Computers and Industrial Engineering Vol. 25, Nos 1-4, pp. 321-324, 1993 Printed in Great Britain. All rights reserved 0360-8352/9356.00+0.00 Copyright © 1993 Pergamon Press Ltd INTiTJ J JGENT CONCURRENT ENGINI:~RING ENVIRONMI~NT Ganesh M. Krishnaswamy, Ahmad K. Elshennawy Computer Integrated Manufacturing (CIM) Laboratory Department of Industrial Engineering & Management Systems University of Central Florida Orlando, FL 32816. ABSTRACT Thoughseveral are the in-roads that have been made towards achievingdesign-for-manufacturability(DFM)and design-forassembly (DFA), speaking of some of the crucial issues associated with concurrent engineering, yet very few comprehensiveblueprints exist that speak of a near-complete concurrent engineering environment. This could mainly be attributed to the multifarious versions of how to successfullysetup a concurrent engineering environment for today's ever changing engineering scenario. An operational concurrent engineering environmentneeds to be addressed in its totalityor near-totality. This paper is one such venture. INTRODUCTION Greater engineering productivity has brought with it more design alternatives and more analyses. Design complexities lengthen the lead time to manufacture, making it more difficult to anticipate the eventual impact of design decisions, place a greater burden on engineering talent, and heighten the need for better design tools. The competitive pressures for more efficient design and manufacturing methods are also strong. In particular, automating access to design and manufacturing data for teams of engineers and administrators has become an important technical priority. To meet these challenges, manufacturers are turning to concurrent engineering. They are forsaking sequential methods of product development and adopting methods that integrate product design and the product development process. Such a compact relationship, spoken of as above, envisages the need to implement automatic data flows between design and manufacturing. This is so central to implementing concurrent engineering that it is rapidly coming to be regarded as one of its key technical enablers. In this era of electronics and computers, it is but natural to employ these hi-tech tools to continually improve time-to-market, quality, and costs; increasingly this means building an automated concurrent engineering environment. Extrapolating this a step further, it could be fabricated to incorporate some intelligence, the later part of this proposal being devoted towards illuminating the same. Past work in concurrent engineering has meant dealing with one or more aspects of the science of concurrent engineering, and never integrating all these into a single coherent process. This could mainly be attributed to the multifarious versions of how to successfully setup a concurrent environment suited for today's ever changing engineering scenario. PROPOSED W O R K Lets consider the implications of considering manufacturing after design. The product development team of a company, consisting of personnel from market research through marketing, works on all aspects of the product, always with an aim to give the customers what they want, i.e. good product quality. Since design has always been an iterative process, at least theoretically there never reaches such a stage when the ideal product has been designed. This goes to say that there exast numerous stages ot product design, the termination at any of which could lead to a different product. Also, when customers feel a need for a product, they almost immediately and invariable they associate such a need with a physical object, which to them is the answer to their wanting. Nothing gives the customer more satisfaction than a product which will satisfy his/her need(s), as well as conforms to the customer's visualization of the same. Aligning once again with the product development team, manufacturing is such a process which tries to machine a workpiece with as close a conformance as possible, to an idea or a drawing. There never arise questions associated with physical product variations arising due to different machining techniques or different machining equipment, unlike the process of design. To give the customer utmost satisfaction and to clear the nebula preventing an unambiguous understanding of whether it is possible to generalize the process of concurrent engineering immaterial of the company involved, the route from design to manufacturing could be reversed and experimented with. In this line of reasoning could lie the answer to a much vaunted question. The ultimate goal of this venture shall be the creation and experimentation of a fully operational and generic concurrent engineering environment, as an answer to the numerous problems confronting the dynamic industrial fabric of today. 321 322 Proceedings of the 15th Annual Conference on Computers and Industrial Engineering For this purpose, the entire concurrent engineering environment will be thought of as depicted in Figure 1. ~ , S S EARC H u~ S Figure, i: SUBC04~: 14Nx" CE E n v i r o n m e n t USER INTERFACE To address the precise & complete understanding of the customer needs, the traditional Quality Function Deployraent'8 tool, the House o£ Quality which addresses the voice of the customer, will be augmented by computer graphics tools such as virtual reality tool kits dealing with the customer's vision of his product. Computer graphics technology enables us to create a remarkable variety of digital images and displays that, given the right conditions, effectively enrich communication capabilities. Real-time computer graphics are an essential component of the multi-sensory environment of virtual reality. Virtual reality technology is an emerging field with certain unique characteristics that lends itself so as to make possible the closest comprehension of a customer's vision by the product development team. Such systems project the human operator into an interactive environment mediated by artificial electronic & electromechanical devices. Moreover a typical virtual world combines multiple input & output devices operating concurrently, with the output servers allowing simultaneous feedback over multiple sensory channels. Though virtual reality would have made for a better user interface, owing to the time and material constraints in which we find ourselves, the next best option was one built upon the design and drafting package AutoCAD. The single most important qualification of this software is its ability to permit the creation and editing of object attributes. ENGINI~RING DATA MANAGEMENT Typically, many different CAD/CAM systems are involved, and often these are widely dispersed geographically and span many departments in one or more companies. Of recent, Engineering Data Management (EDM) providers are stepping in to fill this need. Once viewed as a database application with an engineering accent, EDM is rapidly evolving into a concurrent engineering enabler, or Concurrent Engineering Automation (CEA) product. To be a true CEA application, data management software