Decision

changing scenario in manufacturing s mitra* and a k chatterjee+ this paper traces the development of new approaches to manufacturing management such as mrp/mrp ii, erp, jit and opt during the last two decades. the evolution of these approaches has been discussed in the context of historical development in manufacturing starting from early in this century. earlier approaches typically are “reductionist” in nature, in that they assume that the problem of the entire system can be solved by solving efficiently each of its sub-systems in isolation. but with time, these approaches failed to meet the expectations, and gave way to the adoption of more recent “holistic” approaches like erp, jit and opt, which offer solutions by taking an integrated view of the system. having discussed the new techniques, an attempt has been made to outline the future of manufacturing management, where it is argued that it could possibly evolve to a hybrid system with the systems view of the recent approaches and the optimisation capability of or techniques.

1. introduction the last two decades have witnessed the evolution of new approaches to manufacturing management, illustrated by the growing body of literature in this field. material requirement planning (mrp), manufacturing resource planning (mrp ii), just in time (jit), optimized production technology (opt), and enterprise resource planning (erp) are some of the approaches the companies are adopting to achieve manufacturing excellence. a survey of american industries done by newman and sridharan (1992) shows that out of 185 firms from different manufacturing industries ranging from machine tools, automobile components, furniture, plastics, and medical equipment to computers and defense electronics, 56% were using mrp systems, while 22% were using reorder point (rop) systems. about 8% reported using jit, while 5% used opt systems. 9% of the firms reported using their own in-house developed systems. in india too, these techniques have been successfully implemented by various organisations. jit has been implemented in tvs suzuki, crompton greaves, maruti udyog, eicher goodearth limited, godrej and boyce manufacturing company, hero honda, siemens limited, sundaram clayton, tisco and many other companies (padukone and subba rao, 1993). sundaram fasteners limited and sundaram brake linings, two tvs group companies, have implemented jit, and are moving toward lean manufacturing systems with supply chain integration, cellular manufacturing and kaizen, or continuous improvement. sundaram fasteners limited has introduced total productive maintenance (tpm) in 1995, and it is the first indian company to get iso 9000 (suresh, 1998 and sridharan, 1998). more recently, companies are going for erp solutions. companies like hindustan lever, * fellow student, operations management, iim calcutta + professor, operations management, iim calcutta 1

telco, arvind mills, mahindra and mahindra, ranbaxy laboratories, and larsen and toubro, among others, have opted for erp. the objective of this paper is to put these approaches into perspective by examining them in the context of the historical developments in manufacturing management starting from the beginning of this century to date. an overview of manufacturing problem solving is first presented in section 2. this is followed by a presentation on the new approaches in section 3. in section 4, the new approaches are examined and an attempt to identify the future scenario in manufacturing is made. the concluding remarks are presented in section 5. 2. history of manufacturing manufacturing, by definition, involves transforming inputs into a desired set of outputs. various types of problems ranging from forecasting of demand to planning for production and distribution arise while managing the inputs, outputs, and the transformation process. different approaches to manufacturing management have evolved over time to find solutions to such problems. the manufacturing scenario during this century has also been marked by increased degree of automation in the shop floor. transition from manual operations to transfer machines to numerical control (nc) to flexible manufacturing systems (fms), and finally to computer integrated manufacturing (cim) depicts the evolution of the automation process. f. w. taylor’s contribution, manifested by the scientific management movement at the beginning of this century, provides a landmark in the history of manufacturing problem solving. manufacturing environment at that time was mainly labour intensive. taking time as a measure of efficiency, taylor contributed to the development of the concept of “standard time”, which laid the foundation of planning in the context of manufacturing. “work study”, as it was popularised later, thus consisted of measuring, and also simplifying tasks by critical examinations of the same. taylor’s scientific management had a wider perspective, where the basic objective could be summarized in terms of determining the most efficient way of doing any task. the variation of the efficiency objective in terms of time, cost and productivity soon percolated to different areas in manufacturing. this was manifested in terms of development of different inventory management and quality control models during 1910-1940. a departure from the efficiency objective was, however, noted during 1927, with elton mayo coming out with human relations approach focussing on the role of workers in increasing productivity. the theory of socio-technical system, propounded by miller and rice (1967) reinforced this further by advocating the dovetailing of task needs with social needs of the workers. during 1935-40 operations research (or) came into existence for solving military problems. or, as a mathematical modelling approach, soon found its application in manufacturing problem solving. methodologies available before or had their foundation in mathematics and probability theory that were fairly developed at the beginning of this

