Chapter Eight PACE OF TRANSITION Constant improvement has characterized the evolution of numerical toolmaking. Upheavals have been commonplace as process refinements have been explored and implemented. This dynamic nature of numerical control and computers makes it especially difficult to predict just how fast the transition will take place affected as it will be by so many imponderables. The anticipated pace of these future developments can at best be founded on cautious conjectures based on seemingly relevant factors. Paradoxically, several of these factors act as stimulants in some applications and deterrents in others. THE EFFECT OF FLUCTUATIONS IN THE ECONOMY The pace of the transition to numerical toolmaking in Michigan is surely influenced by economic growth and its concomitant effect on car sales. This pace will speed up during periods of intense sales and drop off as sales abate. The acquisition of numerically controlled equipment and the hiring of personnel trained in the new techniques trailed off when auto sales dropped in the first quarter of 1967. Several companies reported that management authorizations to acquire numerical control had been cancelled before orders could be placed. The cutbacks were part of across-the-board austerity programs. Sales patterns from earlier years suggest that the 1967 pause was not unprecedented. Both the 1954 and 1958 recessions saw lower sales of machine tools. During 1958-an unusually poor year for the auto industry, which buys about 40% of the nation's machine tools-the sales of the machine tool industry fell to $411 million, and the industry's contribution to the gross national product dropped to one-third of what it was during the Korean War. Fluctuations in the economy will have an even greater impact on the acquisition of numerical control by the independent tooling firms. The inordinately high cost of numerical control dissuades small firms from gambling on a new process unless the economic outlook is especially bright. Although the prices for control systems are dropping, and less sophisticated, cheaper machine tools are now being offered; the average prices reported by the U.S. Department of Commerce in 1964 for numerically controlled equipment were: $31,000 for a drill; $100,000 for a boring machine; and $250,000 for a milling machine. In 1966 the average price of numerically controlled machine tools sold by members of the National Machine Tool Builders Association was $74,000, almost four times higher than the average price of $19,000 for all the other machines. These equipment prices would be unprecedented for tooling firms which have been stretching capital by purchasing secondhand machinery. An analysis of the investment in fixed assets of 18 large tool and die firms, all employing over 100 workers, illustrates the severity of the financial problem. The 18 firms were divided into three categories according to sales volume: under $3 million, $3 million to $5 million, and over $5 million. Even though these firms are among the largest in the industry and their investment in plant and equipment is uncommonly high, their average total fixed assets compared to a $300,000 numerically controlled machine are: 30% for Group I, 28% for Group II, and 46% for Group III. Large working capital resources are also required because of the long production cycles in the tooling industry. While progress payments are received by some well-established firms, nearly all rely on short-term financing to support their work-in-process. Their borrowing potential is so frequently used for this purpose that there is little opportunity to finance the purchase of costly new equipment unless a substantial downpayment can be amassed. The long-term business outlook must be especially bright for the firm and its banker to gamble on an additional loan for equipment. As a further complication during periods of intense business activity, banks charge higher interest rates because of the greater demand. The payback period of sophisticated machinery spanning only two to four years, the fluctuating business conditions, and the higher interest rates make the industry's standard financing practice difficult to justify for numerical control acquisition. These companies are left with few alternatives, however, because additional equity capital is difficult to raise for these family-held companies. The purchase of even one machine strains the finances of the large firms and may be beyond the reach of many medium-sized and small shops. The large and medium-sized firms also need more than one numerically controlled die mill or milling machine. Among the firms still searching for financial stability seven or eight years after the recessions of the 1950s, the prospect of such massive capital investments is unnerving. Clearly, the major factor needed for tooling firms to implement numerical control is money—and lots of it by their standards. The financing assistance provided by banks, machine tool builders, and concerned governmental agencies will strongly affect the pace of the transition to numerical toolmaking. The implementation of numerical control will fluctuate more widely than that of EDM with changes in business activity. Numerically controlled equipment is required by many firms to extend capacity. EDM, on the other hand, is often purchased to produce hard-to-machine parts. During recession periods, when capacity is rarely expanded, EDM may still be the only solution to a production bottleneck. THE EFFECT OF SKILLED LABOR SHORTAGES Economic growth during recent years has expanded the demand for tooling and created a shortage of skilled workers. This shortage variously accelerates and impedes the introduction of numerical control. The Department of Labor reported that the tool and die industry averaged almost a 49-hour workweek in May, 1966; the average for all manufacturing was only 41.5 hours. Many Michigan shops were on a 58-hour schedule for much of the year. Some firms, realizing that numerically controlled machines can be operated by semiskilled workers, installed the machines to compensate for the shortage. As they were able to place semiskilled operators on the second and third shifts to run numerically controlled machines, both worker productivity and the output of finished parts increased through its 24-hour a day operation. Although the shortage of toolmakers and machinists encouraged many firms to buy numerically controlled equipment, at least one captive shop had to curtail additional equipment orders for want of qualified maintenance men. Numerically controlled equipment must usually be operated around-the-clock to redeem the investment; therefore it may cut parts 15 to 18 hours a day, or three to four times as long as conventional machines. The loss of such a machine because of a malfunction is an equivalent loss of seven to nine conventional machines operating on a single shift. Understandably, maintenance programs must be very thorough to keep these new machines operational. Maintenance personnel, who previously serviced conventional electric wiring with its relays and circuit breakers, are often bewildered by problems in the photoelectric tape readers, the