Implementing CM From The Top Down

Does implementing CIM require systems integration? Yes, argues CSC’s Hoover. Without it, automation become’s a downhill run to oblivion.

November 1 1987 by William R. Hoover


For a long time, those of us in the computer services business have known we were living in the Information Age. Though it wasn’t until the early 1980s that it became common knowledge, and that was concurrent with the widespread use of personal computers.

We knew, for example, the computer’s enormous capacity and capability of compiling, categorizing and correlating data. We learned to use the awesome and almost instantaneous velocity of these functions. We slowly came to a new realization: It is not the computers or their systems but the information itself that has such tremendous inherent value.

I am reminded of Rear Admiral Grace Murray Hopper’s recent keynote address at the first anniversary celebration of the Pittsburgh Supercomputing Center at Mellon Institute. In her remarks, she pointed out that information is totally inert. It must be fed to another processor -a human being-who takes that information and turns it into intelligence or knowledge. Furthermore, she predicted that someday information will be listed on a balance sheet as an asset. Those two fascinating concepts from this pioneer of the modern computer era bear strong witness to the undeniable fact that we are living and working in the age of information. 

INFORMATION AGE

There is, however, an area of mankind’s endeavors that seems to be stalled in the late stages of the Industrial Revolution. Manufacturing is having difficulty moving into the age of information. We are seeing a deficit situation in the industrial sector, especially in the Western world. Manufacturers must transform information, energy and materials into value-added products. During most of the Industrial Revolution manufacturers devoted their primary attention to the mechanical processes: electricity and steam, machines and manpower. The third ingredient in their formula-information-played a lesser role until the early 1960s.

Postwar technology spread throughout the globe. High-tech products and markets blossomed. Traditional products and markets became unusually complex. In order to serve these new demands even adequately, manufacturers for the first time in their history had to begin assembling and manipulating huge amounts of data. They bought computers and assigned them to everything but manufacturing-payroll, accounting, the filing of historical data, etc. They didn’t use the computer as a tool to run their plants. The result is a serious information deficit in Western manufacturing today. It renders much of our production capacity noncompetitive.

Three or four years ago a potential panacea hove onto the horizon. It is called CIM-Computer Integrated Manufacturing. It seems like an attractive solution to the information deficit. Before taking a look at CIM, perhaps a brief review of where the computer services business has been laboring for the past three decades might shed some light on how manufacturing can move into the Information Age and tackle the challenges and opportunities of CIM.

The late 1950s and early ’60s witnessed the true arrival of the mainframe computer. These early machines had sufficient memory and processing capability to offer management in both industry and government an unprecedented opportunity to assemble and relate large volumes of data. In those days the computer hardware makers were selling both hardware and software but provided minimal training and implementation or application services. Small computer services companies came into being to serve that need. CSC was one of them.

We soon expanded our activities into directly serving the users of computers, and then to developing computer-based services for specific industries and markets. We entered the credit field, health and insurance, income tax processing, and formed a worldwide data communications network. That took us 25 years of highly technical work.

Today, CSC enjoys $1 billion in annual sales, employs 18,000 men and women and ranks 82nd in Fortune’s list of the 100 largest diversified-services companies. The company designs, engineers, integrates, installs and operates computer-based systems and communications systems. Our technical capabilities include: requirements analysis, computer software development, systems engineering and integration, a broad range of engineering disciplines and other technical support for high-technology operations. We have about 200 facilities throughout the U.S. and in 50 international locations. And in the huge multibillion dollar federal market we are the dominant independent supplier of computer services, obtaining about two thirds of our total revenues from there.

Now we find ourselves turning to new markets where services of our kind are desperately needed. It’s not that we are leaving the federal market, we simply believe there are many nongovernment operations that need our level of sophisticated service in order to survive competitively. Before we look at manufacturing, let me share with you the development of formal systems in manufacturing via, of course, the computer sciences. 

