In the digital factory: the next generation.

Lean, Agile, Just-In-Time - all touted as manufacturing cure - alls in the '80s and '90s. But there's a new kid (and a new acronym) in town - and makers of all shapes and sizes will have to adapt to succeed.

George Bernard Shaw once observed that all progress depends on the unreasonable man. His argument was that the reasonable person adapts himself or herself to the world, while the unreasonable person persists in trying to adapt the world to himself or herself. Therefore, for any change of consequence, we must look to the unreasonable person.

Until a few years ago, I thought of myself as an "unreasonable" man each time I presented the vision of the digital factory to one of the big manufacturing companies. I described the digital factory as an enterprise-wide computer solution that enabled manufacturers to plan, simulate, and optimize a complete factory, its production lines, and its processes at every level of detail. The digital factory had to support the development of a product from conception throughout production. Its database had to be available to everyone in the organization.

Some of the pieces were in place and everyone, of course, wanted to improve productivity, but within reasonable limits. Looking at the industrial process as a whole, and its three stages of product design, production engineering, and manufacturing in particular, manufacturers had been investing billions of dollars in Computer-Aided Design (CAD) systems and automated production machinery. But they had only half-heartedly invested in automating the crucial production engineering process which, in my analysis, was creating a major bottleneck in the industrial process. Overcoming this problem by implementing computer-aided production engineering (CAPE) would allow them to tap a formidable reservoir of productivity gains and lay the foundation for a new way of doing business - the digital factory.

In fact, back then in the '80s, nobody could estimate the size of the digital factory market - a market that has grown from $25 million to over $100 million over the past four years. It is expected to reach nearly $250 million by 2000, and eventually become a billion-dollar market, according to Barrington Capital Group.

THE END OF AN ERA AND THE BIRTH OF A NEW MODEL

Historically, manufacturers were monolithic organizations where the objective was to turn out as many units of a limited number of products as cheaply as possible. Production workers were evaluated on how many units they could install per hour and assembly lines were an accumulation of repetitive operations.

Beginning in the early 1980s, faced with fierce competition, U.S. manufacturers began to recognize that this model no longer worked. Competitors from Japan and Europe were offering better quality products in greater variety. It was then that I became convinced that the digital factory was not only the right vision, but a new and revolutionary way of looking at manufacturing.

While the world at large was busy trying to understand the impact of new "wonder" cures being touted, such as lean manufacturing, agile manufacturing, just-in-time manufacturing, design for assembly, etc., many car manufacturers, such as BMW, Ford, Renault, Volkswagen, and Volvo, had started to experiment with computer-aided production engineering (CAPE) tools to simulate robot behavior on screen and to design robotic work cells. They soon began to use "what you see is what you get" displays to address a wide variety of tasks that normally would have had to wait until the plant was nearly complete.

These benefits led more and more companies to use CAPE tools for designing the wide range of manufacturing processes including assembly, machining, quality inspection, welding, and painting, as well as for assembly and ergonomics studies. By the early '90s, production engineering had reached a milestone; most automotive, aerospace, and heavy equipment companies were using the CAPE tools and were reporting improved production and lower costs.

REVVED UP ROBOTS: FORD CLEANS UP

Ford was one of the first U.S. companies to take advantage of this emerging technology. By the end of 1989 it had about 80 robotic design systems, called ROBCAD, installed worldwide and had launched a technology assistance program to aid its key outside production design suppliers to also acquire ROBCAD.

Although initially Ford adopted the tools for work cell simulation anti off-line programming, it soon recognized that the real benefits would come from using the technology early in the design stage to influence decisions, correct mistakes, and optimize systems. In Europe too, companies such as Aerospatiale, Renault, and Volvo were expanding the use of CAPE and strengthening the communication between their design and production engineering departments. The building blocks of the digital factory were being laid - and I, for one, felt somewhat less unreasonable.

The '90s brought another shock to manufacturers: globalization of the markets, increased competition, and higher expectations from customers to produce more models at lower prices. These changes forced manufacturers to address some critically important questions: Could product designers continue to throw their designs over the wall? Could Original Equipment Manufacturers (OEMs) continue to ignore their suppliers' cries to have more influence over the designing process in order to build better lines? Could they afford long plant shutdowns for model changes and last minute engineering change orders? And could they continue to spend millions on physical prototypes for each new model? "Faster, cheaper, better" soon became the battle cry of nearly every manufacturer competing in the new global economy.

