When I joined the group, Hans naturally chose a project in metalcutting for me to work on. He was interested in the friction between the chip and the tool. When a chip moves over the tool face, it will either stick and cause a built-up edge, or it will fracture, or it will go smoothly. He had already classified those three types of chip formation, and he surmised that chip formation must be very much influenced by the amount of friction between the chip and the tool.
He wanted me to study the fundamentals of the friction process in chip formation. Because material tended to weld to the tool face and form a built-up edge, Hans guessed that the friction between the chip and the tool face must be very high. But there was no way of knowing how much friction there was. So, my thesis was a study of friction between chemically clean metal surfaces, and I developed a theory that explains the nature of friction between such surfaces. This yielded the basic equations for the mechanics of friction for either boundary-lubricated surfaces or very clean, dry surfaces that are still used today.
I found that application of the science of mechanics to the chip-formation process resulted in equations that made it possible to calculate machining forces and other variables from first principles. That provided the first step in developing a scientific basis for engineering the metal cutting process.
Those equations kind of broke the ice and opened up the field of manufacturing to science, and demonstrated to everyone that you can apply science to what goes on in the factory. People began to look at their processes and ask questions: were the processes amenable to application of various kinds of science--mechanics, thermodynamics, and other branches?
ME: So you built off the qualitative understanding of the metalcutting process that Hans Ernst had developed, and took that to the quantitative realm?
Merchant: Exactly. He had an empirical understanding.
ME: Along those same lines, your history of machining research in the US describes an empirical phase and a theoretical phase. Where are we now?
Merchant: We are of course continuing to develop the science of manufacturing. But the impact of computer technology on that is a factor that cannot be overemphasized. It allows you to bring in science much more readily than the old "hammer and tongs" way did. So the real revolution began with the introduction of computer technology about 50 years ago.
ME: And that's about the time when, according to your book, there was quite a large gap between the theoretical knowledge that had been developed and what was being applied on the shop floor. What factors helped drive that knowledge from theory toward practice?
Merchant: The requirement to machine aerospace materials is really what drove a lot of the theoretical knowledge down to the shop floor. MetCut Research Associates, where I spent some five years as a research engineer and which later became the source of today's Machining Solutions program at TechSolve, recognized the challenge, and they undertook, with Air Force funding, to change that situation.
And they really did. Their machinability research was quite an undertaking, larger even than the research of Frederick Taylor, who focused primarily on turning.
ME: Speaking of Frederick Taylor, according to your book he felt that his contributions to metalcutting science were overshadowed by the attention paid to his management theories--like the concept of the division of labor. What would you like to be remembered for?
Merchant: For recognizing the potential of computer technology, and recognizing that it was a systems tool and that manufacturing had to be approached and operated as a system. The only way you could really do that, and integrate the system, was by using computer technology. So I believe that started the whole movement, slowly at first, but then continually gaining momentum, toward really applying the capability of the computer as a systems tool to integrate, operate, and optimize manufacturing. After all, as I said at that time, manufacturing is a system, and it should be operated as a system. So I started working more in that area.
ME: And you began talking about the concept of computer-integrated manufacturing (CIM) about 40 years ago?
Merchant: Yes. I wrote and published several papers on it that described the concept. That was the main thing, to get people interested in developing the required software based on that concept. So the idea was to clarify the concept of CIM, and to make sure the possibilities were understood so people could develop meaningful software. I did not get directly involved in software development, but I encouraged it. That was the only way that practical CIM for industry--CAD, CAM, and all the rest--would be developed.
The thing, to me, was to be sure that everything was kept integrated, so that CAD, CAM, and other types of software were not developed as independent technologies. They had to be integrated technologies to get CIM to work. That was a hard fight, to get people to accept that, because they were interested in concentrating on just one area or the other. But that's just human nature, I guess.
ME: What progress is there to be made in the area of CIM?
Merchant: The progress that is being made now can be characterized as leading toward making the whole system capable of operating as a holonic system. That's a system in which every entity--machines, software, people, all the entities involved in a manufacturing enterprise--can communicate and cooperate totally with every other entity in the system, whether it's a machine, a person, a piece of software, or whatever.
That concept requires that entities in a system give up a part of their autonomy and become a partner in making the system operate in an optimized way. The big thing that has to be done to have a truly global manufacturing capability is development of a system and technology that enables anyone in any part of a global manufacturing enterprise to communicate, collaborate, and cooperate fully with anyone else in the system, anywhere in the world, just as though they were in the same room face to face.
Gradually, the technology to do that is coming, but it still requires a lot of improvement. A key is going to be continuing development of virtual reality software. It has so much potential, and it's so necessary to make the concept of holonic manufacturing work on a global basis. When that technology is finally brought to the point where individuals in one location in a global enterprise can be made to feel that they're in the same room with others in a distant location, then we will have realized the full potential of holonic manufacturing.
