Shulou(Shulou.com)06/02 Report--

Abstract: this paper introduces the historical background and definition of product data management technology, analyzes the characteristics of product data, and introduces three ways to realize product data management. This paper focuses on the data management architecture of product design process based on concurrent engineering. Finally, the development prospect of product management technology is prospected.

Overview

With the rapid development of computer technology, automatic control technology and network communication technology, fundamental changes have taken place in modern manufacturing technology and production management mode. The manufacturing mode with CAD/CAF/CAM integration technology as the core (referred to as VC3 technology) is rapidly becoming the mainstream mode of product development and production. The whole development process people begin to change the serial process into parallel mode. Dynamic optimization is used to deal with the problems in the process of product development, so that product developers can take into account all the factors in the whole product life cycle from conceptual design to product scrapping. including quality, cost, job scheduling and user needs.

Therefore, parallel operation has become an important feature of advanced product development, and all kinds of engineers and technicians can carry out their own related product development under the same product model. In order to complete the above process, it is necessary to realize information sharing, which is the basis of enterprise production process automation, which is the inevitable technical background of product data management technology.

1 data characteristics in the process of product development

The product development process can be divided into small stages, such as planning, design, calculation, production and processing, test assembly and so on. In these processes to produce a large amount of data information, because each stage requires to solve different tasks, the data information is also different, but the whole is related to each other, it often takes a long time to calculate to get the final results.

In addition, a design process is often a process of repeated exploration and adjustment, and the operation of the designer is mainly a real-time interactive processing, which requires high response speed and user interface. Therefore, the product data has not only static standard data, but also dynamic construction data, information growth data, experimental temporary data and so on, which are modified with the development of the design process. The data structure is complex and diverse, and the recording data items often change according to the product object.

To sum up, product data has the following characteristics:

The main contents are as follows: (1) there are many types of data and a large amount of data: there are mainly two types of product data. First, design environment data, mainly design environment rules, design methods, standard elements, design conditions and requirements, etc.; second, design object data, mainly design product definition, design constraint control, product performance, production requirements, economic accounting, design scheme optimization and so on. In the two types of data, there are basic data types, as well as unstructured variable length data, such as variable length strings, graphics, images, motion video, sound information and so on. in order to express the product information vividly and realistically.

And because of the inherent characteristics of the systematic and long process of the product object itself, the amount of product data is huge and complex, even the more standard data, such as design specification data, standard tolerances, structural elements, materials and parts, technical specifications and so on, the amount of data is also very large, these objective factors require the product database to have large capacity, multi-variety data storage and fast processing ability.

(2) data dynamic: product data generally has static data part, that is, information prepared in advance to guide actual information, such as design specifications, etc., but more importantly, dynamic data, that is, data generated dynamically with the development of design behavior in the process of product design, and directly related to the final results, such as structural analysis and calculation, mechanical model, part graphics, etc. In addition, process rules, scheme burying and technical document processing are added, so the structural pattern used to store product data also has the nature of dynamic construction, which can not be completed until the final result of the product object is produced.

(3) the data structure is complex: while the product data forms and develops dynamically with the development of the process, the relationship between styles is complex and diverse, showing a network structure and divided into layers; hierarchical design not only has a hierarchical structure, but also has different levels for different design objects, sometimes it may have complex structural characteristics in each node. On the other hand, the structural relationship between the representation and access of knowledge and rule data in the product shows a more complicated scene, especially the fuzziness of description exists in many factual knowledge (such as design and production cases, experience, etc.). Individual design objects or management objects have complex logical connections and uncertain reasoning. It is very difficult to define and manipulate this kind of non-standard data. Therefore, the product data structure is complex.

(4) procedure access and random access: product data is mainly process access, so transaction management is also oriented to long transactions. Product design usually lasts for a long time, the data is not only a simple query or a large number of extracted data, the extracted data often have to participate in the design process, but also use a variety of relevant information in the design process. And the new results (including modifications to existing design data) should be submitted as transactions and stored.

At the same time, because product design is a tentative process, it is repetitive, tentative and developmental, such as some designs are modifications of classical examples, while some designs may have a variety of options, this requires the ability to retain and manage the history of design, different design schemes and dynamically changing patterns, and so on. In addition, there are requirements for version control in the mode of operation of product data. Therefore, there are overall data, professional data, vertical related data and horizontal related data in product design. In an integrated system, the amount of random access data is very large.

