Each (i-1, i-th) operation in the network model is assigned a bar (transition) - d _i . The current state of the network model is characterized by a marker with a signal corresponding to the health / readiness state formed in the drive, indicating the event of the start of the i-th operation - Р _i-1 , the completion of the (i-1, i) -th operation coincides with the opening of the transition - d _i and the arrival of the token, from the drive - P _i-1 to the final drive for this operation - P _i . The network transition - d _{i is} opened after the i-th production element has completed all the required amount of work with the required quality. This is signaled by the appearance at the control input of the transition - d _{i of a} marker coming from the i-th production element. From the aggregate of drives of the network model, generalized drives are formed, which contain n elementary drives, where n is equal to the number of data included in it in the form of a certain signal about operations, and k is the number of elementary drives, equal to the number of data on the execution of operations outgoing from it.

The proposed dynamic mathematical model of an enterprise functions in a cloud computing environment, structured as an artificial neural network of adaptive Grossberg-Carpenter resonance. The cloud computing environment is associated with data transmission channels with a variety of human-machine systems and / or physical objects corresponding to their functional purpose within the enterprise. The implementation of automated control functions using cloud tools with an assessment of production, product status and work progress, according to the proposed mathematical model, with appropriate visualization based on internal information exchange and assessment of current resource indicators with specified (control) indicators.

A single integrated information and control space of cloud computing with a scalable architecture of network flows, clustering and unified software is formed on the basis of hardware and software of the data processing center by combining the existing computing resources of the enterprise and provides information exchange between the entire set of man-machine systems. The hardware and software implementation of a single integrated information and control space includes a distributed data warehouse that exists in a networked computer environment, covering all services and divisions of the enterprise that carry out the processes of information exchange with each other and information interaction with third-party structures at all stages of the product life cycle. At the same time, access to information resources and cluster solutions is provided, including through the formation of a corporate cloud infrastructure, which includes cloud modules of enterprises - co-executors and / or interacting organizations. In a single integrated information and control space, a unified system of rules for the presentation, storage and exchange of information operates, in accordance with which information processes proceed that accompany and support research, development and other engineering and technical work at all stages of the product life cycle. Information that once emerged at any stage of the life cycle is stored in a single integrated information and control space and becomes potentially available to all project participants, which allows you to avoid duplication, re-coding and unauthorized data changes, reduce the number of errors, reduce labor and time costs and optimize all kinds of resources.

The implementation of the intellectual component of the process of intelligent management of big data is carried out using the functional properties of the cloud platform, which, among other things, includes horizontally scalable software tools that provide massively parallel processing of incoming information, such as a database management system, including use of NoSQL graph storages; software based on software frameworks that integrate various software modules; a Hadoop-based HDFS file system for maintaining the project library. In addition, the cloud infrastructure provides the implementation of the analytical function for assessing production, product status and work progress with appropriate visualization based on internal information exchange and assessment of current resource indicators with specified (planned) values. Users of this system are given the right to access services that provide analytical processing of data on the progress of the project, control the current state of resource components and identify signs of violation of the specified production indicators; predictive modeling and automatic assessment of the state of products in pseudo-real time mode throughout the life cycle of scientific and technical products, including the progress of work on development, creation, operation, up to disposal; computer-aided design, logistics support, control and accounting of all types of resource components; automation of work with design, technical and operational documentation, including the function of documenting, accounting and monitoring the implementation of network planning and management activities that determine the state of the product at each stage of the life cycle; automatic data exchange and comparative analysis of accounting indicators, economic planning and analysis models of indicators of the current state of resources in comparison with the specified (planned) values; automated production management, including high reliability and efficiency of control and assessment of qualitative and quantitative indicators.

Cloud infrastructure ensures data synchronization of the entire set of corporate cloud platforms and thereby ensures a high level of security and at the same time availability of meaningful information. In a single integrated information and control space, functional modules of cloud computing are allocated, including: "Management of virtual machines based on the operating system of the Linux family"; "Data storage management" (including multimedia storage management); "Management of network resources" (including IP-addresses, domains, etc.); "Algorithms for automatic scaling of resources"; "Monitoring of resources and information support of the life cycle of scientific and technical products and resource components of the manufacturer"; "Monitoring the integrity and ensuring integrated security"; "Linux operating system templates for WEB"; "Management of virtual machines based on Windows OS"; "Monitoring the status and testing of cloud applications and WEB-sites"; "Management of a network geographically distributed data storage system"; "Services of geographic information systems"; "Counterparties".

