National Academies Press: OpenBook

Design and Analysis of Integrated Manufacturing Systems (1988)

Chapter: Material Handling in Integrated Manufacturing Systems

« Previous: Flexible Machining in an Integrated System
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 46
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 47
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 48
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 49
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 50
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 51
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 52
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 53
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 54
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 55
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 56
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 57
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 58
Suggested Citation:"Material Handling in Integrated Manufacturing Systems." National Research Council. 1988. Design and Analysis of Integrated Manufacturing Systems. Washington, DC: The National Academies Press. doi: 10.17226/1100.
×
Page 59

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS JOHN A. WHITE ABSTRACT A systems view of manufacturing is presented from a material han- dling perspective. A foundation for considering the next generation of material handling systems is established by first describing the role of material handling in manufacturing. The need for integrated systems is then considered. Subsequently, various aspects of designing, selling, specifying, and implementing integrated systems are treated. The subject of intelligent material handling is considered by focusing on its individual com- ponents: movement, storage, and control. Underlying this discussion is the assertion that the best material handling is no material handling. Finally, the status of the analytic techniques for treating material handling and the development needs for material han- dling are assessed. The paper concludes with the identification of a group of research and development tasks that can improve the analysis and design of integrated manufac- turing systems. INTRODUCTION In most U.S. corporations, manufactur- ing is not accorded the high prestige given to product design or product engineering. And within the manufacturing community, material handling is not highly regarded. The best and brightest people in industry are seldom given manufacturing assign- ments, and it is a rarity that one of these is assigned to material handling. The career path to the position of chief executive offi- cer neither begins with nor passes through material handling assignments; in fact, it is unusual to find it passing through a manu- facturing assignment. Although a vice pres- ident for manufacturing might exist in a U. S. corporation, it is quite unlikely that he or she will be a member of the board of directors. This can be contrasted with the frequent election of the vice president for marketing or finance to the board of direc- tors. 46 Not only has material handling failed to achieve recognition within the industrial community, but also it is seldom recognized by the U.S. research community as a bona fide research field. Since material handling is treated as a critical link in the manufac- turing chain in Europe and Japan, it is not surprising that technological leadership in the field is to be found there. Although the technology for the automated guided vehi- cle originated in this country, the major advances over the past decade have been made by European firms. Major innova- tions in the design of such items as lift trucks, conveyors, and automated storage and retrieval systems also have come from outside the United States. Despite this long neglect of material han- dling, there is reason for optimism. In- creased attention is being given to it, and attitudes toward it are changing. It is now recognized that improvements in material

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS handling must occur if manufacturing ca- pability is to improve. Improvements will be critical in reducing inventories, improv- ing quality, reducing cycle times, increas- ing productivity, and lowering costs. Another reason for a changing attitude regarding material handling is an improved understanding of its role in integrated man- ufacturing systems. Once defined simply as "the handling of material," a systems view of material handling leads to a definition of "using the right method to provide the right amount of the right material at the right place, at the right time, in the right se- quence, in the right position, in the right condition, and at the right cost" (Tompkins and White, 1984, p. 166~. Because a mate- rial handling system cuts across cost centers and departmental boundaries, it functions as an integrating agent for manufacturing. For this reason, many consider the mate- rial handling system to be the systems inte- grator. In the discussion that follows, the "sys- tems view of manufacturing" is presented by addressing the definition of an inte- grated system, the need for integrated sys- rems, 1nclualng their aeslgn, specification, and implementation, the various aspects of automation, and finally material handling. The movement, storage, and control of ma- terial are examined in conjunction with in- tegrated manufacturing systems. The paper concludes with a critique of the analytic tools most commonly used in designing ma- terial handling systems and an identifi- cation of development needs for future- generation manufacturing systems. . . .. . . . WHAT ARE INTEGRATED SYSTEMS? A key issue that must be addressed is the definition of an integrated system. The term has been used for decades, yet considerable confusion exists regarding what is and what is not an integrated system. At the heart of the issue is the meaning of a system, which is defined here as a set of objects or ele- 47 meets, with relationships between them or their attributes, organized in such a way as to achieve a predetermined objective as they interact within their environment. The elements of a system can be con- nected loosely or tightly. Although they can operate independently, so long as collec- tively they achieve a predetermined objec- tive, an integrated system is generally one that is tightly connected. Put another way, the individual elements in an integrated system are synchronized. The connections can involve physical linkages by means of hardware or information linkages by means of computers and humans. The emergence of computer-integrated manufacturing (CIM) has increased the at- tention given to the use of integrated sys- tems. Unfortunately, much of the promise of CIM, though great, has not been real- ized. Some firms are advertising they sell CIM; others promote themselves as the turnkey CIM suppliers. With each playing a different game and using different rules, the only thing they have in common is the use of the same label (CIM). A term-by-term examination of CIM, in reverse order, is useful. First, CIM applies to manufacturing. In this context, manu- facturing is defined sufficiently broadly to include design and operating systems, as well as production and distribution activi- ties. All value-added and support functions for the manufacturing operation are in- cluded, as are both direct and indirect labor. The scope of CIM can be described best by a two-dimensional matrix. In the verti- cal dimension it includes the functions of product, process, and schedule design. In the horizontal it includes production, as- sembly, material handling, packaging, pur- chasing, quality control, production con- trol, inventory control, maintenance, and distribution. CIM includes both planning and execution. Second, CIN4 depends on integration. The synergistic benefits of systems integra-

