Software and hardware differentiation of information flows by areas of activities of a machine-building enterprise, as well as the temporary inconsistency of these flows, is one of the serious problems in managing such an enterprise. In this case, operational management is not linked to the strategic objectives of an enterprise, which generally affects its sustainable development. The construction and use of the enterprise information model serve as a tool for coordinating information flows. The study addresses topical issues of building an enterprise model based on the integrated approach to information management. Technologies formed a unified information environment are based on structuring and describing (modeling) various aspects of enterprise activities and creating its information architecture. Existing corporate information and management systems often have a limited scope that does not cover all enterprise aspects. This study aims to analyze the existing technologies for modeling machine-building enterprise activities and develop proposals for forming its information model. The authors analyzed the methodologies for creating an enterprise model as a combination of its technological, auxiliary, and business processes models. More than that, the authors compared software components written in the unified modeling language and object-process language. According to the results obtained, the authors proposed using the object-process methodology as the basis for creating an integrated or combined information model of a machine-building enterprise.
The study focuses on constructing an opportunity analysis of an engineering * Copyright © 2021 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). enterprise model based on existing software complexes. This type of activity began to develop in the 1980s. However, due to the complexity of elaboration and insufficient development of information technologies, practical implementation of such models has narrowed to certain areas (e. g., business process management or service management). There is a renewed interest in the development of holistic enterprise models and the use of such models in enterprise management using artificial intelligence.
Theoretical approaches in this direction have been well studied and quite developed. Practical implementation, especially in Russia, faces difficulties at the initial stage, namely, choosing a direction and a developed model. The proposed study will enhance knowledge about the purpose and application of enterprise models.
Enterprise modeling is an abstract representation, description, and definition of
the structure, processes, information, and resources of an organization. An
enterprise is a complex system, which description consists of a model variety that
describes the main characteristics of real objects, processes, and phenomena.
These models and their components can be presented in the form of electronic
documents, databases, and software files. An enterprise model is a representation
of what an enterprise intends to do, how it works, and, possibly, how it is
organized
E x ploitation
of Design and Production
indicators sign e D
Fig. 1. Processes in the production environment. Source: Compiled by the authors.
A model of any kind can be reduced to an information model, that is, a comprehensive standardized representation of objects, processes, properties, parameters, and links in the form of information data sets. An enterprise information model can contain mathematical and process models, storage, management, and data exchange models, model transformation rules, and model graphical representations.
An enterprise model can function only in the integrated environment with certain information architecture. The integrated approach used to create an enterprise model is based on the following principles: • • •
All information in a model is presented according to the rules of the semantic network; Simulations should accompany an enterprise life cycle; Quality control of the decisions is performed using simulation models.
Enterprise models are used as tools for the systematic description of enterprise
activities
Simulation models allow one to do the following: • • • • • • •
Form strategic development prospects;
Perform dynamic analysis of possible development scenarios;
Determine the impact of external and internal enterprise factors;
Determine the efficiency and effectiveness of enterprise processes.
Enterprise modeling includes the following aspects
The integrated environment combines information from multiple independent sources into a single logically sequential data set. Implementing the integrated environment and creating an enterprise information model are possible based on the integration, unification, and consolidation of these model components.
Integrated models are characterized by using a standard model form based on a common template for presenting information, for example, according to the Information Resource Dictionary System (IRDS).
A general structure template of the meta-model is used for unified models since it allows transforming private models and establishing the semantic equivalence of composite models. Unified models are used, for example, to model enterprise business processes.
A combined model is created in a situation when it is impossible to establish
the semantic equivalence of all private models. A certain degree of unification of
private models and their dynamic and terminological harmonization are used in
this case. A combined model is most likely for engineering enterprises in their
current state when most private models are not standardized or in the unified form
The basis of an enterprise information model is a conceptual model, that is, an informal model, in which formulation notions and representations of the subject area knowledge are used. In a broad sense, a conceptual model refers to a meaningful model based on a particular concept or viewpoint.