An integral part of the system, with the help of which the configuration and management of the functionality of generalized production elements is carried out, is the automated workstation of the system administrator (figure 4), which is the hardware and software complex of the head performer of the project. In the implementation of the function of system administration, information and analytical support and management, hardware and software complexes of the data processing center are involved in: "Modeling and control unit", which provides modeling, analytical processing of data on the quantitative and qualitative characteristics of scientific and technical products, the formation of control signals and pr. (Q _i - processes), "Block of resource supply", which controls the provision of the production process with resource components (F _i . - processes) and "Block of monitoring resources", which performs the function of monitoring and analytical assessment of the state of production readiness of resource components of performers. From the automated workplace of the system administrator, the distribution of available and planning of the required resources is controlled, including the stimulating and motivational components of the moral and psychological resource, ensuring the timely and high-quality execution of planned tasks.

In terms of hardware and software solutions, a single integrated information and control space is characterized by the following features that ensure high efficiency in the use of computing resources of existing services, including: a two-tier file storage system, which automatically transfers the most frequently used files to hot storage based on semiconductor non-volatile SSD storage; automatic and / or controlled allocation of the required amount of information resources and permanent storage devices via the WEB interface, depending on the load; displaying the current parametric characteristics of the state of virtual machines and the cloud as a whole; interface for access to information resources and application programming REST API based on application-level protocols of data transfer and text formats HTTP / JSON.

Thus, the proposed invention provides a high level of automation of production management processes and / or research and development activities of an enterprise, as well as the life cycle of scientific and technical products, based on cluster analysis and predictive modeling of an artificial neural network of adaptive resonance, simulation of production processes, computer network administration and project management in a single integrated information and control space and cloud computing environment, which ensures high efficiency of production management, reliability of monitoring and assessing the state of qualitative and quantitative indicators of products at all stages of the life cycle.

Claims (4)

An intelligent enterprise management system, including a hardware and software complex that ensures the functioning of a mathematical model based on the Cobb-Douglas production function in the formalism of Petri nets, built according to personnel, material and technical, infrastructural, geographic, technological, organizational, information, moral and psychological characteristics, connected by data transmission channels with a plurality of "man-machine" systems corresponding to a given functional purpose, characterized in that the mentioned hardware and software complex ensures the functioning of the dynamic mathematical model of the enterprise in a single cloud computing environment for the enterprise with a scalable architecture of network flows, clustering and a single software structured on the basis of an artificial neural network of adaptive resonance, which is controlled by the hardware and software complex an administrator who provides training of an artificial neural network and the distribution of enterprise resources, and is connected by data transmission channels with a variety of "man-machine" systems and equipment that records the state of objects and processes at the enterprise, Moreover, to build a mathematical model of an enterprise, in addition to the listed personnel, material and technical, infrastructural, geographical, technological, organizational, informational, moral and psychological characteristics, they use its financial characteristics, taking the moral and psychological characteristics as the main one, at the same time, the artificial neural network of adaptive resonance is made with the ability, in the event of inconsistencies in the observed state of objects and processes at the enterprise, of the current values of quantitative and qualitative indicators to the specified indicators, to generate signals with the corresponding control vectors that form control commands for performing corrective actions to maintain and / or restore compliance of the current values of indicators with the specified ones, while in the event of an emergency situation in the production process and changes in the values of quantitative or qualitative indicators that differ from the specified values above the set values, a control signal is generated and the corresponding information is displayed on the information display facilities, and the control vectors signal the need for prompt resolution of issues replenishment of the corresponding resource components and / or signal about the prompt solution of the problem of withdrawal from the production the process and disposal of components, the state of operability of which does not correspond to the conditions necessary for the fulfillment of the set production task.

Images