These challenges forced manufacturers to seriously consider re-engineering their business processes while adopting new methods to speed up the introduction of new models. The race to market meant that every player had to start running at the same time. Manufacturers started implementing concurrent engineering procedures; product designers started sharing data with process designers; digital mock-ups began to replace physical models; and everyone started to talk about a seamless process from design to production. Manufacturing worldwide was on the brink of a transformation. More and more companies were adopting CAPE - not as a specialist tool but as a mainstream production solution.

Major companies were reporting significant gains from wider use of CAPE. For instance, GM reported saving one-third in manpower and about six months of time on the entire implementation in tooling, design, construction, and installation on its P90 Malibu project. GE reported that it saved more than $300,000 per engine in reduced scrap. And Volvo was already introducing new models on the shop floor ()vet the weekend instead of the typical eight weeks.

ATOMIC ENERGY SAVES A BUNDLE WITH ROBCAD

Automotive manufacturers are not the only ones enjoying this new technology. Virtual environments are particularly beneficial where reality' can be hazardous to health and safety. Take, for example, the Atomic Energy of Canada (AECL). It was chosen by a nuclear power plant in New Brunswick to clean the steam generators of their nuclear reactor. Cleaning steam generators is a regular maintenance task that requires the highest level of precision.

In the typical cleaning process. chemicals are used to remove the magnetite deposits from the one-inch diameter tubes. The reactor is first powered down. Then the primary side of the steam generator is filled with chemicals. After the required time, the chemicals are removed and the system is filled with water. Chemical cleaning produces large amounts of contaminated liquid waste that must be handled and decontaminated by a qualified company. The liquid remains radioactive for many years and must be stored at a special site, which results in expensive transportation and storage costs.

AECL was the first to attempt automated mechanical cleaning in the nuclear industry. Before bidding on the job, they tested their ideas with ROBCAD. Once they had proved the procedure internally, they showed the customer a video of the simulation and won the job. AECL showed a less complicated procedure than traditional methods that saved money by producing less radioactive waste.

Before placing engineers and equipment on-site, AECL's engineers used ROBCAD to test the mechanical system's fitness for use. Only a few years ago, engineers would have spent thousands of dollars and hundreds of man-hours building physical prototypes of the new tools and a model of the nuclear reactor's steam generator. It was necessary to take precautions so there would be no surprises when they came to do the actual job.

AECL also used ROBCAD to optimize the operations and ensure that the procedure would take the same amount of time as the traditional method. Timing was critical as each hour the reactor was down, it cost the operating company between $15,000 to $20,000 dollars.

After engineers drained the primary system, they inserted the manipulators into the two sides of the primary system. One robot blasted stainless steel beads into each heat transfer tube, and the other robot arm collected the waste and beads on the other side. ROBCAD was used to monitor the entire procedure. The precision of the software and manipulating robot arm allowed AECL to pinpoint each of the one-inch diameter heat transfer tubes. The collection device separated the beads from the waste and recycled the beads back into the process. Engineers stored the 789 kg (1740 lbs) of magnetite waste in lead containers on-site. This saved a significant amount of money because it did not involve any third-party companies to handle, transport, and store the radioactive waste.

Today, these companies and more have moved beyond using CAPE for various manufacturing activities. They require integrated, interoperable, and enterprisewide manufacturing solutions that fit into their information technology architecture. CAPE providers are their partners in optimizing their production process and, in turn, improving their overall business.

For the digital factory to be effective, the software must be an integral part of the host IT infrastructure and be able to communicate both upstream with the CAD tools and downstream with controllers of the production facilities. Advanced technologies and methodologies are enabling seamless integration and communication between CAD, CAPE, and shop floor environments. Process databases and PDM (Product Data Management) systems are providing central repositories of all the company's information.

With all of these pieces now in place, the digital factory is likely to become even more important in the future. Now that all automotive and aerospace manufacturer are experienced users, more and more large implementation programs are being announced like Ford's C3P, Mazda's MDI, and the digital factory programs of BMW, Nissan, Renault, Rover, and the Airbus Consortium. Heavy, equipment and consumer electronics manufacturers are also increasing their implementation of the technology.

While manufacturing has taken great leaps forward during this decade, the revolution has only just begun. As product-design lifecycles continue to shrink and manufacturing operations become more costly and complex, flexibility will be the door to success and the digital factory the key.

Harel Beit-On is president and CEO of Tecnomatix Technologies Ltd., a company that develops, markets, and supports software tools to computerize the industrial process and achieve seamless transition from design to production.