The conceptual model should be correlated with an enterprise strategy and solve its strategic tasks:
Enterprise models provide a detailed organizational context relevant and
necessary for strategic planning. <...> Enterprise models identify factors that
contribute to and hinder strategic changes in an organization at the detailed
level, particularly regarding the interaction of business models and information
systems. Thus, they provide a promising basis for the methodological support
of strategic decision-making processes
The consistency of systems and goals of an organization can be enhanced
through the use of interoperability models within and outside an organization.
Therefore, an enterprise that supports and uses an integrated set of management
models can be considered a model-driven organization (MDO)
The choice of a particular model type depends on the type of enterprise
activities, tasks it faces, automation of the existing information flows, and
resource restrictions. Although the relevance of enterprise modeling is undeniable,
a holistic concept of an enterprise model has not yet been developed. New
concepts of enterprise modeling are developed depending on the application area
and the underlying concept. For example, there is Model-Based Systems
Engineering (MBSE)
The main objective of this study is to analyze the existing methodologies for modeling enterprise activities and develop proposals for forming an enterprise information model of the machine-building profile. The solution to this problem is associated with choosing private quality indicators for the software systems and criterion forms of their suitability and optimality. 2
Analyzing the existing methodologies and corresponding software systems, which help implement them and build an enterprise conceptual model, is the main research method.
A functionality indicator of a software system is as follows:
K pmr = k1pr , k2pr , ..., k pr , m where kipr , i – 1, ..., m are private indicators of functionality.
The criteria of functionality include the criteria of suitability (G), optimality (O), and superiority (S).
The authors consider n software systems with m private criteria, where kij, i = 1, ..., m, j = 1, ..., and n is an indicator of i property of j system; K (mj ) =
k1 j , k2 j , ..., kmj {kiaj } is a permissible value set of kij indicator.
is a vector indicator of the functionality of j system; The suitability criterion is as follows:
m G : (kij ∈ {kia }) ⊕ U , j ∈ [1, ..., n],
(kij ∀l∈{l}m0 = klopt ) ⊕ U , j ∈ [1, ..., n], m where l is a number of properties to be optimized; m0 is a quantity of properties to be optimized; {l}m0 is a set of properties to be optimized; klopt is the optimal value of l property index.
The superiority criterion is as follows:
Fig. 2 shows an integrated enterprise information environment that includes a resource management environment.
A single space of a multidimensional model designed for synchronized data exchange and real-time dynamic enterprise management cannot be built on a single type of simulation system. There is a significant functional and information gap between these systems that is eliminated by various methods. Therefore, the solution to the problem of business process interaction requires using the PLM (Product Lifecycle Management) concept, which provides the possibility of integrating CAD/CAM/CAE/PDM systems into a single information space through structured data files or APIs (Application Programming Interface). Fig. 2. An integrated enterprise information environment. Source: Compiled by the authors.
In the development of enterprise models, structural, hierarchical, behavioral,
and other aspects of enterprise activities are considered. Through the structural
(architectural) aspect, an enterprise model is considered in the architecture form of
an information object. An information model describes enterprise architecture
using formal methods (graphs, matrices, lists, tables, and charts). The semantic
aspect allows defining and connecting the concepts used in the model to construct
elements and link maps
The hierarchical aspect allows defining hierarchical abstraction levels of an enterprise model. In this case, objects of the real world and abstraction can be ranked and arranged as a hierarchy of parts determining a model decomposition or a hierarchy of shapes representing the levels of generalization and specialization for building simulated entities.
The behavioral aspect concerns the consideration of functional relationships between system elements and the identification of variables. Behavioral description can be static and dynamic. In static behavior, the model is implemented as a series of object links. Dynamic description uses the information about dynamic characteristics and time dependency of object behavior, attributes, and relationships.
An integration service model, which ensures the functional consistency of information objects in heterogeneous environments, is one of the most important private models. In this case, the integration service definition should refer to standard definitions of information exchange protocols, such as STEP, XML, and EDI.