48 tion that result in 2 plus 2 being greater than 4 is one of the promises of CIM. How- ever, integration must be defined very broadly; it includes the interface and coor- dination of functions, the linkages of phys- ical components, and the information hand- offs that occur, both vertically and horizon- tally, throughout the entire organization. The breadth of integration requires more than the integration of two or three ma- chines or workstations or the integration of two or three departments. CIM, when im- plemented to its fullest, will incorporate the entire manufacturing enterprise, including multiple production plants, suppliers, and customers. Third, CIM depends critically on the use of the computer to perform the integrating function. Although it might include the physical integration of hardware compo- nents, integration of the information sub- systems is essential to CIM. It is worth noting, perhaps, the attri- butes that are not required of a CIM sys- tem. The production processes can be labor-intensive in a CIM environment, so long as they are integrated by means of the computer. One can achieve CIM without automating production or material han- dling, and one can achieve integration without the use of the computer, as, for example, in the Toyota just-in-time JIT) system that relies on kanbans. Many observers believe it is necessary for computer integration to occur using a highly centralized software architecture. Computer integration can be achieved by networking microcomputers, and informa- tion linkages can be performed using a combination of people and computers. THE BARRIERS TO CREATING INTEGRATED SYSTEMS The need for integrated systems in manu- facturing has long been recognized (White, 1982~. However, few truly integrated sys- tems have been installed. Instead, automa- tion islands, information islands, and even lOHNA. WHITE organizational islands have been formed. Some of the more common reasons for this follow (White, 1986b). First, it is not easy to design and imple- ment integrated systems. A detailed under- standing of systems requirements and inter- actions must exist; many details must be considered; nothing can be left to chance. System complexity increases exponentially with the number of operations. Second, designing and implementing in- tegrated systems is a radical departure from tradition. Since the early 1900s, the ap- proach to designing a system has been to divide the system into its basic compo- nents e.g., operation, inspection, trans- portation, storage, and delay and then to analyze the system in terms of these com- ponents, to use a division of labor, to design hierarchical organizations, and to form or- ganizations into cost centers. To create an integrated system design from these many elements is not only difficult but has often proved to be impossible. Third, there are few apparent rewards in designing and implementing integrated sys- tems. Managers are frequently responsible for managing individual organizational units "by the bottom line." Few, if any, incentives exist to use team or systems ob- jectives. Despite claims to the contrary, in- dividual rather than team performance is frequently rewarded. Fourth, "insurance" is preferred, often in the form of inventories, additional space, redundant equipment, and excess person- nel. Minimizing risks seems to be preferred to maximizing gains. Thus, high inventory costs are often preferred to costs of stopping production. In addition, many believe a highly integrated system has no cushion for error; they believe a finely tuned "inte- grated engine" will experience considerable downtime (White, 1986c). Fifth, organizational barriers must be overcome. Organization charts tend to cre- ate boundary lines. Many individual man- agers are reluctant to attack the overall sys-

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS tern when they are being evaluated on the performance of the individual segment. Sixth, the concept of the integrated sys- tem is not well understood. It has different meanings for different people, depending on their background and experience. To a hardware supplier, manufacturing systems are integrated if they fit together physi- cally; to a computer supplier, they are in- tegrated if the information systems share a common data base and provide real-time control; to a design engineer, integrated systems represent the combination of prod- uct design, process design, and schedule de- sign (White and Apple, 1985~. Each group views system integration differently, and each believes that system integration has been achieved by satisfying their view. Each frequently tends to think of an integrated system as a pipeline, rather than as a net- work of pipes. Consequently, integrated systems are frequently viewed as tightly connected, inflexible, and risky. Seventh, the achievement of an inte- grated system requires a champion. A strong leader with a strong commitment to an integrated approach is necessary. Eighth, few success stories and numerous horror stories exist. As Skinner noted, "The new, computer-based 'total systems' ap- proaches to production management offer the promise of new and valuable concepts and techniques, but these approaches have not overcome the tendency of top manage- ment to remove itself from manufacturing. Years of development of 'the factory of the future' have left us each year with the promise of a great new age in production management that lies just ahead. The promise never seems to be realized. Stories of computer-integrated manufacturing (CIM) and new automated equipment dis- asters are legion; these failures are always expensive, and in almost every case man- agement has delegated the work to experts" (Skinner, 1985, p. 55~. Despite the highly publicized systems that have been success- fully implemented by some leading firms, 49 the belief continues to exist that systems in- tegration is expensive, risky, and complex. Ninth, resources are often limited. In ad- dition to a scarcity of capital, a critical shortage of qualified people frequently ex- ists. The systems integrator should possess technical, management, and financial skills and an understanding of operational re- quirements. Among the technical skills needed is competency in developing control systems, for it is the control aspect that tends to determine success versus failure. Tenth, some individuals are threatened by the use of integrated systems. In many firms, direct labor represents less than 15 percent of manufacturing cost. Materials and indirect costs contribute about 50 per- cent and 3S percent, respectively. Because integrated systems are aimed at eliminating gaps in the manufacturing process, they will result in a reduction in the labor that is used to close those gaps. Indirect costs should be reduced through the integration. The evidence, however, is overwhelm- ingly on the side of the potential advantages for integrated systems. The promise contin- ues to be great, even though current expe- rience has not always been good. Computer hardware and software have been devel- oped that allow information to be provided accurately and in a timely fashion for deci- sion making. As a result, management's span of control has been expanded. Finally, integrated systems provide a mechanism to reduce the "fat" in today's manufacturing and distribution systems and to increase competitiveness . Aristotle is credited with the observation that the whole is greater than the sum of its parts. In the case of integrated manufactur- ing, this claim is frequently repeated but with little evidence to support it. The costs and benefits of the whole cannot be ac- counted for by summing the costs and ben- efits of the component parts of the system. An overall assessment of the benefits must be made if the financial justification is to be made convincingly.

