The authors proposed a new method for automating the of complex technical systems re-design based on workflows allowing to reuse design solutions, modify them to meet the conditions, using the “Reuse” concept. The method differs from the existing ones with project workflows ontology in the automation to design complex technical systems.
The key problem of couputer-aided system (CAS) design and development is to create a successful project. According to the up-to-date Standish Group data, only 40% of software projects are successful (i.e., the projects completed in time and budget, with all the specified features and functions). And the especially significant role in achievement of the CAS development’s success is given to the diagrammatic models in visual forms of business process artifacts, particularly at the concept phase of CAS design. For this purpose, the visual languages (UML, IDEF, ER, DFD, eEPC, SDL, BPMN, etc.) were developed, and they are widely used in practice. Such models’ use sufficiently increases an effectiveness of the design process and a quality of the design solutions, through the unification of interaction language of the CAS development participants, the strict documentation of the project-architectural, functional solutions, and the formal control of diagram notation correctness.
In recent years the large industrial companies and enterprises actively use
distributed dynamic workflows of designing and manufacturing activities. For
example, according to [
However, in theory and practice of corporate use of diagrams there are no effective methods and tools for monitoring diagram representations of dynamic distributed workflows of CAS, that results in the serious design errors. Thus, the analysis, monitoring and processing of distributed dynamic workflows in CASs’ design and operation, presented via their diagram, is an important scientific and technical task.
With each new stage in the technology development, the complexity and enhancement of designed products and their components grows up. New production mastering and corresponding release of changes leads to the increasing of information volume. Such information can be both structured (3D models, electric schemes, etc.) as well as unstructured (office and project documentation in DOC, XML and other formats). Last appears at different stages of the engineering process and is stored in separate repositories in electronic or paper form, often without attaching to a specific project. These circumstances complicate the right data searching task. Thus, products data organization and updating in the formal description problem exists. It is worth pointing out that Entity-Relation (ER) data presentation model is frequently used in the information systems of large enterprises for the formal description of the product.
Exploring the scope software and hardware computer-aided design shows that
upto-date researchers are mostly focused on design methods and their components
improvement. The progress has been made in the field of mathematical modeling,
engineering calculations, optimized data structures and user interfaces [
The dynamic design processes development has become one of the latest trends in
the science of business process management (BPM). Researchers and practitioners are
improving the tools, methods and theory of flexible design processes [
[
Control, analysis, synthesis, transformation and interpretation of design workflows are engaged in scientific schools of HSE, MSTU STANKIN, N. Bauman MSTU, SPb Department of Steklov Mathematical Institute of RAS, Computational Mathematics and Cybernetics, Lomonosov Moscow State University, Institute for System Programming of RAS (Russia), Carnegie Mellon (USA), VERIMAG laboratory (France), as well as scientists like Afanasiev A. N., Karpov Yu., Sosnin P. I., Lifshitz J. M., Yarushkina N. G., Kalyanov G. N., Konev B. Yu., Shalyto A. A., Savenkov K. O., Kulyamin V. V., Okhotin A. S., Mikheev A. G. (Russia), as well as Neda Saeedloei, Gopal Gupta, Clarke E. M., Booch G. (USA), Yuan Wang, Yushun Fan (China), Van der Aalst.
Gavrilova [
The complex technical products design is a complex workflow of various specialists of the design bureau (DB). Each specialists group performs part of the product design work, going through the implementation chain of a variety of the business process tasks design preparation. Each task can be performed using a large amount of specialized computer-aided design (CAD) systems, so the complex technical product design documentation (DD) is a set of design solutions of independent CAD, which must be collected in a single design solutions database. Product data management (PDM) system, as a rule, acts as a unified design solutions information base. The complex technical product design process is presented in the Figure 1 in the form of conceptual model.