transistorized control logic, the power supplies, the solid-state circuits, or refined servomechanisms. The aerospace industry, which averages 0.53 electronic and 0.45 mechanical and hydraulic maintenance personnel for every numerically controlled machine, has had to institute new maintenance methods to build a staff capable of coping with the electronic and servomechanism problems. In the search for a more technically qualified maintenance staff, job classifications have been upgraded. Aerospace personnel managers frequently recruit military veterans with training in electronics and adapt their knowledge in special, concentrated training programs. Despite upgraded classifications and special training schools, a skills shortage still exists. The automotive manufacturers also improved their maintenance training program, but as the emergency summonses to vacationing electronic maintenance technicians indicate, the training efforts have been only partially successful. Advancement has been impeded by two frustrating elements: numerical control, like most forms of automation, has reduced the skills required of operators and has increased those of maintenance men; the automotive industry has also found it difficult to recruit qualified maintenance trainees and to realign worker pay scales. Machine operators in captive shops, who now only supervise numerically controlled machining, are still paid higher wages than maintenance personnel. In contrast, we found that 14 firms in the aerospace industry paid maintenance men an average of 23 cents an hour more than their machine operators. Union contracts require automotive companies to recruit trainees from the existing maintenance staff, restricting the acquisition of the new talent necessary to cope with solid-state circuits. The largest user of numerical control and one automotive firm reported that two out of three members of their maintenance staffs fail to pass the tests on the minimum qualifications for admission to numerical control maintenance training programs. Seniority rights which protect the more experienced workers also make it difficult to retain younger maintenance men hired for their knowledge of the new equipment. Because of these problems the automotive industry is having difficulty limiting machine downtime to the 5-10% level of the aerospace industry. Poor equipment utilization has reduced possible profits and discouraged some from expanding their numerical toolmaking until trained men are available. Some captive shops have contracted with the equipment builders for the maintenance service. To allow outsiders in areas where new styles are being developed is unusual and indicates the seriousness of the shortage. New control systems using integrated circuits are being introduced which may circumvent some of the maintenance problems. Besides being lighter and more compact, the new systems are more reliable. Experience with computer equipment has shown that a logic circuit using silicon chips in place of conventional transistors and resistors improves reliability about ten times. Numerical control also requires a staff of skilled programmers. The aerospace industry has averaged 0.72 programmers for every numerically controlled machine. The inadequate supply could have been more sparse if, as in maintenance apprenticeships, the training of new programmers were subject to union approval. The extent of union control is still disputed; e.g., it was an issue in the 1967 automotive strikes. The few decisions by umpires judging union-management disputes have not as yet established a pattern. The spread of numerical control, however, will be restrained by union practices; to what extent is not yet discernible. Until the supply of programmers catches up with the demand, a rapid switch to numerical control will be inhibited. Many managers are slow to recognize this condition in the labor market and frequently wait too long before authorizing funds for the training of their own staff in part programming. Some managers have committed as much as $350,000 for numerically controlled equipment without allocating an additional 1-2% to train their staff. While our study did not disclose any company stymied by a lack of part programmers, many openings exist and much job hopping occurs.1 There are also presently unfilled jobs for 50,000 computer programmers who provide essential support to numerical control part programmers. The programmer shortage also limits the use of computers for experimental projects like computer-aided design and computer-graphics-both of which are closely related to numerical toolmaking. The need for a more efficient programming language for automotive die applications also deters the spread of numerical toolmaking. The complexity of programming languages for non-conventional geometric shapes has deterred some potential users from acquiring numerical controlwhich may indirectly have kept the demand for programmers from completely outstripping the supply. THE EFFECT OF PROFITABILITY The early users of numerical toolmaking were disappointed with their reductions because of the unexpected, and sometimes bewildering, start-up problems. More recently, improvements in equipment and methods have greatly improved profits. Still, many job shops are waiting even now for the publication of more concrete and convincing examples of the economic use of numerical toolmaking. Since users of numerically controlled equipment are circumspect in discussing savings, the financial information available, including that gathered by this study, was insufficient for an explicit economic analysis of current numerical toolmaking in one-of-a-kind operations. The misgivings of the skeptical independent firms must be dispelled with concrete economic data if the pace of transition outside the captive shops is to quicken. 3 Besides being reluctant to release the cost information, automotive companies do not judge the economic benefits of numerical control in the same fashion as the independent shops who are less concerned with product design. The automotive companies evaluate numerical toolmaking for its effects on both the time and costs of the total model changeover cycle. The Mustang and the Edsel show that swings in consumer style preferences so affect profits that cost evaluations which isolate toolmaking in the total changeover cycle are incomplete. Market research information available just prior to the Edsel's intro 1 One company reported the loss of ten part programmers who received a 75% increase in salary from their new employer. 2 Many exceptionally skilled programmers believe that the part programming languages available today are adequate, but are just too complex for the average programmer to use. 3 One captive shop reported that the advent of numerical toolmaking is forcing the standardization of die design to assure that one-of-a-kind items are minimized. This has already resulted in tapes to machine such items as supporting members for the tooling of a 1967 model to be useable on the tooling for 1968 and 1969 models. Such a practice will help the captive shop, but may not aid the independent unless he can be sure that the repeat order will be given to him. |