MANUFACTURING MARKETPLACE

About the same time the computer services businesses came into being, a new formal system of running a manufacturing plant began to emerge. We know it today as MRP II Manufacturing Resource Planning. In its embryonic stage, this system was a simple incoming inventory control program. IBM called it PICS, or Production Inventory Control System. Then it evolved into a more sophisticated inventory control including ‘MP or work-in-process inventory and then finished product inventory. Along the way the system reached out to include BOM or a planned Bill of Materials, CRP or Capacity Requirements Planning and Shop Floor Control. With that it became a scheduling tool and the Master Schedule became its central engine. It was known then as a closed loop system or just MRP-Material Requirements Planning. The late Oliver Wight expanded the concept to embrace purchasing, marketing, engineering and finance, and changed its acronym and name to MRP II. This evolution of a system has spawned a whole new computer software industry with about 250 MRP II software packages now available on the market.

Despite the fact that MRP II is a proven system that has brought success and profit to its users, there is a serious lack of widespread acceptance. Thousands of plants own the software. Only a few thousand are using it to any extent, and most of them use it as an inventory control only. Only a few hundred use the system at its highest level of efficiency. These users have improved their cashflow, deliver their product on time, 90 percent have reduced their total inventory and keep their delivery promises, improved their share of market, and no longer allow engineering change orders to disrupt their production schedule, etc. These are newsworthy accomplishments and millions of positive words in the business press have been published, yet the system is just now gaining some credibility in the executive suite and in board rooms across the nation. 

JIT AND CIM

I believe the climate of acceptance is changing due to two new factors: JIT and CIM. Just-In-Time techniques, learned from the Japanese, proved to be extremely efficient, especially in those plants which had implemented some form of MRP or MRP II. Any plant which had a formal system of scheduling resources: material, men, machines, energy, time and money, could put JIT practices to work in vendor scheduling, in work-in-process and in delivery of the finished product on time to the customer. JIT, on the other hand, did not work well in plants where an informal expediting mode was the way of life. JIT became an enhancement of MRP II, and it proved to be a competitive advantage for those MRP II user companies. Like MRP II, this practical tool has not gained widespread acceptance in the American manufacturing world.

Enter CIM. This is the second factor which is making MRP II more popular among CEOs. Unfortunately, CIM came on stage disguised as robotics and automation. Thousands of CEOs, thinking, in all sincerity, that their problems could be solved by a quick fix, embraced CIM by spending huge sums of money on the very latest robots and automated production lines. Many of these plants that thought they were getting into CIM were the very same ones that had not installed MRP II or JIT. Their shiny, new robots and high speed automated lines only continued to make the wrong product at the wrong time more efficiently-and, of course, they were not in any way involved with CIM. The robots did their work as did the automated lines, but things were just as chaotic as they ever were in the plant.

So here we are today, having seen the science of computer services grow, and we are about to witness CIM become a reality. In many ways the large computer services companies like CSC have started to institute CIM-like services in the nonmanufacturing sector.

For example, over the past two years we have provided the federal government with several fixed-fee, one-time projects which included: computer hardware, software, communications and operations -an entire integrated system. This is a significant switch from the traditional cost-plus pure computer services approach we have provided over the years. It is this type of turnkey system the leaders in the computer services industry hope to bring to the manufacturing world.

Large manufacturers who have lots of computer equipment and thousands of machine tools, presses, welders, robots and automated assembly lines need help in making their machines work together in a comprehensive, precise (yet flexibly scheduled), and computer-driven system. First the computers must be linked together and be capable of passing data back and forth. Then, that network must be able to communicate commands to a second network made up of people and machines in an on-line, real-time fashion that meets the demands of the manufacturer’s customers.

I believe CIM is an idea searching for a definition. If you read any of the current material being written about the idea, you’ll see what I mean. What I have decribed above is really the result of CIM or is the demand which CIM will fulfill.