50 DESIGNING INTEGRATED SYSTEMS In designing integrated systems, a deli- cate balance must be maintained in terms of the degree of risk one is willing to take with respect to leading-edge technologies versus proven technologies. A trade-off is required between the possibility of obsoles- cence of yesterday's technology and the pos- sibility that tomorrow's technology might not be available on time, might not func- tion as required, and might cost more than predicted. A concern with the risk of the unknown has perhaps been responsible for the lack of unproven technology in most integrated systems. A number of challenges face the designer of integrated systems. The system must be kept simple and "requirements-driven." The ultimate user must be involved in the design process. The various design func- tions must be integrated. The system must be designed to accommodate automation, especially when the automation may be in- troduced at a later time. Excess insurance must be avoided. Systems discipline must be achieved and a long-term focus must be maintained while remaining responsive to short-term needs. And finally, one must avoid the "not invented here" syndrome (White, 1986a). SELLING INTEGRATED SYSTEMS In selling integrated systems to corporate management, a number of lessons have been learned. Among them are these: De- sign the whole, sell the whole, and imple- ment the parts; be sure you know what you are doing; do your homework; thoroughly analyze the "do-nothing" alternative; in- volve the user; analyze thoroughly; sell aggressively; be realistic and thorough in estimating costs and benefits; ensure accountability; and find a champion (White, 1985). SPECIFYING INTEGRATED SYSTEMS For many, a significant change is needed in the way integrated systems are procured. .IOHNA. WHITE When a firm purchases an industrial truck, a conveyor, or a palletizer, detailed design specifications are used routinely. They are often written at the component level. How- ever, when an integrated system is pur- chased, greater emphasis must be placed in the specification on what, when, where, and why, rather than on how, who, and which. The objectives are seldom defined well enough to allow the use of detailed design specifications. As firms obtain expe- rience in implementing integrated systems and as increased standardization occurs in data bases, protocols, and communication interfaces, detailed design specifications will become useful in procuring integrated sys- tems. Until then, functional specifications will be the preferred procurement instru- ment. IMPLEMENTING INTEGRATED SYSTEMS Careful planning is required if an inte- grated system is to be successfully imple- mented. This includes installing and debug- ging the system, training employees, and auditing the system to ensure that it meets requirements and is used properly. A1- though it is important to perform top-down design, it is critical that implementation be bottom-up. Designing from the top will en- sure that the system is compatible with, and supportive of, the firm's business objectives. Implementing from the bottom up ensures that individual components can operate in- dependently. The top-down design ensures that the operations will be synchronized. As with most major changes, a transition from a segmentalist approach to an inte- grated approach will most likely meet with opposition. Hence, care must be taken to protect against premature rejection of the system. It is often the case that an existing system must continue to function while an improved system is being designed and in- stalled. A phased implementation plan is required to avoid interrupting production. It is also difficult to maintain a long-term

