Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 RECOMMENDATIONS - The recommendations that follow are based on the need to support progress in factory data management techniques and increase the use of CIM in response to the declining competitiveness of the U.S. manufacturing base. It is estimated that fewer than 1,000 U.S. companies are seriously pursuing CIM efforts. One objective of these recommendations is to ensure that the companies considering or just beginning a transition to CIM have access to the knowledge already gained by those few manufacturers that have in some cases as much as 20 years of experience with the concept. A second and more signifi- cant objective is to ensure more collective action than has occurred to date in implementing CIM. Implemention of CIM is beyond the capabilities of most individual manufacturing organizations. This fact implies the need to develop and disseminate knowledge on factory data management issues by both industry and government working toward the same ambitious goal--CIM more widely applied in U.S. manufacturing. 1. The Committee recommends that the National Aeronautics and Space Administration adopt a strategy of computer-integrated manufacturing for its Manned Space Station program. Major programs that NASA undertakes have limited schedules and budgets, rely on unproven product technologies, and operate in an environment where quality is critical. These circumstances point to an increasing need to automate design and production operations to achieve program goals. This Committee believes that the Manned Space Station program is not only a logical but an essential place for NASA to initiate a concerted computer-integrated manufacturing effort. The use of CIM for the Manned Space Station is essential to meeting the program's administrative goals--a manned station in space in 1992 at a total cost of $8 billion--because manual coordination of the data to support the design and manufacturing of systems components would be time consuming and prone to error. Adoption of a clear CIM strategy by NASA would avoid the evolution of incompatible approaches as each manufacturer or supplier prepares to respond to NASA's space station requirements. Unless the efforts of the hundreds of companies supporting the space station program are coordinated by means of CIM, design engineering and production errors are likely to proliferate. , ., . . . . errors 40
41 Finding and correcting these errors would entail substantially higher costs, longer development time, and perhaps reduced operational capability. Because many components in the space station will be unique or produced in small batches, the most obvious benefits of CIM would be realized in engineering design. However, including production in the CIM system would provide a significant measure of flexibility in the project configuration and would ensure high quality. A secondary reason for the use of CIM in the Manned Space Station program is to further the development and use of the technology. NASA has a legitimate role in sponsoring the development and diffusion of technology, such as CIM, that is important to the country but that initially, at least, is beyond the resources of a single company to deliver. The Apollo and Space Shuttle programs provide clear precedents for NASA-funded creation and diffusion of new technologies. NASA can influence a large portion of U.S. industry because the program will involve many companies, in both aerospace and other manufacturing sectors. Investment in technologies such as CIM is driven by the marketplace; with an $8 billion space station program, NASA will be a powerful market force. The program managers will need to plan carefully to use that power constructively. Adoption of a strategy of computer-integrated manufacturing would require implementation of CIM throughout the pertinent operations of both NASA and its contractors. Such an effort would involve many combinations of computer equipment from various vendors. To permit communication among these heterogeneous systems, NASA would have to adopt standards for data definition, data formats, languages, and protocols. The communications problem would be difficult but, in the Committee's opinion, surmountable. In solving it, NASA could draw on four existing federal efforts: the Integrated Program for Aerospace Vehicle Design (IPAD), funded by NASA and the Navy; the Air Force's Integrated Computer Aided Manufacturing (ICAM) program; the Product Definition Data Interface (PDDI) effort under ICAM; and the National Bureau of Standards' Automated Manufacturing Research Facility (AMRF). NASA also could benefit from the experiences of companies now well along in implementing CIM. From information supplied by five such companies, the Committee has extracted a set of guidelines for CIM programs (Table 1, at the end of this chapter). Although the guidelines are oriented toward industry, they are applicable in important respects to NASA. They indicate, for example, that to implement an effective CIM program, the Administrator would instruct the manager of the space station program and the directors of the NASA laboratories involved to prepare a master plan. Execution of the plan would be directed by the Administrator through the program manager and laboratory directors. The Administrator also would direct efforts to take full advantage of the experience of companies who have been impl ementing CIM.