2

century. however, the applications were limited to relatively simple systems. for example, economic order quantity (eoq) model for purchasing decisions, statistical quality control (sqc) model for quality inspections were solved using calculus and sampling theory respectively. production planning and distribution problems, on the other hand, were ill solved, and or, by 1950, provided the requisite methodology for formulating and solving these complex problems. with commercialization of computers during 1951, it was possible to exploit or methodologies to a greater extent. this also made possible solutions of large scale manufacturing problems. thus, from the beginning of this century till about 1960, manufacturing management was dominated by piece-meal applications of work study, or, and related techniques catering to the problems of different subsystems. the environment, on the other hand, changed from labour intensive to fixed automation (eg. transfer lines) to programmable automation (eg. nc). the fixed and programmable automations were characterized by high volume and low to medium volume production respectively, and accordingly were associated with mass production and batch production systems. in the late 1960s, several ideas began to surface to improve manufacturing. the marketplace was also undergoing changes both in terms of increasing demand of variety products from the customers and increasing competition in the following decades. for manufacturing this meant a shift from mass production system to batch and job shop systems. by then, the advantages and economics of designing and planning for mass production system vis-a-vis batch or job-shop production system had become quite apparent. automation took the front seat with the objective of increasing efficiency by cutting down the throughput time. application of group technology (gt) for manufacturing systems with a variety of products having similar parts helped to exploit the benefits of mass production in the context of batch production systems. single minute exchange of dies (smed) made jit system a reality by cutting down the set-up time drastically (shingo, 1989). in the early 1970s, flexible automation (flexible manufacturing systems and robotics) emerged to offer solutions for job shop. flexible automation helped in producing variety products in small batches with virtually no time lost for changeovers (groover, 1989; asfahl, 1985). computer integrated manufacturing (cim) advocated systems oriented approach where effort was directed towards achieving integration of the different activities in an organisation through a common data base and flexible technology such as cnc (computer numerical control) machines, automated material handling systems (rehg, 1994). the first writing in manufacturing strategy also came up during this time (skinner, 1969). other writings (hayes et al, 1978; wheelwright, 1978) followed and the need to spell out clearly the manufacturing action plans, and integrating them with the overall business strategy to gain competitive advantage were soon established. thus a shift from the piecemeal approach to the holistic approach towards manufacturing problem solving was seen during this time. this and the following decades were marked by the evolution of the new approaches. material requirement planning (mrp) followed by manufacturing resource planning (mrp ii) were developed during this period and offered as solutions to the complex problem of production, material and resource planning. if or among the earlier approaches was integrative, these were supposed to be more so. these approaches were 3