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS focus and simultaneously avoid making mistakes in the short term. Yet, these con- straints must be accommodated, since a "greenfield" system is seldom feasible. AUTOMATION'S REPORT CARD Automation, in the form of automated storage and retrieval systems, guided vehi- cles, and automatic identification, has been used extensively in production environ- ments for more than 20 years. Despite a lengthy record of proven successes, "horror stories" about the implementation of auto- mation continue to appear. Reports of schedule slippages of months and years are not uncommon slippages coupled with the doubling and tripling of cost estimates. Claims and counterclaims abound as to whether automated or conventional sys- tems are the best approach. Why is auto- mation such a controversial topic? A num- ber of observations may help answer this question (White, 1986a). First, the most significant lesson learned is that the primary benefit of automation is the systems discipline it imposes. An auto- mated storage and retrieval system (AS/RS) is a highly disciplined technology when compared with the conventional system. To the extent that the same level of discipline is obtained using more conventional ap- proaches, it is difficult to justify automa- tion. In fact, it is difficult for automation to compete against a well-managed, highly motivated work force. But it is also true that very few well-aged, highly motivated work forces exist in the United States. Second, the technology of automation should not be singled out for blame. Rather, the design, implementation, and use of the overall system should be examined before making a judgment of culpability. It is an uncommon exception when the technology will not work. It is more usual that the scope of the application was too broad, the wrong technology was adopted, or the technology was being applied incorrectly. Third, too many automated systems are 51 designed "for show, not for dough." Too little attention is paid to the tangible bene- fits derived from capital investments in au- tomation. There is a tendency to buy more hardware or software rather than to invest in a careful system design to ensure that the investment will pay dividends. As a re- sult, automation is often overpromised and underdelivered (White, 1984~. Fourth, rather than being designed to meet a set of basic requirements, system designs are often determined by existent or, in rare cases, anticipated technology. They are solution-driven systems instead of re- quirements-driven systems. Fifth, the real system requirements are seldom well defined. The supplier is often asked to do "free design" and quote a price for satisfying a user's needs, without those needs being specified quantitatively in terms of performance requirements. Sixth, the automation bottleneck is often software, not hardware. The control sys- tem is often designed with an excess of con- trol. Rather than err by undercontrolling the system, the designer tends to over- control. Instead of designing the control system on the basis of the maximum amount of information possible, it should be de- signed on the basis of the minimum amount of information required. Seventh, a new automation system be- comes the vehicle for solving tangential problems for example, purchasing, re- ceiving, shipping, accounts payable, ac- counts receivable, inventory control, and other functions. The result can be that a major software task is assigned to a firm whose area of expertise is represented by less than half the total effort. Suppliers are often expected to satisfy the users' glutton- ous appetite for sophisticated software. "Software-smart" suppliers typically are not "material handling hardware-smart," and vice versa. Eighth, the user often lacks the discipline to "freeze the design.' following the pro- curement decision. As a result, require- ments are often added while the system is

52 being designed. Design changes are both expensive and disruptive to the delivery schedule. Also, it is often the case that changes made during the design and imple- mentation process are "spur of the mo- ment" responses to a rare event. Ninth, the computer hardware to be used is frequently specified before the computing requirements are defined. As a result, it is frequently the case that the computer is undersized in the initial quotation and that the error is not discovered until the system is installed and fails to handle the require- ments. Tenth, unreasonable, unnecessary, and expensive requirements are placed on the system to "never fail!" Reliability, avail- ability, and maintainability values are specified arbitrarily, rather than by consid- ering the cost impact of purchasing redun- dant computers. The cost of having "full- system recovery" within x seconds of an unscheduled computer stoppage should be assessed before specifying the value for x. INTELLIGENT MATERIAL HANDLING Desirable attributes of a material han- dling system were previously given as the right method, amount, material, place, time, sequence, position, condition, and cost. A material handling system can also be defined as moving, storing, and control- ling material. Despite recent improvements in material control that have transformed it from a mundane task to a sophisticated, state-of-the-art activity, one must not lose sight of the fact that the best material han- dling is no material handling. Although material movement, storage, and control are often taken for granted, an idealized manufacturing system would be one in which material did not have to be moved, stored, or controlled. Although the ideal of completely eliminating material handling may not be possible, it is certainly true that handling less is best. More specifically, less material movement is best, less material JOHN A. WHITE storage is best, and less material control is best. Less Material Movement In what sense is "less material movement best"? For some applications, "less" might mean less frequently, less material, less dis- tance, fewer interruptions, less manual in- tervention, or less variability. The objective of continuous flow manufacturing is to transform discrete parts manufacturing into a nonstop process in which material moves steadily through the production process. In general, quality tends to decrease and costs tend to increase as material movement in- creases. For these reasons, moving material fess frequently is best. The successes of the Toyota Production System and JIT have resulted in the elimi- nation of much of the material inventory. As a result of a renewed focus on the amount of material being moved, the size of the unit load and the number of moves performed have both decreased. Under these conditions, moving less material has proved best. Another result of JIT is improved "pipe- line" management. Suppliers are locating production facilities closer to their custom- ers, and the distances between successive production operations are being reduced. The reduction of the amount of inventory in the "production pipeline" reduces overall costs and increases the ability of the pro- duction system to respond to changing re- quirements. It has become better under- stood that moving over less distance is best. To eliminate waiting, the speed with which material moves must match the speeds of the processes it feeds. In the past, little attention has been given to the mate- rial waiting for processing. Instead, the fo- cus was entirely on maintaining constant Production. It was acceptable to have large buffers of material, if this was necessary to ensure that machines would not wait for material. Parts shortages and processing in-