42 2. The Committee recommends that companies form consortia to pursue research and other projects in CIM not readily undertaken by . individual companies. advanced sell CIP The technological accomplishments required to achieve CIM entail enormous expense, creativity, planning efforts-, and amounts of time. Only the largest U.S. manufacturers, or those responding to a specific marketplace requirement for CIM, can be expected to apply the needed resources to their CIM efforts. Yet the Committee believes that a majority of U.S. manufacturers will not be able to remain competitive in the quality, timeliness of delivery, and cost of their products unless they use CIM. Therefore, cooperative efforts among companies will enable a broader base of U.S. industry to achieve CIM. The companies that have started using CIM have created their own systems. It will be some time before the technology is sufficiently and potential users sufficiently sophisticated for vendors to _ systems. At present, user companies still need to do their own development, a task that is beyond the capabilities of many smaller companies that need CIM to remain competitive. To speed the rate of application of existing knowledge about CIM, and to identify priorities for research, groupings of companies should organize to pool their efforts. Individual company applications of CIM technology will vary in hardware, software, and order of implemen- tation. However, many issues in data communication, data bases, and process modeling surpass the abilities of single companies and could best be resolved cooperatively. The Department of Commerce's R&D Limited Partnership program may offer a useful mechanism for forming consortia. Any company that purchases a large number of manufactured components is a candidate for a cooperative effort. As a specific instance, the Committee recommends that the automotive industry--auto producers and their principal suppliers-- embark on a joint program of development and implementation of techniques and standards in the area of CAD/CAM integration. This activity might be carried out by the Automotive Industry Action Group (AIAG), a nonprofit industry association that is disseminating just-in-time technology to the industry's suppliers, or by another consortium with comparable legal and support arrangements. The automotive industry already is pursuing a number of elements identified in this report (GM's Manufacturing Automation Protocol, the IGES standard, and component releasing/shipping/inventory systeme), but further work is required to extend communication standards to the CAD area, to technical vendor/customer communication, and to manufacturing data bases. . In the spirit of Recommendation 2, the Committee believes that companies and industrial sectors should build close working relation- ships with universities in education, research, and technology transfer related to CIM. Industry also should seek analogous relationships with the relevant federal programs, drawing on ICAM, IPAD, and the AMRE for appropriate technology.
43 3. The Committee recommends that the Computer and Automated Systems Association compile knowledge of CIM technology, drawing on both industrial and governmental sources, and make it available to industry, to universities, and to governmental agencies. All of the companies involved in the Committee's site visits had spent a good deal of time planning and organizing for computer- integrated manufacturing. Companies that are not as far in their thinking about CIM should not have to develop their plans from scratch. If existing information were available in an organized form from a central place, diffusion of CIM technology would accelerate. The intent of this recommendation is that existing information on CIM be collected, organized, and disseminated to current and prospective users. New information would be added as it became available. The Computer and Automated Systems Association (CASA) of the Society of Manufacturing Engineers has the constituency, mandate, and authority to carry out this recommendation. CASA's membership comprises engineers with an interest in and experience in computer-integrated manufacturing. Although some of the information developed by the leaders in computer-integrated manufacturing is proprietary, much of it is not. The availability of such information in well organized formats could be extremely helpful to manufacturers who are interested in CIM but lack the resources to initiate and develop programs unilaterally. Adoption of Recommendation 1 would create a large demand for this type of information. Detailed understanding of CIM resides largely in industry, and progress in the field has outpaced our universities' ability to play their traditional roles--research and teaching--in this area. Curricular materials in the field are extremely scarce. Effective efforts to make current information on CIM readily accessible to faculty members would yield important benefits and, in the long range, are probably indispensable to the nation's well-be~ng. 4. The Committee recommends that the federal government continue to undertake research to resolve fundamental technical issues related . to computer-integrated manufacturing. Resolution of the technical issues in Chapter 3 requires more research than a single company can undertake on its own. Data communication in a heterogeneous system, validation and consistency of data, representation of textual and geometrical data, expert systems, and analytical models of manufacturing processes are all risky areas of research, requiring multiyear, cooperative efforts. Solutions to these problems are needed to advance work in computer-integrated manufacturing. We believe that the national research agenda should be revised to incorporate these needs and that the efforts initiated should be given long-term support. The Federal Coordinating Council for Science, Engineering, and Technology, chaired by the President's Science Adviser, might be the appropriate group to manage federally funded