Each workflow task in the conceptual model is an embedded implementation procedure, which involves not only the subdivisions of the design bureau, but also the various departments of the enterprise participating in the coordination process. The result of each nested procedure is an agreed electronic document (ED) in text format or in the CAD specialized format. In some cases, for example, in the “Technical task development (TT)” task, the execution result is presented as a created PDM system object, in which necessary attributes are filled. The same technical task in paper form is generated from the system as a report. It should be noted that the design documentation and information about the product design in the PDM-system does not fall once, but appears as the design at each stage of the complex technical product design workflow. Ontology-based product presentation semantic model development
One of the promising areas of the product formal description is the ontology-based
semantic model design of the engineering design components. The main purpose of
such model is to systematize data about a product, create a conceptual diagram of a
product in the ontology form [
The ontology-based structural design organizational and technical components semantic model includes a structure for describing information and rules for its interaction. Conceptual diagram of product representation using an ontology-based semantic model is presented in Figure 2.
Each designed product is represented in the semantic model as a “DAC” concept, denoting a detail or assembly component (DAC) and having properties: type, class, designation, name, state. It should be noted that the product is a complex organizational structure, which has several hierarchical levels of nesting, represented in the form of a tree structure, therefore, the composition of the product is presented as a set of concepts “DAC”, having a reflexive relationship “Consists of ...”. The “DAC” concept is connected with the “Documentation” one by the “Contains” relationship. Each “DAC” concept, as a rule, contains several “Documentation” concepts. The “Documentation” concept has properties: DD type, state, approval signatures. Since any product and its components are developed on the basis of a technical task (TT), the “Technical task” component is introduced into the semantic model, which is associated with the “DAC” and “Documentation” concepts by the “Based” relationship. The “Technical task” concept has the following properties: customer, work description, execution period, state. Within the developed technical specification framework, the technical requirements for the designed product are defined, which in the semantic model are represented by the “Technical requirements” concept with the properties: requirement type and value. According to the design process results based on the technical task and technical requirements set in it, the “DAC” concept has technical characteristics presented in the semantic model as the “Specifications” concept, which has properties: specifications type, value, measure unit. The “Technical task”, “Documentation” concepts have a direct developer, which is introduced into the semantic model as the “Developer” concept and has the properties: developer full name, position, division, design date. The complex technical product design cannot be accomplished without detailed elaboration of composite components and drawing up design work schedules. The introduced “Operating schedule” concept is associated with the “DAC” one and allows to create the entire works list decomposition. The “Operating schedule” concept contains the properties: work description, work executor, the planned work commencement date, the planned work completion date. The “Operating schedule” concept is also associated with the “Developer” one.
n l t i t e a n n h ic e n tm u t
n m io r t c g e o e A t d d e
h D c m m g n i r u t c a f u n a n i s e t s y s l a e t x e l p o c g n i t a e r c r o f s o l f k r o n g i s e w w D . 1 e r u g i F , s s k e s n
i a l t
d f a o e e d c l n o
r a t u n s o
Is c m e t s y s t n e e g a n a m m t c e j o r P , s s k e tsa li
n fo ade e d c l
o n r a t u n s o Is c , e r d o e tr u l d p ro p o A n io l t a a v in ro c n ro o p c l g a n v i o m r it p p A s e f n o il e , d c s a n k e a s d u ta l s ro s t , I n s s o k e c tsa li n d fo ae n g i s e
D e d c l
o g an tr n u n li s o u Is c d e h c
S , s s k e sa li
n t fo ade , e d s s c l
o k e n r sa lin a t t u n fo ade Iss co e d c l
o n r a t u n s o Is c w o l f k r o W re -3D ign s sap eD 3E oKm litum A g h o c l r a a n e
A S eD fdo t n g s enm iseg -s3D mren iren iton lpo ftd aopm lituA isegD ine lau ve ra K g lc n a
E c m T p T lo f e o v t e n
D e n s o t i t n a e in f m
o e m r r i e u t
q e
e D r irng son -3D iregn e it sa se sy abL i opKmlituDm snA tah
a ne lu g lc M n a E c A y r ina ign ilmse re d P re -3D ign sa se sy abL opKmlituDmsnA tah
M A
D C E s n o it a l u c l a C s n o it a l u c l a C T T E s lca tne i n m ch ire e u T q e r e l u d e h c S s r led tee oMraam
P l is e s d ly o a M n
A 4 The library of workflows
The autohor’s ontology-based method is presented in Figure 3. The semantic model formation takes place using the data extraction method. It is presented in different formats from documents to PDM system objects. The semantic model formation is launched at each stage of the design process through a project management system and requires effective algorithms.