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS terruptions would not be tolerated, regard- less of the cost. In the identification and elimination of sources of variation in pro- duction, the interface of the material han- ~ing system with the production processes has proved to be a fertile area for improve- ment. It has been shown that moving ma- terial with fewer interruptions is best. Although U.S. manufacturers have avail- able vast arsenals of sophisticated technol- ogies, a high percentage of human activity in manufacturing is engaged in performing material handling. An analysis of the basic work elements performed by human oper- ators reveals that most could be categorized as material handling. Simple pick-and-place activities continue to be performed by peo- ple. For many firms, the impact of manual material movement on quality, productiv- ity, and cost is such that for them, moving less manually is best. A major impediment to improved mate- rial handling is the lack of standardization. Standard containers, pallets, tote boxes, cases, and cartons do not exist. Except for bar code standards adopted by the Depart- ment of Defense, the automotive, health, pharmaceutical, and meat-packing indus- tries, little standardization has occurred in the past 20 years. One example of a failure to standardize is the design of cases to hold size 303 cans, a standard can in the food industry. The Food Merchandisers Institute and the Grocery Manufacturers Association have identified 32 separate sizes of cases that are used to contain 24 of these cans (Sims, 1986). Few material handling systems actually handle material. Instead, they move, store, and control containers (pallets) in (on) which material is placed. Hence, the basic building block of a material handling sys- tem is the container or pallet being han- dled. A failure to standardize on this unit creates a wide assortment of rack openings, conveyor widths, and lift truck sizes. Be- cause the material handling system must be designed to move so many different con- 53 tainers, moving less differently is best. Stan- dardization would produce dramatic re- ductions in material handling costs. Less Material Storage Just as intelligent material movement is no material movement, so intelligent mate- rial storage is no material storage. Although it may not be possible to eliminate material storage, an intelligent material storage sys- tem should be designed with the objective of "zero storage." As with moving material, "storing less is best." This can mean less frequently, less material, less distance, less volume, less money, or less routinely. Storage is required when there are im- balances in the flow rates of successive op- erations. A balanced production system and a reduction in the number of parts, subas- semblies, and assemblies that require stor- age will reduce the requirements for mate- rial storage. As an example, if several products are identical before a particular production operation, such as painting, then any in-process inventory storage should occur before the painting operation. Parts standardization is another way of reducing the number of parts requiring storage. Stor- ing less frequently is best. The emphasis on storing less material in- cludes all types of material e.g., raw ma- terial, subassemblies, in-process material, finished goods, tooling, supplies, equip- ment, and other support facilities. Storing less material is best. Because of the need for improved re- sponse time in delivering materials to the point of use, a number of organizations are shifting from centralized storage to distrib- uted storage. Past arguments that central- ized storage reduced aggregate inventory levels are not necessarily valid in a real-time material control system; with real-time control, it is possible to manage distributed storage systems to the same degree that cen- tralized storage systems were once man- aged. By using distributed storage, material

54 is stored closer to the user. Hence, storing at less distance from (or closer to) the user is best. Material storage involves more than just storing material. Space is required for clearances, aisles, storage racks, and con- tainers. It is not uncommon in a well- designed pallet rack storage system for the material being stored to represent, on the average, less than 10 percent of the cubic volume of space consumed. As the cost of space increases, it becomes obvious that storing less volume is best. The cost of storing material tends to be underestimated. Too many believe the only cost of material storage is the opportunity cost associated with the cost of the material itself. Thus, as noted previously, invento- ries have become pseudo-insurance policies. More discriminate management is needed; increased differentiation is needed between high-cost and low-cost materials to reduce inventory investment. By recognizing that materials are fiscal resources, it is evident that storing less material means storing less money, and storing less money is best. In many cases, material storage has be- come a routine activity. In particular, the very existence of material storage has be- come a "given" when manufacturing sys- tems are being designed. When major cap- ital investments are made in automated storage systems, subsequent modifications of the production system are designed around the storage system; too often, the "sunk cost" invested in the storage system continues to influence future design deci- sions. Furthermore, the storage methods used are frequently routine; in many cases, manual storage methods are used, with subsequent inefficiencies in use of space, difficulty in maintaining accurate invento- ries, and excess time spent looking for ma- terial. As a result, it has been concluded that storing less routinely is best. Less Material Control Although many operators agree that the best material movement is no material JOHN A. WHITE movement and the best material storage is no material storage, few agree that the best material control is no material control. The trend in designing material control systems is to overcontro] rather than to undercon- trol. Since software has frequently become the bottleneck in implementing automated material handling systems, it is concluded that "controlling less is best." In this case, "less" means less centrally, less collectively, less complexly, or less frequently. In a highly centralized control system, the production system becomes extremely vulnerable to start-up delays, software "bugs," human resistance, and information overload. In the extreme, the centralized control system is expected to be omniscient, immutable, omnipotent, and veracious. To the extent it is not, difficulties arise. Thus, controlling less centraZ!y is best. Several years ago, it was recommended that different things be moved differently, stored differently, and controlled differ- ently. The objective of the recommendation was to recognize the inherent differences in movement, storage, and control require- ments when designing a material handling system. The differences in the physical characteristics make it obvious that an item weighing 1 ounce should be moved differ- ently than an object weighing 1 ton, and an ice cube must be stored differently than a diamond. Although many material han- dling systems do move and store different things differently, they seldom control dif- ferent things differently. One can find ex- amples of real-time control being imple- mented for an inventory consisting of more than 1 million part numbers when less than 100,000 of these experienced any activity over a 2-year period. The system should control material selectively, not collectively. Thus, many believe that controlling less collectively is best. As noted earlier, the tendency is to over- control rather than to undercontrol. Simi- larly, there is a tendency in designing com- puter control systems to make them overly complex. Complexity can occur with either