2 the concept of product data management

PDM (Product Data Management) technology first appeared in the early 1980s to solve the management problems of a large number of engineering drawing documents. In the past, engineers used to dive into the data heap and spend a lot of time looking for information that should have been readily available. In order to solve the management problem of a large number of engineering drawings, the engineering drawings are converted into electronic images by using graphic scanning technology, and these electronic images are browsed and modified by software to generate new engineering drawings. This kind of software is the embryonic form of TI PDM.

With the development of document storage technology, the object type of software management is gradually broadened, and it begins to deal with a large number of heterogeneous documents (such as CAD graphics, images, word processing files, etc.). Combined with the reality of enterprises, it begins to simulate the frequent and delayed "change order" management process in enterprises, which is gradually extended to the field of product development process. Such as design drawings and electronic document management, automation project change order management, bill of materials (Bill o f Material,BOM) management and so on. In recent years, PDM has developed rapidly, and it supports a wide range of applications.

In the process of product development integration, PDM is an indispensable software supporting environment, and it is the bridge and link to connect people, tools, information and processes in the integrated environment. In February 1995, ED Miller, president of CIM Data, which specializes in PDM and CIM related technical consulting business, gave the definition of PDM in the article "PDM Today": PDM is the technology for managing all product-related information and processes; all product-related information refers to all kinds of information describing the product, including parts information, structural configuration, documents, CAD files, approval information, etc. All processes related to the product, that is, the definition and management of these processes, including the approval and dissemination of information. This definition means that Qingrou Yingtai TI PDM will have a broad application prospect in industry.

In September 1995, D. of Gartner Group Company. In his "CIM Strategy Analysis report", Burdick defines PDM as: PDM is a key enabling technology for enterprises to design and produce a concurrent product art environment (composed of supply, engineering design, manufacturing, procurement, sales and marketing, and customers). A mature PDM system enables all people involved in creating, communicating and maintaining design intentions to share and transmit all heterogeneous data related to the product freely throughout the information life cycle.

This definition accurately shows the relationship between PDM and concurrent engineering. Here, "heterogeneous data" means that as the knowledge structure involved in product design is becoming more and more complex, and experts in various fields participate in the design with the help of different hardware, different operating systems and different application software tools, product information is often generated or used by a variety of application software systems provided by different software vendors. This will result in different types of design data And because the people involved in the design may belong to different departments or even different regions, the dispersion of design activities makes the information stored in different departments or locations. " Product-related process "refers to the product development process, product change process and other workflows."

3 structure of PDM

In the physical structure, the traditional PDM adopts hierarchical structure. Before the 1980s, single-layer structure was widely used. The typical feature of single-layer structure is GUI (Graphic User Interface, graphical user interface). Processing logic and data storage are included in the system as a whole.

A two-tier structure is popular in the mid-1990s, which divides the application into two layers: the client layer and the server tier. The user interface and processing logic reside on the client side, while all the related data is stored on the server side. Its biggest feature is the simplification of the client, the general client browser is only a tool to display the interface, and the main business logic is all processed in the Web server, this feature makes the installation and maintenance of the system particularly convenient, only need to change the server side, you can achieve the update and maintenance of the whole system.

However, Web itself does not provide a good development mode, it is only a running mode, according to the current common Web development mode, that is, writing CGI (Common Gateway Interface Common Gateway Interface) programs in C, it is still difficult to develop. And because the business logic programs run on the server side, when the scope of application becomes more and more complex, the load of the Web server will become heavier and heavier. Therefore, it is necessary to develop distributed Web programs to distribute the computing load.

At the beginning of 1996, the emergence of three-tier or multi-tier structure has fundamentally changed the application of computer networks. this open, distributed object method divides the client's huge program into two parts: user interface and logical processing. This three-tier structure responds to and processes user requests faster.

With the gradual maturity of distributed object technology, multi-tier distributed application architecture has been used more and more. Only when the application system changes to multi-tier distributed system, can the problems existing in Client / Server structure be solved finally. Under the multi-tier architecture, applications can be distributed on different system platforms, and the object communication between heterogeneous platforms can be realized through distributed technology, and the application system can be integrated on the distributed system, which can greatly improve the scalability of the system.

The emergence of CORBA (General object request broker Architecture, Common Object Request Broker Architecture) solves the problem of distributed computing, which enables distributed applications to interoperate with each other. The ORB (Object Request Broker) mechanism provided by CORBA enables transparent communication and access between objects. As a middleware to connect client-side and server-side objects, ORB provides an effective way for distributed interoperation between product heterogeneous platforms and heterogeneous systems. Different ORB can communicate through IIO P protocol.