Private enterprise models include relevant elements of syntax, semantics, and
ontology. The model syntax is associated with allowed view relationships. The
model semantics extends to objects and links in accordance with a conceptual
model of an enterprise
The ontology includes a conceptual apparatus and a thesaurus developed on its
basis, and collection and processing information about the system and its
constituent objects
The modern approach to enterprise modeling is to consider an enterprise from
two points of view: (1) as an open set of harmonized business processes aimed to
achieve goals and solve the tasks of an enterprise and (2) as the integration of
functional entities (software and hardware, personnel, and technologies)
Enterprise architecture can be considered as an information base of assets that
determine the mission and strategy of an enterprise, the necessary technologies to
implement the strategy, and asset adapting processes and technologies for
changing needs of the mission
Developm ent Strategy fIr n o m a tS ito ra n tyge cheT n o l o g y Software ental Impact Environm reu ssse tc e u co trS rP iton sse iza isn n u rgaO dnaB Hardware
Enterprise t-uoA itanom Information
Model Information Information concerning architecture, the architecture description language, and the architecture structure. The combination of enterprise models at the conceptual, block, and element levels is a part of enterprise architecture.
Business Strategy Information Architecture and
Information Processes
An enterprise conceptual model is a compact statement of the business essence and its presentation by users and developers within the accepted ontology. The block-level includes reusable reference models based on language modeling designs. An enterprise model structuring at the block level is performed with consideration of various standpoints of the conceptual level.
Research focusing on the development of the enterprise information architecture has led to the creation of various methodologies, such as Zachman Framework, Structured Analysis and Design Technique (SADT), Enterprise Architecture Planning (EAP), Model Driven Architecture (MDA), Enterprise Wide Information Technology Architecture (EWITA), Global Returnable Asset Identifier (GRAI), Integrated Methodology (GRAI-GIM), GRAI Evolution Methodology (GEM), Federal Enterprise Architecture Framework (FEAF), Department of Defense Architecture Framework (DoDAF), Strategic Architecture Model (SAM), Extended Enterprise Architecture Framework (E2AF), Computer Integrated Manufacturing (CIM), Architecture of Integrated Information Systems (ARIS), Computer Integrated Manufacturing Open Systems Architecture (CIMOSA), Technical Architectural Framework Information Management (TAFIM), Technology Neutral Architecture (TNA), Purdue Enterprise Reference Architecture (PERA), Department of Defense Architecture Framework (DoDAF), The Open Group Architecture Framework (TOGAF), Generalized Enterprise Reference Architecture and Methodology (GERAM), Semantic Object Model (SOM), Dynamic Essential Modeling of Organizations (DEMO), and Method for Multi-perspective Enterprise Modeling (MEMO).
Many modeling methodologies are no longer supported. Limiting their ability to manage certain aspects of the simulated system is the main reason for the decrease in interest in these methodologies.
The enterprise information architecture concept developed to date is a holistic approach that integrates business and information technology capabilities. Most modern enterprise information architectures have four presentation levels: business, application, technology, and data architecture.
Enterprises widely use architecture description languages (ADLs), such as
AADL, ARIS, EADL, C2 SADL, ArchiMate, BPMN, SysML, to implement
architectures and models. The Unified Modeling Language (UML) is a
wellknown language for describing enterprise architecture and models
Functional models (by creating user-defined class diagrams); Object models reflecting the structure and hierarchy of system objects, attributes, operations, and relationships (by implementing object class diagrams); Dynamic models reflecting the system behavior (by using dynamics diagrams and process states).
The following requirements are imposed on enterprise models: Maintaining the possibility of joint analysis and information system and action system development taking into account the main points of view; Providing abstractions corresponding to the professional level of potential users and model representations in the diagram form; Comparability of model concepts with implementation level concepts; Ensuring the compliance of multilingual diagrams and the existing tool environment; Convenient and safe design of enterprise-specific models; Planning and ensuring the economic effect of enterprise modeling; Availability of enterprise reference models; Provision of appropriate modeling tools; Availability of convenient specification and enterprise model adaptation. The authors consider some standard enterprise architectures.