ME: Help our readers better understand the concept of holonic manufacturing. Is, for example, a cellular system consisting of horizontal machining centers that replaces a transfer line for machining engine blocks an example of a holonic system?
Merchant: No, not in itself, but it can be made to be such by enabling all its items of hardware and software and all the humans who interface with it to fully communicate and collaborate with any of the other entities in that system.
ME: But we need to expand and extend that concept to the manufacturing enterprise as a whole?
Merchant: That's correct. And in particular, to be able to integrate humans into that system. That's the difficult part, and to provide the incentives and the capabilities and the tools needed for them to work in total collaboration with all other humans in the system.
ME: Integrating humans into a holonic system is not only going to take hardware and software tools, but education. How are we doing in terms of developing and educating people to fit into that kind of system?
Merchant: SME's Education Foundation provides grants to colleges and universities for their development of unique programs of manufacturing-related engineering education, and, as such, we're of course interested in defining what the manufacturing industry's requirements are for graduates of academic programs in engineering. We want to encourage and reward the development of curricula that supply the kind of graduates that have the competencies that really match the manufacturing industry's needs. I chair the committee that evaluates proposals, and that's the main criterion: will the proposed program of curricula and supportive items make a major contribution to the development of graduates that really fit what the manufacturing industry requires, and will require in the future?
So, we seek feedback from industry in terms of what competencies are lacking in graduates that they're currently getting from engineering programs. Based on that, we define what the competency gaps are that have to be filled--what are the competencies that are most important to manufacturing companies that they're not seeing in new engineers? So we have developed a list of current competency gaps, and if the program under evaluation does not target at least one of the gaps, we don't fund it.
ME: What are some of the competencies that industry is looking for that engineering graduates lack?
Merchant: There are about a dozen all together on the current list, which gets revised every two or three years to keep up with the changing needs of industry. On the current list, the top ones are general business knowledge and skills, project management, written communication skills, and knowledge of supply chain management.
ME: We've discussed manufacturing and engineering education at the university level. What do you think about the status of manufacturing at lower levels, in primary and secondary education?
Merchant: There are many very wrong ideas being held out there about the nature of manufacturing and manufacturing jobs. These reflect an old-fashioned view of industry--one that hopefully will soon be gone forever.
The current result is that students have less interest in going into engineering than into other professions. So, we're not producing enough engineers of any kind in this country to satisfy the coming needs of industry. So of course we're importing engineers. It's fine to have a diverse technical workforce, but we can't depend on that as a way of continuing our manufacturing leadership in the world.
And, if you look at the employment of engineers in all the different kinds of industry, 50% of all engineers are employed in the manufacturing industry. Then, if you look at industries that produce real, tangible wealth, there are only four: agriculture, mining, construction, and manufacturing. And 90% of engineers in those four industries are in the manufacturing industry. Yet, very few students actually set out to study manufacturing engineering as such. So, engineering curricula ought to be highly manufacturing-oriented, because the chances are very high that a person who graduates with an engineering degree will work in manufacturing.
That's the big challenge that I see for SME and for manufacturing in general--to make engineers realize that their best opportunities are going to come about if they have a real understanding of, and background in, manufacturing. So, engineering curricula should really be strongly manufacturing-oriented.
ME: Going back to the concept of holonic manufacturing, and the notion of all entities of the system being able to fully collaborate and cooperate with all other entities: how does that concept fit into the management side, for example, with lean manufacturing?
Merchant: Holonics almost automatically helps the enterprise move toward lean manufacturing, because it really enables the human entities in the system to communicate and collaborate with all other entities in the system. A guy on the shop floor, for example, can be trained to see exactly what kind of bottlenecks there are, what things are standing in the way of production flow, and can then fully communicate and cooperate with the other entities in the system to identify plans to eliminate those bottlenecks or backlogs.
ME: What would you like to tell our readers about your life, your career, metalcutting, manufacturing in general?
Merchant: I just very much enjoy the whole idea of looking at things in the field of manufacturing from as many different points of view as I can. It's a matter of generating ideas that can lead people in new directions in manufacturing. I didn't get in and actually do many of these things--for example, I didn't develop CIM software. I mainly just proposed some ideas and some ways of thinking that people could then take and use to develop whatever tools and software were necessary. Dick Messinger, Vice President for Research for my later years at Milacron, used to call me "The Philosopher of Manufacturing". Perhaps he was right.
It's been fun, from the day I cut my first chip on a lathe at the University of Vermont's machine shop. I've had a wonderful time. That's why I'm still here working--because it's so much fun.
This article was first published in the July 2004 edition of Manufacturing Engineering magazine.