different from or in that no explicit mathematical programming was used. rather, an integrated computerised framework or platform was provided which implicitly tied up the different activities involved in planning through perhaps the simplest of equations depicting the relationship among the variables of interest. application of or in specific modules within the overall framework was not ruled out. augmentation of mrp ii functionality incorporating all other functions such as finance, marketing, personnel etc. led to the development of erp. the other major approaches that came up during these decades were jit and opt. these approaches had the same basic objective, that of achieving efficiency in manufacturing by integrating the same with the other activities of the organisation. the means adopted and the focus however were different. jit as a philosophy focussed on striving for excellence in all departments of manufacturing. excellence was spelt out in terms of zero defects, zero inventory, zero setup time, zero downtime etc. computerised system was not a necessity, rather the solutions or operationalising the philosophy were rooted to a mix of technological development and behavioural science principles. opt on the other hand was similar to mrp in that it was also a computerised approach. focussing on bottleneck concept, advocating balancing of flow rather than capacity, and having its root in a strong scheduling package, opt removed some of the drawbacks in mrp and gained some of the advantages of jit. the above presents an overview of the history of development in manufacturing during this century. it should be clear from the above that the predominant approach for manufacturing problem solving before 1960s has been marked by viewing the whole system as a collection of several smaller subsystems. the implicit assumption was that the problems of the entire system could be taken care of by solving the problems of the different subsystems in isolation. the approaches evolving after 1960, on the other hand, consisted in offering solutions through an integrated approach by looking at the entire system rather than by looking at the parts. for a better understanding of this shift from the “reductionist” to the “holistic” approaches, it is helpful to have a closer look at the new approaches. these are presented in the following section under the three broad heads of erp, jit and opt. 3. the new approaches erp mrp originated in the early 1960s in the us as a computerized approach for the planning of materials acquisition and production scheduling. starting with the requirement of different finished products over the planning horizon, the approach involved determining the quantities of different components or raw materials to be manufactured or purchased in different periods. the approach was essentially top down exhibiting a centralised, hierarchical control structure (browne et al, 1989). mrp system was extended to mrp ii to include rough cut capacity planning (rccp) and capacity requirement planning (crp) to check for feasibility and to update the master production schedule (mps) until a workable schedule is developed. erp originated from mrp ii. however, unlike the latter that attempted to fit in the system to the existing structure of the enterprise, erp generally advocated a re-look into the company’s business processes, and could be applied both to manufacturing and service

4

organisations. erp was also meant to be more all-encompassing, in that besides taking into account the company’s internal issues, it also considered factors external to the company such as competition, time-to-market and so on. erp embodied the planning process for efficient deployment of an enterprise’s resources – man, machines, materials and money. with focus on customers, implying better quality products and faster delivery, erp helped in real time control of operations with a greater degree of accuracy. the erp software incorporated the best business practices in industry, and tried to link the different elements of an enterprise so as to allow the organisation to cope with the changing environment. today, many organisations are undergoing transformation from function-oriented businesses to process-driven entities. erp enables this both at the information systems level and at the applications level. erp has a few problems. first of all, it is capital intensive. considerable investments for implementation and training of erp software are required. secondly, the implementation process is time-consuming. the gestation period varies from one year to one and half year, and still at the end of the road it may not function properly. it has been seen that one-third of the total erp installations worldwide fail to deliver because of poor planning, lack of implementation details, lack of top management support and a host of other factors. since it involves reengineering and transparency, top management support is required at every step. also, dedication and commitment from all users at all levels in an organisation are required for setting up and smooth operation of an erp system (kohli, 1998). jit jit system started in the early 1960s in the toyota automotive plants in japan. it attracted the attention of the world in the early 1980s after the success of the japanese firms. currently it is being used in a variety of industries like automotive, aerospace, machine tools, computer and electronics. jit, like the other new approaches, attempts an integrated solution to manufacturing problem solving (korgaonkar, 1992). however, unlike mrp, it is a pull control system with the flow of materials synchronized to the end item usage rate. advocating production based on the requirement, and striving for zero inventory, zero defect, zero set-up time, etc. in the same breath, jit essentially implied advocating “excellence” in all manufacturing activities. it was imperative that the philosophical preaching be followed up to operationalise the objectives. methods (smed) to drastically reduce set-up time helped in reducing batch size and hence inventory. installing limit switches at relevant places on the machines so as to prevent defects rather than detect defects helped in moving towards zero defects. since jit originated in the automotive industries, one tends to associate its application with mass production and repetitive manufacturing systems. however, today the jit philosophy and jit manufacturing techniques, marked by integrating supplier into the system, modular product design and cellular layout in the shop floor etc., are applicable to all types of discrete parts manufacturing. jit, too, has its failing. when it becomes difficult to accurately predict the demand and restrict the product variety, implementation of jit solutions is almost impossible.

