plans. For the most part they remain uncoordinated and too small to make the impact required to restore and retain a successful competitive position for the United States. A mandate from Congress will help speed up the process and increase the

effectiveness of government activities in manufacturing research

and development.

Thank you for the opportunity to testify today. Given the time limit of ten minutes I have tried to keep my remarks general and to avoid the specific facts, figures and examples which would draw my testimony well beyond the time limit. I will be happy to answer questions and amplify on my remarks at your convenience.

Mr. ROE. Thank you, Dr. Nagel. Mr. Kramer.

STATEMENT OF SAMUEL KRAMER, DEPUTY DIRECTOR, NATIONAL ENGINEERING LABORATORY, NATIONAL BUREAU OF

STANDARDS

Mr. KRAMER. It is a pleasure to appear before the Committee to describe the National Bureau of Standards program in automation and in particular the program being conducted at our Automated Manufacturing Research Facility. With your permission, I would like to submit my prepared statement for the record, and summarize the remarks; and hopefully then respond to questions.

Before I start my comments, I would like to say that the Bureau of Standards has always, in the National Engineering Laboratory, stressed the importance of quality control and interacting with industry and in tech transfer. By coincidence, Dr. Nagel, who just testified, I would like to note, we have had the privilege of having him with us for a few years in the very early stages of the formation of our automation program. Dr. Nagel contributed significantly, and then left us to join Lehigh, and if I may say so, Roger represents one of our better tech transfer efforts. And we still maintain a very close collaboration of working between Lehigh and the Bureau.

Another thing is, I might note, I did my homework before coming to see who the other witnesses were, to note that Eastman Kodak Company, and Air Products, both corporations at many levels, but including from their CEO's, are participating with the National Bureau of Standards, and in particular in our latest effort, the Malcolm Baldrige Quality Award,3 in trying to recognize industry and encourage it. So we thank those corporations for their contributions. But more, I would like to point out they merely exemplify what we consider to be a very important role for the Bureau and that is to be in continual dialogue with industry and academia and not to operate our program in isolation from them. I guess if I end there, it would be the best part of my testimony.

The Automation Program at the Bureau is part of a coordinated NBS technical effort to support U.S. industry's improved productivity, quality control, and international competitiveness. It began in the 1970's with research on robotics and development of computercontrolled coordinated measuring machines. In the 1980's, we first introduced the twin concepts of an Automated Manufacturing Research Facility [AMRF] and a plan, a definite plan for transferring advanced manufacturing technology to American industry, especially the small and medium-sized manufacturers.

We have completed the construction of the initial AMRF. The facility operates now under a computer control system using advanced control approaches pioneered at NBS. In December, 1987, we demonstrated its ability to perform and at that one run alone we had over 500 people from industry, academia and other Government agencies participate. To date, we had over 70 research associates representing about 40 companies and Government agencies who have worked alongside the NBS staff in the activity. In addi

3 The Baldrige Award was established by Public Law 100-107, the Malcolm Baldrige National Quality Improvement Act of 1987. It commemorates the late Secretary of Commerce.

tion, we have been affiliated with about 50 universities in various forms of research. Industry's commitment in this program, I think, can be displayed by 36 companies who have loaned or donated equipment at a value of over $5.7 million to the National Bureau of Standards' AMRF, and that equipment operates in our facility with our research.

The AMRF is a unique engineering laboratory in that it provides a basic array of manufacturing equipment and systems, a testbed that researchers from all sectors use and experiment with-experiment with new standards and study new methods of measurement and quality control for automated manufacturing.

The AMRF incorporates some of the most advanced and most flexible automated manufacturing techniques in the world. NBS is the Nation's primary laboratory for measurement science and engineering, and has two principal goals for its automated manufacturing program: to supply the industry with new ways of making precisely machined parts-with dimensions that can be referenced to the national measurement standards-and to encourage the modernization of American manufacturing by providing the technical information necessary to develop standardized interfaces between various types of equipment.

The National Bureau of Standards is also using this testbed and will do so even more as the testbed for research on the next generation of "knowledge-based" manufacturing systems-automation systems that incorporate artificial intelligence capabilities.

Let me go back a little bit, historically. Manufacture and measurement have always been two separate processes. A machinist would cut a part on a milling machine and stop periodically to check dimensions with calipers and gauges. As manufacturing techniques became more and more efficient, the measurement part of the operation consumed an even greater percentage of the time required to produce a part. The development of the automated coordinating-measuring machine in the 1970's helped somewhat, but measurement still used up about 50 percent of the total time required to produce precision parts.