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS human or computer control. However, it tends to occur most easily when using com- puter control. Simplicity must be the hall- mark of control systems design; controlling less complexly is best. "The best material control is no material control" can mean that the best material control occurs when there is no need to con- tro} material. Eliminating the need for con- trol simplifies the design of the material handling system. Control is needed to deal with variances, by identifying and elimi- nating the sources of variance, the need for control systems is eliminated. System disci- pline should be achieved without the use of computer or human control. Also, combi- nations of human control and computer control are preferable to computer control alone. If the frequency with which control needs arise can be reduced, then controlling less frequently is best. MATERIAL HANDLING: ANALYSIS AND DEVELOPMENT We now review the status of analytic tools for material handling and the devel- opment needs for future-generation mate- rial handling systems. Both design and op- erational issues are considered. The Status of Material Handling Analysis As noted previously, relatively little re- search has been performed on material handling problems. Although little atten- tion has been given to a number of funda- mental design and operating issues, consid- erable effort has been devoted to the development of analysis tools applicable to the design and operation of material han- dlin~ systems. Among the various analytic tools com- monly used to facilitate the design of mate- rial handling systems, simulation is the most popular. Simulation is also being used to assist in the management of operating sys- tems. Recent advances in simulation mod- 55 cling have included animated output and the interface of simulation with computer- aided design (CAD). As a result, a model of a material handling system can be used to observe its simulated performance. The combination of color graphics, animation, and the CAD interface has transformed simulation into a powerful design, operat- ing, and marketing tool. Advances in the ability to solve large- scale simultaneous equations quickly have enhanced the feasibility of modeling ana- lytically large-scale material handling sys- tems. Recent work in queueing analysis has been particularly important in increasing computational capability. The relevant re- search also includes studies on closed queueing networks (Solberg, 1977; Suri, 1983), open queueing networks (Shanthi- kumar and Buzacott, 1981), the control of queues (Bares et al., 1985; Gonheim and Stidham, 1985), networks of queues (Whist, 1983a, 1983b), and mean-value analysis (Surf and Hildebrant, 1984; Hoyme et al., 1986~. Mathematical programming formula- tions of material movement and storage sys- tems have been developed. However, since most real-world problems result in non- linear, integer programming formulations, these formulations are usually solved using heuristic approaches. The design of a closed-Ioop conveyor network linking mul- tiple machine tools is an example of a situ- ation that is difficult to mode} and solve exactly, since it requires that decisions be made on the buffer sizes at each machine, the number and locations of "by-passes" or "crossovers" on the network, and the con- trol logic to use. A number of scheduling and sequencing decisions arise in operating a material handling system. Unfortunately, little attention has been given to the operation of such systems. For example, little work has been performed on determining the "optimum" sequence of storages and retrievals for a rotary rack with robotic loading and unloading. The same is true for a microload automated storage and

56 retrieval system. Even less attention has been given to determining "optimum" con- trol strategies for material handling subsys- tem combinations. Beyond the explicit con- sideration of material handling problems, there has been little attention to material handling in the formulations of related problems. For example, few formulations of scheduling problems have as their objec- tive the minimization of material move- ment and material storage. Unfortunately, there is little concern a priori for the impact of scheduling decisions on material han- dling. Although inventory control has received considerable attention from researchers, the problem formulations have often failed to incorporate a number of significant aspects of the inventory problem as it occurs in manufacturing. For example, the determi- nation of production lot sizes and unit load sizes, the use of centralized versus distrib- uted storage, and the conditions under which hitting should be used have not been addressed from a systems perspective. In- ventory formulations have tended to be too narrow in scope. In general, too much academic research has been "solution-driven" rather than "problem-driven." So-called applied re- search has been performed on contr~vect problems rather than on real problems. Rather than becoming intimately familiar with the problem and defining a research agenda to address it, the initial focus has been on the solution methodology. As a re- sult, the problem has been viewed from the perspective of the solution and formulated accordingly. In terms of material handling technolo- gies, little research has been devoted to the selection of the appropriate technology for specific applications. As an example, the use of asynchronous material handling equipment has become quite popular. Asynchronous alternatives include the use of automatic guided vehicles as assembly platforms and for performing general trans- port functions, "smart" monorails for mov- JOHNA. WHITE ing parts between workstations; transporter conveyors for controlling and dispatching work to individual workstations; robots for performing machine loading, case packing, palletizing, assembly, and other material handling tasks; microload automated stor- age and retrieval machines; cart-on-track equipment for moving material between workstations; and manual carts for low- volume material movement activities (Na- tional Research Council, 1986~. Choices are being made by the automo- tive industry, for example, between the use of power-and-free conveyors and auto- mated guided vehicles in assembling cars. Yet, no comprehensive models exist to sup- port such decisions. The decisions tend to be based- on intuitive appeal and anecdotal evidence. Electronics firms store compo- nents in carousels, in miniload automated storage and retrieval systems, in microload automated storage and retrieval systems, and in bin shelving. Although each of these has different performance characteristics, investment costs, and operating costs, little has been done to guide the material han- dling system designer in making selections among the alternative technologies and in configuring and operating each alternative optimally. - ' Experience has shown that it is difficult to develop economic models of technology selection decisions. Although it is easy to assess the economic impact of incremental changes in the design configuration, the sales price for each technology alternative is dynamic and is determined by prevailing market conditions, and the decisions are of- ten based on consideration of multiple cri- teria. Hence, prescriptive models of tech- nology selection decisions are, at best, aids to decision making. Even though some attention has been given to interstation handling, little atten- tion has been given to material handling at the workstation. As noted previously, a considerable amount of human activity in manufacturing is used to move, store, and control material. In many cases, the opera-