must have strong networking features that support a wide range of open and proprietary networks, workstations, and personal computers as well as software drawing and model formats. In other words, it must be able to reach and service all participants in product development, regardless of what hardware and software they use. In addition, the product must provide varying levels of viewing and update privileges for individuals and groups. It must enable or automate project tracking and control features that tie into project planning and management software, and should include milestone posting, electronic approval mechanism, and percent-complete estimating. It has to accommodate document annotation and redlining. And of course, it must provide traditional EDM functions such as document storage and retrieval by meaningful engineering or manufacturing reference, search and retrieval, and automated data archiving and backup. The need for an automatic database update facility cannot be underemphasized. Amidst numerous databases, those of design and manufacturing demand to be explicitly mentioned in that these are at the heart of any concurrent engineering methodology in practice today. Consequently, getting these two to communicate successful and constructively, with an aim at effective EDM, may well be the keystone of concurrent engineering automation. Such an industrial perspective shall be adopted. Fourth generation languages like Superbase or Oracle could be employed, the virtue of these being their compatibility with CAD databases, such as the ones from AutoCAD. A Relational Database Management System (RDBMS). such as Microsoft Access, made available to the Product Development Team members across a computer network, has the ability to quickly search for, find, and bring together meaningful information stored as data in separate tables, these tables being related in a direct or an indirect fashion. This will be used as a real-time model management system, to manipulate and interface the object models of AutoCAD with a working domain which can control the process, based on the decisions taken in an expert system session. KRISHNASWAMYANDElshennawy: Intelligent Concurrent Engineering Environment 323 THE JOURNEY The salient features of such a process can be viewed from Figure 2. The product development team helps the customer communicate his requirements to them, by interpreting his product vision as a CAD drawing in AutoCAD. The creation of such a drawing could be envisaged as merely the collection of predefined features or objects, which could be parametrically manipulated to bring out the image realistically. Moreover, these pre-defined features or objects could be so defined as to reflect the machining capabilities of the manufacturing industry in question, thereby always providing due consideration to the manufacturability of the product. This drawing serves as the ideal product, the delivery of which to the customer, after its safe design, will make the customer feel that the product is truly his/hers. U s e r I.uteriace~ (Au~CAD) j 0 [w°rki.~l _. [ _ i ~ a_n a g e m e n t 1 _ . ,--, [ ob~t/~,.tur, uomam ~ . ~ystem . psl----I~[..J p Databases I (W~do,~] f [[SuperDase/Orae]e)J_ |(.]ne3udm8attributes) ... oj,,o.o o, F i g u r e 2: D e s i g n i n ~ for M a n u f a c t u r a b i l i t y An object oriented expert system like Level 5, containing the heuristics of DFM and DFA could be created, which would suggest intelligent changes to be made to the drawing so as to give it a design basis. Since all the above mentioned softwares can be set running under the windows environment, it will be possible to Open an output file to dump the suggestions of an expert system session, and make it communicate with the AutoCAD database through Microsoft Access, so as to alter the CAD drawing suitably. Against such a backdrop, the engineering data management system also keeps track of $uch other information as market survey reports to help standardize the product to stretch its appeal to encompass a larger customer clientele, information about available manufacturing processes, design analysis algorithms, and vendor inventory details. The relational database management system would be employed to relate all these discrete packets of data, so as to be made available to the product development team in a meaningful mode. The role of such a data management system is depicted in Figure 3. / / // ~-------~ /- t ....._.0~ - -lli \ .~--,-~ \ _ ~ ¢ \~ I o4/ //) / :s.., ,,.Figure 3: Engineerin~ Data Motor. 324 Proceedings of the 15th Annual Conference on Computers and Industrial Engineering CONCLUSIONS & RECOMMEMDATIONS FOR FUTURE WORK Engineering data management (EDM) has come a long way since its humble beginnings as a simple file manager. One can expect this technology to be greatly improved and grow in importance as an enabler for simultaneous engineering and systems integration, while at the same time giving users much-needed automated archiving, retrieval, and updating systems for engineering and manufacturing documentation. Global macros with jurisdiction over various software applications, could lead to the successful fabrication of a dynamic concurrent engineering environment, in which, apart from the phase of user interface, the rest of the product development process would have been gifted with intelligence enough to turn out products that both the customer(s) and the organization can proudly associate with. REFERENC'F_~ Clausing, D., 1991, "concurrent Engineering", Proceedings of the Design Productivity International Conference, pp. 35-44. Erevelles, W., Alptekin, S., and Ragsdell, K., 1991, "Design for Assembly/Design for Manufacturability (DFA/DFM): An Industrial Survey", Proceedings of the Design Productivity International Conference, pp. 229236. Chen, Y., and Miller, R., 1991, "On Implementing an Integrated Design-Manufacturability Assessment Environment", Proceedings, Computers in Engineering, pp. 407-413. Vujosevic, R., and Kusiak, A., 1991, "Data Base Requirements For Concurrent Design Systems", Engineering Databases: An Engineering Resource, pp. 109-113. Hardwick, M., and Downie, B., 1991, "On Object-Oriented Databases, Materialized Views, And Concurrent Engineering", Engineering Databases: An Engineering Resource, pp. 93-97. Krishnaswamy, G., and Elshennawy, A., 1992, "Finite Element Methods in Customer's Quality Function Deployment", IASTED, Washington D.C., November 04-06. Heiler, S., and Zdonik, S., 1990, "Object Views: Extending the Vision", Proceedings, Sixth International Conference on Data Engineering, pp. 86-93. Keller, AM., 1985, "Updating Relational Databases Through Views", Ph.D. Thesis, Stanford University, Stanford, CA. Autodesk, Inc., 1990, "Introduction", AutoCAD Development System: Programmer's Reference Manual, Autodesk, Inc., pp. 1-10. Bricken, M., 1991, "Virtual Reality Learning Environments: Potentials and Challenges", Computer Graphics, July, Vol. 25, No. 3, pp. 178-180. Microsoft Corporation, 1992, "Macro Basics", User's Guide: Microsoft Access, Microsoft Corporation, pp. 506514.