Systems integration experts define CIM by listing its components as follows:

·               MRP II: Manufacturing Resources Planning

·               DRP: Distribution Resources Planning

·               EDI: Electronic Data Interchange

·                CAD/CAM: Computer Aided Design/ Computer Aided Manufacturing

·               CAE: Computer Assisted Engineering

·               CNC: Computer Numerical Control

·               SPC: Statistical Process Control

·               JIT: Just-In-Time

·               Bar coding

·                Programmable controllers

·                Robotics

·                Automated warehousing

These components are not competitors. They are members of the CIM team and are all needed to make the system work. CIM is a radically new way to run a manufacturing business. It requires a holistic approach. Its main concerns are total cost, total throughput time and total quality. These concerns must be shared by all departments of a corporation attempting to implement CIM. This is why systems integration is so vital to the success of this leading concept in what I call the NeoIndustrial Revolution.

Two final thoughts about integration. CIM facilitates three segments of business that have traditionally been autonomous to work together for the first time: engineering, manufacturing and marketing. CIM not only links industrial processes together but creates a linkage from vendors to manufacturer to customers which can be an irreplaceable competitive advantage far outweighing price and unique product characteristics.

A few brief words of caution. In researching this article I asked our CSC experts to find a few examples of CIM. Like old Mother Hubbard, they found the cupboard bare. Not because CIM does not exist, but there are only bits and pieces of it at work in our nation’s plants. We all know that IBM, DEC, Xerox and Hewlett-Packard use very highly sophisticated manufacturing methods in their plants, (i.e., MRP II and JIT). We know that Parker Hannifin has a bridge between its MRP II and CNC systems. In the 100 or so user plants that have our MAN-FACT II software, we know several with CAD links. This points toward the very real need for systems integration to take the parts and put them into a total company operating loop.

Is there hope? I firmly believe there is, but CIM will not arrive full-blown tomorrow morning. We must find the proper bridges and gateways for data connectivity between machines, computers and software, and this requires generally accepted standards. A beginning can be seen in the fact that GM’s MAP (Manufacturing Automation Protocol) is based on the IEEE 802.4 standard. Although it is shop-floor oriented only, this standard does provide an open architecture for linking devices through the use of a token bus. It allows equipment to be linked over long distances such as those encountered in large factories. (Even though this is a breakthrough accomplished by our competitor, I mention it because it is a significant step on the road to CIM. And many others can use this same standard.)

A recent poll published by Manufacturing Forum shows a strong market for CIMrelated services. Respondents predicted that the share of the industrial automation market enjoyed by the major manufacturers of computers and software would decline from 70 percent to 51 percent from 1985 to 1995. Systems integrators who had nine percent of the market in 1985 would rise to 19 percent in 1995. And turnkey suppliers with 14 percent in 1985 would increase to 20 percent in 1995. It is in the systems integrator category that CSC plans to take a leadership position.

To prepare for this market we acquired CSC Compufact in 1986. This division is the creator and marketer of MAN-FACT II-an MRP II-based computer software providing a total manufacturing management system that is the cornerstone of CIM. We have set aside funds for additional acquisitions, including companies with systems integration experience with CIM (see list of CIM components above).

If some of my fellow CEOs are eager to launch into CIM, let me warn you in the words of the late 011ie Wight: “If management tries to automate existing informal processes, they likely will not get the improvement they expect. Without a firm grasp of what to build and when to build it, automation becomes a downhill run to oblivion.”

Since I opened this discussion with a thought from Rear Admiral Grace Murray Hopper, I shall close with another of her penetrating comments. It rings true for me because we are becoming more and more conscious every day that this is the Information Age and that systems integration or CIM is no more than the harnessing of all the information required to produce products that comply with the business plan of an enterprise. In that light, Admiral Hopper said: “We have failed to look at the value and priority of information. There are no criteria for the value of data. If someday information will be listed as an asset, I asked the IRS how they were going to depreciate it, since some is good overnight, and some for a thousand years.”


William R. Hoover is president and chief executive of Computer Sciences Corporation, a California-based independent professional-services company in the computer industry, and a leader in the development of computer-communications systems technology.