5

opt opt, developed in the early 1980’s by goldratt and cox, took care of a number of drawbacks in the mrp paradigm. as a philosophy, it focussed on the bottleneck management, pointing out the importance of time saved in bottleneck. as a logical extension, it advocated balancing flow and not capacity thereby gaining advantage similar to jit. it may be noted while the fact that bottleneck governs the output rate was known right from ford’s days, the traditional focus was on designing shop floor to balance capacity and not flow. opt also posed a challenge to the traditional accounting focus on efficiency and utilization measures, calling for not only a revision to accounting practices, but also for a rethinking of the assumptions on which they were based. by highlighting the difference between process and transfer batch, opt showed the way to reduce the throughput time. lead time for production in opt was thus not taken a given constant. instead, as it would normally be in practice, lead time was taken as dependent on the schedule. this implied that opt provided an in-built scheduling module for the purpose of implementation of the same. in fact, apart from being looked at as a philosophy, opt can be taken as a tool equipped with a good scheduling package for solving manufacturing planning problems. opt has a few problems. it claims to generate an optimum schedule through its proprietary algorithm. hence, there is no participating or learning approach; also it is very expensive. 4. future scenario in manufacturing from the above discussion, it is apparent that manufacturing problem solving has travelled a long way from the beginning of this century till date. during the course of this journey, variety of tools and techniques have been offered as solutions. the major difference between the techniques of yesterdays and the new solutions offered through erp and the like today, lies in the effort of the latter for striving for an integrated business solution rather than solving a manufacturing problem per se. in this section, the genesis of the new approaches has been further examined to obtain an insight into the future of manufacturing problem solving. as noted earlier, techniques used in the early days of manufacturing problem solving were typically based on industrial engineering, mathematics and probability theory. operations research was developed along the same lines, in that it had its foundation in mathematics. however, unlike the other approaches, integration was embedded in the modelling methodology of or. as manufacturing became more and more complex, or modelling of manufacturing problems turned out to be more and more difficult. multiple products manufactured through multiple stages in production processes requiring different kinds of resources with different types of interactions and sub-systems’ objectives led to large monolithic mathematical programming models difficult to solve. simplified models, eyeopener at times, were often of limited practical value. simulation as a part of or methodology offered a trade-off in providing a “good” solution with more of real-life incorporated, in terms of non-linearity and uncertainty. this perhaps called for a rethinking, resulting in the development of holistic approaches, where the hopes were pinned on the fact that in the face of growing complexity in manufacturing, these

6

approaches will, at the least, help preserve the overall perspective. complexity in manufacturing mainly arose from the changing nature of demand from the marketplace. demand for variety products, changing with time and often also marked by uncertainty, posed a challenge altogether different from the earlier period, when much lower variety was required. increased competition also implied that unlike taylor’s time, mere cost minimisation objective is not enough. strategic priorities were redefined to include not only cost but also quality, flexibility and delivery. gaining competitive advantage implied taking competitor into account in the modelling. pursuing optimal solution in a hierarchical fashion or piece-meal optimisation of sub-systems soon became less important. instead, what was called for was technological breakthrough in all facets of manufacturing, ranging from machines that will enable quick production of variety products, ensuring quality, to automatic material handling, storage and retrieval systems that will reduce the overall cycle time. on the software side, it was imperative to operate with a single data base with all the sub-systems linked up through a computer. holistic approaches such as erp, jit and opt thus followed as a natural sequel. while a number of studies are reported on mrp and mrp ii, formal studies on erp are limited, the same being the most recent development. jit, on the other hand, has been under scrutiny as a part of the overall japanese manufacturing technique (schonberger, 1982) and has also been examined in various ways relative to mrp and mrp ii. opt in this context, remains proprietary in nature with applications reported from uk and usa (koziol, 1988). though to start with, mrp and jit were typically meant for mass production system and opt for job-shop system, today such barriers have given way, and the effort has been to make erp, jit and opt universal. as has already been noted, each of these approaches has advantages and disadvantages. on the positive side, by and large all these approaches do successfully provide integrated solution. erp, as it stands today, perhaps provides the most integrated approach. however, it is confined to more of transaction reporting with most of the benefits accruing from its solution being in terms of reduction in lead time by adoption of best practices. the process of tapping the other major source of benefit, by applying optimisation for planning and scheduling is yet to be reported on a large scale. the same refrain is found in the context of mrp (baker, 1993) when it is pointed out that mrp is more of a management information system rather than a decision-making system. opt supposedly scores better in this respect by having the scheduling package at its heart. studies have reported the similarity between the scheduling logic of opt in its early stage and jit system (goldratt, 1988). the potential drawbacks of jit are perhaps embedded in the same “simplified lessons” in jit (schonberger, 1982) for which the approach has been found to be most beneficial. future manufacturing in this context would perhaps be a hybrid system with these approaches complementing each other, to take advantage of strengths from each of them (browne et al, 1989; giauque and sawaya, 1992). recently, money resource planning or mrp iii, has been coined as an information integrating tool concerning both money and material flows with focus on financial aspects rather than the traditional production logistics aspects (gautam, 1996). buzacott (1995) while sharing his views on future 7