As long as the interface of any object is described by IDL language (Interface Definition Language), other systems that follow CORBA specification can be accessed through ORB without considering the physical location, running platform and concrete implementation of these systems. This realizes the transparency of access and interoperation, and solves a series of problems caused by the heterogeneous environment of product data distribution.

PDM technology is a general extension of engineering data management (Engineering Data Management, EDM), document management (D ocu ment Management, DM), product information management (Product Informa-tion Management, PIM), technical data management (Technology Data Management, TDM), technical information management (Technology Informant Management, TIM), image management (Image Management, IM) and other product definition information management technologies. The new generation PDM system integrates object-oriented technology, network communication technology, client / server structure, distributed database technology and graphical user interface. Many enterprises regard PDM as the integrated framework that runs through the whole enterprise, and it is a strategic decision for enterprises to maintain competitiveness.

As an integrated framework in the field of engineering, TI PDM has the following functions:

① supports heterogeneous computer operating systems, databases and networks

② adopts object-oriented mechanism, provides object-oriented modeling method, and has good system openness and expansibility.

③ has a convenient integration plug-in function, which supports the integration of heterogeneous application systems and PDM, and realizes transparent invocation of tools.

④ provides serial and parallel process management for the whole product life cycle.

⑤ provides unified management of distributed data and supports information sharing.

Figure 1 shows the architecture of the product concurrent design support system based on PDM (the thick wireframe part of the figure). Among them, PDM framework is a network-oriented design and development team, which is composed of control layer, application layer, application support layer and basic environment layer.

(1) Control layer: located at the upper layer of the system, it defines the process model of product concurrent development. Data management in concurrent engineering revolves around the process. The process determines how participants (Team) are organized, their responsibilities and authority, the allocation of equipment resources, the use of tools, and who should submit data to where and when. Therefore, when determining the architecture of the product data management framework that supports concurrent engineering, the process is regarded as the top control, which defines the executable design process through the corresponding software operation interface to realize the process optimization.

(2) Application layer: it is the generator and receiver of data. It faces different application fields and accomplishes different tasks (such as CAx, DFx, etc.). It can encapsulate the object in the management framework through the corresponding software application development interface, and realize the encapsulation or integration of application tools and PDM.

(3) Application support layer: it is the core layer of data management. on the one hand, it provides user operation interface to realize the definition and operation of personnel organization, product structure and process management, and on the other hand, it also provides application development interface. so that users can modify and supplement the product data model to the encapsulated objects, and realize the customization tasks such as product structure management and application tool encapsulation. All industrial standard models and customized models are managed uniformly in a graphical object browser by using object-oriented technology, which shields the network, operating system and database. Users can manipulate the data objects in the whole environment intuitively and conveniently.

(4) basic environment layer: provide support for the above two layers, in which the database mainly manages the metadata (meta data), and the network is mainly the client / N. The architecture of server (C / S) provides support.

In order to ensure the smooth flow, consistency and sharing of upstream and downstream information in the product life cycle, on the one hand, the PDM integration framework needs to manage and control the product information generated in the process of product development. On the other hand, it is necessary to realize the integration with different application systems, which includes two levels: one is the encapsulation and integration of PDM system and CAx / DFx tools, and the other is the information exchange between design system based on PDM system and production plan management system based on EPR. On this basis, it provides a collaborative working environment for each member of IPT to ensure that they cooperate with each other in a different environment, exchange information in a timely manner, and speed up the product development process.

4 conclusion

Combined with the above concurrent engineering architecture, a collaborative work environment supporting multi-functional concurrent development is established in a certain software environment, the product life cycle is fully digitally defined, and the complete computerized management of information is realized; define a team organization model with optimized structure, establish the authority of information sharing operation at different levels, provide hierarchical interfaces with different application tools, and effectively manage tool resource information. On the basis of optimizing the enterprise process model, the enterprise process model is transformed into a controllable workflow to realize the concurrent design of the product. This is the main content of product data management.

At present, the more advanced development stage of PDM technology is called Product Lifecycle Management PLM (Product Lifecycle Management). It is a technology to manage products, processes and resources involved in all aspects of products from user requirements, product design, manufacturing and assembly, maintenance services to scrap recycling. If PDM is focused on the product design phase, PLM greatly expands the scope of PDM. This article is from Qingsoft Yingtai PDM. For more details, please click: www.plmpdm.cn

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.