44 research related to CIM. Federal science policy-makers need suffi- cient understanding of the generic research issues outlined in this report so that a small percentage of the federal research budget can be effectively directed toward building the cumulative knowledge base necessary for progress on there issues. As this report states, the federal government can be pleased with the results of its initial modest investments in CIM technology. In the past 10 years, the Air Force and NASA have sponsored the two largest government programs that have furthered technological develop- ments in support of CIM. While the Air Force's ICAM is concluding by tying its findings together in the demonstration of a sheet metal center, the NASA IPAD program has ended without any such wrap-up. IPAD's use of Hyperchannel was a step toward communication between heterogeneous computers, and the program had begun to extend its work to the transfer of geometric data within heterogeneous systems. It would be a shame to lose this valuable research; it should remain on this country's agenda of research related to CIM. 5. The Committee recommends that federal agencies that purchase manufactured goods accept digital data sets compatible with the Initial Graphics Exchange Specification, rather than requiring conventional drawings, as a deliverable item under contracts. This recommendation applies to all federal agencies that procure manufactured goods with high tolerance specifications and require the ability to replicate these products. These agencies include NASA, the Departments of Defense and Energy, and others. Some federal contractors already handle graphical material inter- nally in the form of digital data and create conventional drawings solely to satisfy federal contracts. Besides the inefficiency of this procedure, errors that creep into drawings produced only to satisfy contracts may not be found until another contractor tries to use the drawings. Acceptance of digital data by the government would obviate this problem and, more importantly, would promote the creation of computer links between organizations as well as within them. The government's acceptance of digital data, instead of drawings, as a deliverable item under contracts would also increase the market demand for CIM technology and expedite the transition to computer- integrated manufacturing in this country. The requirement for data in a form compatible with the Initial Graphics Exchange Specification (IGES) is recommended because IGES is the only communications protocol as yet widely accepted in industry. It establishes an initial basis for direct digital exchange of graphical data and has been adopted as a standard by the American National Standards Institute. Although IGES currently applies only to limited situations, it is an important first step in accepting digital data. Enhancements to IGES should result from current research such as the Product Definition Data Interface (PDDI) project. PDDI is based on IGES,
45 thereby reducing the risk of system obsolescence. As enhanced versions of ICES or other digital standards gain acceptance in industry, they should also be used by federal agencies. 6. The Committee recommends that manufacturing companies considering investment in product design or manufacturing process technology consider computer-integrated manufacturing. Adoption of CIM technology is essential to the maintenance or recovery of competitiveness by U.S. manufacturers in domestic and world markets. Companies regularly find themselves losing ground competitors who are introducing CIM. Nevertheless, far too few companies in this country are working seriously to adapt the concept to their operations. Companies that are about to invest in product design or manufac- turing process technology should be aware of the potential benefits of CIM, as well as the consequences of postponing the decision to plan and gradually implement an integrated system. Competition in manu- facturing can only become more intensive--witness the considerable development effort abroad (Appendix C)--and companies that do not move into computer-integrated manufacturing, the Committee believes, face a dim future. The necessary extent of the top executive's involvement in such a transition is suggested by the guidelines in Table 1. TABLE 1 GUIDELINES FOR CIM Each organization must create and execute its own plan for computer-integrated manufacturing, but the experiences of companies that have done so successfully indicates that the following guidelines are generally applicable. Management (guidelines · The chief executive officer (CEO)* decides whether CIM is the right technology for his organization, makes the decision to invest in CIM, and gives the transition his full support. Because computer- integrated manufacturing cuts across organizational boundaries and affects all people in an organization, these decisions must be made by the CEO. · The CEO bases his decision on performance goals, such as reduced work--in-process, increased market share, and quality. Conventional justification methods for investments cannot capture the benefits of CIM, which are cross-cutting and difficult to measure. * In conglomerates, the highest executive in a business unit will be more appropriate than the literal CEO.
46 · The CEO directs the development of a companywide, long-term plan that will guide the CIM effort; the plan includes yearly targets, budgets, and measures of performance. · The CEO assigns a senior executive, reporting directly to him, to manage the CIM effort. · To work under the CIM manager, the CEO assembles an inter- disciplinary team of highly competent people from all areas of the organization. · The CEO ensures that employees at every level in the company understand and support the CIM plan. Good communications must exist among people before computer integration of their areas of responsi- bility is possible. · The CEO recognizes major barriers to CIM--adverse labor/ management relations, financially oriented managers, and a narrowly trained work force--and takes steps to mitigate them. · The CIM manager implement e the plan gradually, adjusting the pace of adoption to the company's ability to assimilate change. · The CIM manager seeks opportunities for cooperative efforts where appropriate--with universities for technology transfer, education, and research; with national efforts such as ICAM, IPAD, and the AMRF; and with professional societies to share knowledge, educate employees, and encourage common data communication standards. Technical Guidel ines o Define and locate data bases shared by product engineering and design, production engineering, and production operations. · Define protocols and conventions used in data definition and handling. The evolving ICES standard and results of the PDDI working groups are suggested as guides. · Recognize that machine tools, computers, and other engineering and production equipment are parts of a total system, and select and maintain them accordingly. · Establish as a goal of the integrated system the ability to use hardware and software from any supplier.