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS for's vision is used to locate a part and de- termine its orientation in order to grasp and position it for the next operation. In all like- lihood the orientation of the part was known at some previous point in the process. How- ever, either the orientation changed or the information was not captured. Hence, the operator must regain physical control of the part. The cost of regaining versus retaining physical control of material has not been adequately addressed. Although the need exists for automatic storage and retrieval of individual items, few equipment alternatives are available, and those that are available are not widely accepted. The technology void has existed for a number of years but does not appear to represent an area of current interest to material handling equipment suppliers. Material Handling Development Needs The assessment of the development needs for material handling is limited to the focus of this volume namely, the design and analysis of integrated manufacturing sys- tems. Hence, only those material handling technology gaps that affect the design of integrated manufacturing systems are iden- tified. The following list of development needs is organized into three general categories: material handling systems design needs; material handling interface needs; and ma- terial handling hardware and software needs. Systems Design Needs The following items relate to material handling systems design needs: · Engineering workstations for designing material handling systems · Expert systems for designing material handing subsystems · Preprocessors that will create simula- tion programs from material handling sys- tems designs 57 · Preprocessors that will create "opti- mum" control systems designs from simu- lation programs · Increased understanding of the per- formance characteristics of material move- ment and material storage technologies · Performance models for collections and combinations of material handling technol- ogies · A method of determining the ease or difficulty of moving, storing, and control- ling a particular part or product · Decision rules for retaining versus re- gaining the physical orientation of individ- ual parts · Decision support systems to assist the designer in determining the sizes and loca- tions of material storage points and the unit load size to be moved between workstations · Network generators for a variety of material movement alternatives, both syn- chronous and asynchronous Interface Needs The following items relate to the inter- face of material handling with product and process design, manufacturing systems, and shop-floor control: · The incorporation of material han- dling considerations in the decision support systems used in product and process design · The incorporation of material han- dling considerations in the formulations of manufacturing systems models · Integration of distributed material handling control with shop-floor control systems · Human supervisory control systems for distributecl, automated material handling systems Hardware and Software Needs The following items acldress hardware and software needs: · Automated material handling systems that recover automatically from significant disruptions

58 . Modular and flexible material han- dling equipment for use in moving and stor- ing a variety of components and products · Direct identification technologies · Automated storage and retrieval sys- te~ns for individual items · Path-free automated guided vehicles · Container and hardware interface standards Within the first group of development needs, the targeted result is the develop- ment of an engineering workstation for use in designing material handling systems in manufacturing and distribution. To accom- plish the objective, expert systems must be developed. The development of expert sys- tems will be aided by the emergence of pre- processors to create simulation programs from CAD-generated material handling systems and to create the control system directly from the simulation program. Before expert systems can be developed, there must be increased understanding of the performance characteristics of material movement and material handling technol- ogies. To gain that understanding, per- formance models must be developed for logical collections and combinations of ma- terial handling technologies. Among the decision support systems needed in designing material handling sys- tems are those that provide assistance in determining unit load sizes, material stor- age locations, buffer sizes, material flow paths (networks), and positional control re- quired for parts movement. A major impediment to developing ex- pert systems for designing material han- dling systems is the paucity of metrics for gauging the ease or difficulty of moving, storing, and controlling parts of various de- signs and configurations. "Design for han- dling" research similar to that performed in support of "design for assembly" is needed. As noted previously, because of the serial nature of manufacturing systems design, it is generally the case that few degrees of JOHNA. WHITE freedom remain for the material handling systems designer after the product and process designers have completed their work. One solution to the problem is to provide the product and process designers with design tools incorporating material handling considerations. The second cate- gory of development needs addresses this issue. In addition to the need to interface ma- terial handling design decisions with pro- cess and product design decisions, it is also important to incorporate material handling considerations in scheduling algorithms, shop-floor control systems, and inventory control models. Finally, any consideration of material handling interfaces must include the hu- man interface. In 1987 the human is still the predominant material handler in indus- try. However, that is not the interface issue addressed here. Instead, we are concerned with the human operating as a supervisor for the automated material handling system in a highly distributed environment. Among the hardware and software de- velopment needs are the need for automatic recovery, automatic identification, auto- matic ranging and guidance, and auto- matic storage and retrieval of individual items. In addition, a critical need exists for the development of standards for containers and hardware in material handling. A RECOMMENDED APPROACH With the material handling development needs defined, what approach should be taken? The following five recommenda- tions are given as research guidelines. First, a problem-driven approach is rec- ommended. Hence, a joint industry- university research team is needed. The team should consist of individuals who un- derstand the problems, individuals who un- derstand the methodologies, and those who can function as bridges between these two groups. The research should be conducted