It would be many times more efficient if the machining process could be made to produce accurate parts without being interrupted by the measuring processes. Not only would it take less time, but fewer parts would have to be scrapped for being out-of-tolerance. Some surveys have shown that in some cases one-third of the work force is engaged in either rework or correcting out-of-tolerance parts.

The National Bureau of Standards research suggests that the program can be solved by use of today's computer-controlled machine tools, because the position of the cutting edge of the tool is known and controlled at all times, at least in theory, by the computer. The computer could then be programmed to compensate for known errors in the machine's movement, using sensors that feed back information on the machine's condition.

The concept of feedback and process control is well known in some industries, such as oil refining and chemical production. In discrete parts manufacturing, however, it will require the development of a whole new generation of sensors and control systems.

Now, this is not just theory, because NBS researchers have already applied some of these ideas to commercial machine tools and improved their performance in terms of accuracy and control five to tenfold. Some of this research already is finding its way into the marketplace.

If I may note one thing, and it was not in my talk. As an example of it, is a major project that is going on in the AMRF in what we call the Mare Island Work Station, which is being performed for the Navy. The Navy has a problem in producing what I call RESIC parts. It stands for "Resident Enabling Sound Isolating Connectors," that are used in nuclear submarines. It is a very precision machine part. It comes in about 40 different sizes and shapes. It is made up of a special material that is extremely hard to machine. Currently, these parts must be machined on order as the submarines come in for rehab, and it takes approximately 17 hours to make a part.

We are jointly working with the Navy, with NAVSEA, with Warner-Swazey, with Westinghouse. Two industrial companies have developed what we call the Mare Island Work Station. It has been set up at the Bureau of Standards. The part that took 17 hours to produce is now being produced on demand in 20 minutes. We will run this machine, producing the parts for the Navy-it is anticipated that this one work station will be able to supply the parts of the entire Navy, and they even anticipate that the capital funds expended on it when it goes into full operation will be recovered in less than 2 weeks. This fall, if it is successfully run, we will then pack up the machines because we are not in the manufacturing mode, transfer it to Mare Island, CA, turn it over to the Navy for full operation. And we will already take off-we have in mind another project for them. I might say though, the result of this while it is going to DOD to the Navy, already the private sector is aware of it, we have tremendous interests from private machine tool manufacturers and others. We are looking to the transfer of this type of technology to the private sector. We are quite proud of it, and we hope it exemplifies some of the tech transfer.

The future for the manufacturing automation or robotic effort appears to be just as interesting as the past, as I noted. The AMRF will be used to do more research on automation, and we will also use it as a working research facility for tech transfer.

We expect to have an increase in collaborations with the private sector as our work becomes more directly applicable to the factory floor, and as industry becomes more automated. Based on the belief that this technology must diffuse rapidly if the nation is to compete internationally, we plan to continue to give priority to our efforts to transfer this technology to industry; we will look for more research associates and visiting scientists. We will make more videotapes, give more talks and write more articles. We continue to seek efficient ways to move the technology to small firms through intermediaries. Intermediaries are extremely important to us-intermediaries such as equipment vendors, trade associations and professional societies.

The National Bureau of Standards views technology transfer as an inherent part of our research program. Our Organic Act cites our responsibility to support the U.S. industry, and this aspect is

built into all of our programs in the National Engineering Laboratory. For example, in addition to the previously noted AMRF activities, we conduct seminars and training sessions, we present talks to the private sector groups, academia, professional societies and participate in numerous standards committees. In fact, Roger mentioned Dennis Swyt. I don't think Dennis is home, he is on the road with industry all the time, for the one reason, it is important we get the feedback from them as much as they know what we are doing.

The intensity of interactions can be exemplified by the seven runs that we had on the AMRF in which there was over-almost 4,000 participants from outside the Bureau participating in these demonstrations.

Our program addresses technological issues with a view to increased U.S. productivity and competitive positions in the complex world marketplace. However, we cannot do it alone. The cooperation and support of all parties in the public/private sector, academia and all other government agencies are essential if we are to be successful in this effort.

I would like to stop there, and then if there are any particular questions try to respond to the committee, et cetera.

[The prepared statements of Mr. Kramer and Mr. Lyons follow:]