TOGAF
CIMOSA
GERAM
ARIS
SOM
DEMO (Dietz, 2006) focuses on the language of interaction and individual
action. DEMO, built on organizational semiotics, philosophical ontology, and a
communicative action theory, provides complex concepts for analyzing
communication. Different levels of DEMO abstraction allow the reuse of
enterprise models. Unlike other methodologies, DEMO does not take a top-down
approach to enterprise analysis. It focuses on individual actions, interaction
models, and the role of the language
The comparative assessment of developing methods for enterprise information architecture shows the presence of common properties, for example, high-level structures for various points of view or languages and modeling methods. Nevertheless, since DSML and its methods of use are specific to each approach, there are clear differences between methods, and there is no common metamodeling language that supports DSML integration.
Enterprise modeling languages not only support the presentation and analysis
of business process models. Some of them, such as BPMN, simulate workflow
diagrams. However, business process models cover only the narrow focus. The
development of information systems that correlate with business and consider
strategic options requires the inclusion of more enterprise aspects
Under the creation of an enterprise model, modeling languages are used since
they allow the reconstruction of language concepts typical for different points of
view of stakeholders. These languages include General Purpose Modeling
Languages (GPML), such as UML
Different approaches to enterprise conceptual modeling as a tool that helps decrease the model complexity aim primarily to design software systems but not to create an enterprise information model that includes all aspects of its activities.
In this regard, the method for multi-perspective (cognitive) enterprise
modeling MEMO and the corresponding meta-modeling environment
Special attention is paid to the analysis of enterprise prospects in this method. MEMO language architecture meets the requirements of adaptability, extensibility, and integration. It consists of the meta-modeling language MEMO MML and the specific DSML. MEMO is characterized by complex language specifications and method engineering support.
MEMO covers the engineering, management, economic, and social aspects of enterprise activities and illustrates current and potential scenarios. An enterprise model includes object and component models integrated with conceptual, strategy, or business process models.
MEMO architecture is based on developing the concept of multidimensional
corporate models and extensible language architecture. The appropriate modeling
environment integrates editors for different DSML into multilingual model
editors. MEMO includes a meta-data editor that enables the user-friendly
development, expansion, and use of the supported DSML set and upholds the
creation of appropriate graphics model editors
A set of modeling methods and languages used is not always sufficient due to the variety of problems encountered in enterprise modeling. MEMO introduced support for customizing modeling methods, which allows creating fundamentally new enterprise models to cover a broader range of issues.
The object process methodology (OPM) represents a conceptual approach,
language, and modeling methodology applied in industrial enterprises
OPM uses two types of elements: entities and links. Entities are objects, states, and processes. The links used in OPM can be structural and procedural. Structural links form relationships between two objects and are constant. Procedural links depend on the state of objects and processes.
The structural ОРМ-model of an enterprise is a set of elements connected by
structural relationships. The dynamic ОРМ-model reflects mechanisms of element
transformation
The object can be in one of its states or during the transition between them. The process converts objects by creating, using, or changing their state. The processes complement objects by creating a dynamic behavioral aspect of a system. At the same time, OPM does not explicitly define a process state model.
With OPD, it is possible to build a system diagram (SD) of the upper level that provides the context of system target functions. When using OPM, objects with a time-constant state and processes that transform them form the universal minimum ontology.
In order to define DSML, OPM has developed a visual meta-modeling language using a static structural meta-model of the language and a set of rules for checking it. This approach was used to determine a subset of the modeling language using the Eclipse platform and technologies. Thus, OPM allows integrating widespread logical DSML, developed, for example, using EXPRESS.
EXPRESS is a language of information requirement specifications for data structuring, which provides the machine-oriented representation of product data and its exchange. EXPRESS schema is the basis for determining and structuring the interrelation of elements presenting product data.