The initial task of the workflow for designing a complex technical product is “Determining Requirements”, within which technical requirements for the entire product as a whole are defined. At this stage, the document containing technical requirements appears in the PDM system, mostly in text format. After that the product semantic model forming mechanism is launched, during which the “DAC” and “Technical requirement” concepts are created using the data extraction method. Filling in the properties of the “Technical Requirement” concept occurs by selecting the “Type of Requirement” and “Value” properties from the document text. According to the general product requirements, a technical task for the development of a draft design or an engineering documentation is being developed, which is carried out in the next stage of product design. The technical task in PLM-system is represented as the object with the same name, which has the following attributes: “Customer”, “Work Assignment”, “Technical Requirements”, “Contractor Subdivision”, “Design Date”. The data extraction method allows to select several concepts from the PLM-system object: • “Technical Task” concept, with the filling of properties: execution period, work description, customer, state. • “Developer” concept, which has the properties of a division, position, full name, design date. • “Technical Requirement” concept with the filling of the corresponding properties.
The result of the “Draft design development” or “Engineering documentation design” task is a design documentation released, which is represented in the PLM system as “Document” object with attached project files. The data extraction method highlights the concepts: • “Document” concept in which the properties are filled: document type, state and approval signatures. • “Specifications” concept is obtained from document files, for example, 3Dmodels, with selection of the “Type of specifications” property and its value in a certain measure unit. • “Developer” concept as described earlier.
Opera ting schedule s a h tI Techni cal requirem ent s e n if e
D technical task
The “Engineering Calculations” stage execution leads to a number of documents development containing calculated values reflecting the DACs characteristics. Using calculations, the main characteristics of the designed product are highlighted, which are compared with the “Technical Requirements” for compliance. The result is a message to the user about the (non-)compliance of the developed DD with the stated requirements, indicating the difference in deviation. Completion of the “Draft design development” stage leads to the start of work on the decomposition of the product into its component parts, planning the stages and development terms. As a result, the “Operating schedule” concept is filled out with the “Work Description”, “Planned start date”, “Planned completion date” properties and the work performer is recorded with the filling the corresponding property. The development timing specified in the work schedule is affected by the presence or absence of the product component parts analogues. For this purpose, the design solutions ontological base is applied for carrying out such an analysis.
In this paper, authors investigated the design workflow of complex technical products and developed noval method to create a database of these workflows. To systematize data about the product and its organizational and technical components, an ontolodgy-based product representation semantic model is presented. Authors have offered method allowing structural-parametric analysis of products in the PLMsystem according to the necessary requirements. The product data extraction method from the diagram ER-model of the PLM-system to form the design solutions ontological database is developed, which will significantly reduce the time to search for ready-made similar solutions. A verification method of the developed product for compliance with the requirements is proposed, which allows to analyze the product semantic model and improves quality of the designed products by eliminating errors in the early design stages. Efficiency is confirmed by the relevant experiments, and authors revealed the superiority in comparison with the existing PLM-systems. The future direction is ontology-based semantic model creation of production design technological preparation that will allow to understand an essence of business processes of the enterprises and will provide decisions reuse.
Acknowledgment
The reported study was funded by RFBR according to the research project № 1707-01417. The reported research was funded by Russian Foundation for Basic Research and the government of the region of the Russian Federation, grant № 18-47730032. The research is supported by a grant from the Ministry of Education and Science of the Russian Federation, project No. 2.1615.2017/4.6.