MATERIAL HANDLING IN INTEGRATED MANUFACTURING SYSTEMS on factory floors and in laboratories rather than in offices and conference rooms. Second, a cross-d~sciplinary approach is needed. The development needs listed will require the expertise of computer scientists, electrical engineers, industrial engineers, mechanical engineers, and systems engi- neers, among others. Third, the research should be results- oriented. Specifically, a particular devel- opment should be targeted as the end-item deliverable from the research. Throughout the research, the targeted goal should be kept in mind. Fourth, the first end-item deliverable should be the engineering workstation for designing material handing systems. The first ten development needs listed should provide a starting point for the research team. However, it is likely that additional needs will be identified in the conduct of the research. Fifth, the research should be generic for discrete parts manufacturing. However, to facilitate the research it will be helpful to focus initially on a particular scenario. Ex- perience indicates that the material han- ~ing problems encountered in industry are quite similar, despite the differences that exist in materials, tooling, processes, and levels of technology employed. REFERENCES Baras, ]. S., A. l. Dorsey, and A. M. Makowski. 1985. Two competing queues with linear costs and geo- metric service requirements: The tic-rule is often optimal. Advances in Applied Probability 17:186. Gonheim, H., and S. Stidham, Jr. 1985. Control of arrivals to two queues in series. European Journal of Operational Research 21:399. Hoyme, K. P., S. C. Bruell, P. V. Afshari, and R. Y. Rain. 1986. A tree structured mean value analysis algorithm. ACM Transactions on Computer Sys- tems 4~2~:178. 59 National Research Council. 1986. Toward a New Era in U.S. Manufacturing: The Need for a National Vision. Manufacturing Studies Board. Washington, D.C.: National Academy Press. Shanthikumar, ]. G., and ]. A. Buzacott. 1981. Open queueing network models of dynamic shops. Inter- national Journal of Production Research 19~3~:255. Sims, J. R. 1986. Food distribution: A material han- dling opportunity. Material Handling Users Con- ference. Atlanta, Gal: Georgia Institute of Tech- nology. Skinner, W. 1985. Manufacturing: The Formidable Competitive Weapon. New York: Wiley. Solberg, I. I. 1977. A mathematical model of comput- erized manufacturing systems. P. 1265 in Proceed- ings of the 4th International Conference on Produc- tion Research, Tokyo. Shari R. 1983. Robustness of queueing network for- mulas. Journal of the Association of Computing Machines 30~3~:564. Suri, R., and R. R. Hildebrant. 1984. Modeling flexi- ble manufacturing systems using mean-value anal- ysis. Journal of Manufacturing Systems 3~1):27. Tompkins, J. A., and J. A. White. 1984. Facilities Planning. New York: Wiley. White, J. A. 1982. The automated factory and inte- grated systems in the 80's. Proceedings of the 4th IIE Managers Seminar. Norcross, Gal: Institute of Industrial Engineers. White, T. A. 1984. Design for automation. Modern Material Handling 39~1~:29. White, J. A. 1985. Selling integrated systems. Modern Material Handling 40~3~:29. White, J. A. 1986a. Becoming the systems integrator by the year 2020. P. 372 in Proceedings of Fall Industrial Engineering Conference. Norcross, Gal: Institute of Industrial Engineers. White, J. A. 1986b. Impediments to system integra- tion. Modern Material Handling 41~10~:23. White, I. A. 1986c. Time to re-evaluate your insur- ance. Modern Material Handling 41 (7) :25. White, I. A., and J. M. Apple, Jr. 1985. Material handling requirements are altered dramatically by CIM information links. Industrial Engineering 17~2~:36. Whitt, W. 1983a. Performance of the queueing net- work analyzer. Bell Systems Technology Journal 62~9~:2817. Whitt, W. 1983b. The queueing network analyzer. Bell Systems Technology Journal 62~9j:2779.

Next: Designing an Information System for Integrated Manufacturing Systems »
Design and Analysis of Integrated Manufacturing Systems Get This Book
×
Buy Hardback | $55.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Design and Analysis of Integrated Manufacturing Systems is a fresh look at manufacturing from a systems point of view. This collection of papers from a symposium sponsored by the National Academy of Engineering explores the need for new technologies, the more effective use of new tools of analysis, and the improved integration of all elements of manufacturing operations, including machines, information, and humans. It is one of the few volumes to include detailed proposals for research that match the needs of industry.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!