Along with the widespread EXPRESS language, there are other languages for engineering information modeling designed to solve industrial automation problems, for example, AutomationML. This XML-based language is designed to support the engineering instrumentation data exchange in equipment projection and process management.
Typical object information models include information about the topology, geometry, kinematics, and logic. Simultaneously, logic considers process sequence and order, and behavioral and management aspects of objects. 3
Based on the results of considering concepts and modeling language capabilities for building enterprise information models, the authors have found that MEMO and OPM have the most functional potential.
There are many essential differences between these concepts. For example, MEMO offers its modeling language and its formal model for each service. In OPM, the formalization level is higher, which reduces development costs and increases the reliability of systems and processes. In MEMO, the relationship between objects is presented as a sequence of object actions, while in OPM, it is presented as a sequence of object states that also increases reliability.
The use of the environment for presenting graphical programs, followed by automatic generation of the object code, is the most common component in the methodology description of enterprise information architecture. The authors have compared OPL and UML used as graphical modeling tools.
Each of these languages uses an automatic object code generator to improve performance and program development speed. A way to compare modeling languages was to study the possibility of generating an object code. For this, the authors have used freely available packages of software systems StarUML v1 with a code generator from PostgreSQL and OPCAT with an OPM-GCG code generator. Output results were obtained in XML codes.
The authors have considered graphical representations of UML and OPL programs for providing a milling workstation with a tool as an example and generated object codes. When comparing the UML results, the authors have found the object code incompleteness that fully reveals the structure of objects and only partially their behavior. Some redundancy of the generated code was also detected. The code size generated in OPCAT was 1.3 times less than the code size produced in StarUML.
Various DSML can be used as languages for describing physical objects and processes, including EXPRESS as the most common in engineering enterprises and AutomationML. The latter allows direct interaction with universal languages.
The importance of OPM is gradually increasing, as evidenced by the
publication of a preliminary standard PNST 173-2016
Implementing a unified enterprise information model is often impossible because it involves too many resources and risks. Nevertheless, it is necessary to develop approaches that determine the evolutionary implementation of an integrated or combined enterprise model at each enterprise.
The construction of a complex information management system for industrial
enterprises based on methodologies not discussed in this paper, such as
serviceoriented architecture (SOA), service-oriented modeling and architecture (SOMA),
and Internet of Things
Systems-oriented design methods do not consider interconnection and
collaboration issues. OPM is also intended for integrated modeling and design of
different complex dynamic systems and the provision of an interface infrastructure
and relationships at the different detail levels. It facilitates a smooth transition
from a set of different system-oriented views to a consolidated, integrated model
that considers the integration aspects, structure and behavior of the interface and
payload, interconnection processes and services, and ultimately new
interoperability capabilities
Based on the above, the authors have concluded that OPM semantics is more adapted to the system design than MEMO and other enterprise modeling systems since OPM allows modeling information, hardware, people, and management and creating a complete enterprise information model at the conceptual level.
Although the analysis volume does not prevent fully imagining the advantages
and disadvantages of the considered software systems, the authors have assumed
that there is no evident superiority between them in most characteristics. However,
the authors have also concluded that OPM surpasses other modeling systems in
two properties (reliability and program execution speed). In addition, the
possibility of using one methodology for developing software, hardware, and
information system in general, using one design concept, principles, and
environment is an important step towards integrating and optimizing the design of
new systems
The analysis of software tools has shown that there are multidirectional trends in approaches to enterprise model development. The choice of the most promising directions in the sphere of machine-building enterprise modeling is related to the automation level of these enterprises and the presence of certain managing, transmitting, and information processing means, that is, existing restrictions, including financial ones.
Machine-building enterprises in Russia need not separate local management systems but indivisible management systems that consider all enterprise activities and provide the possibility of forecasting, risk management, and optimization. Simultaneously, the development and implementation of information management systems with modeling functions remain a serious problem for many organizations.
Product and enterprise lifecycle management, based on MBSE (Model-Based
System Engineering)