manufacturing felt that manufacturing systems would be either integrated or cooperative in nature. for make-to-order or make-to-stock situations with known customer demand, integrated manufacturing, based on computer and information technology, was felt to be most appropriate. cooperative system, based on human interaction for problem solving, appeared to be the solution for situations that are difficult and expensive to anticipate, as in the case of innovations and new products introduction. based on the above discussion and historical development traced earlier, it seems that besides a hybrid or an integrated/cooperative system, the other distinct possibility of future manufacturing could be an approach resulting from the combination of the reductionist approaches of the past and the more recent holistic approaches. this would perhaps be manifested in terms of a computer-integrated platform with erp or some hybrid as the major driver linking different sub-systems with user interface and built-in optimisation capabilities. 5. conclusion in this paper, the gradual evolution of different production management paradigms was discussed, and the main aspects of their differences were highlighted. it is observed that the evolution took place to adapt to the changing environment. an attempt has been made to outline the future of the manufacturing systems based on the changing customer requirements. nowadays, manufacturing is faced with great challenges, particularly in terms of variety of products, shorter product life cycles and faster delivery to the market. in this situation, probably no one technique can exist in isolation. in fact, mrp/mrp ii, erp, jit and opt are, perhaps, not competing with each other. taken together, they will help understand the complexity of modern manufacturing systems, and form the base of future manufacturing planning and control.

8

references 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

asfahl, c. r, robots and manufacturing automation, john wiley and sons, 1985 baker, k.r., “requirements planning”, handbooks in or & ms, vol. 4, elsevier science publishers b.v., 1993, pp. 571-627 bowonder, b, “competitive and technology management strategy: a case study of telco”, int. j. of technology management, vol. 15, nos. 6/7, 1998, pp. 646-680 browne, j; j. harhen and j. shivnan, production management systems – a cim perspective, addison-wesley publishing company, 1989 buzacott, j.a., “a perspective on new paradigms in manufacturing”, journal of manufacturing systems, vol. 14, no. 2, 1995, pp. 118-125 chhachhi, v, “rebuilding mahindra and mahindra”, business today, oct-nov, 1995, pp. 72-81 darlington, j and c. moar, “mrp rest in peace…”, management accounting-london, vol. 74, iss. 9, oct., 1996, pp. 14-20 dugdale, d and c. jones, “the theory of constraints”, accountancy, vol. 116, iss. 1225, sep., 1995, pp. 134 garvin, d. a, “manufacturing strategic planning”, california management review, vol. 35, iss. 4, summer, 1993, pp. 85-106 gautam, v, “erp: the new mantra for competitive edge”, dataquest, april 15, 1996, pp. 60-68 giauque, w.c and w.j. sawaya, “strategies for production control”, production and inventory management journal, vol. 33, iss. 3, 1992, pp. 36-39 goldratt, e. m, “computerized shop floor scheduling”, international journal of production research, vol. 26, no. 3, (1988), pp. 443-455 goldratt, e. m, the haystack syndrome, north river press, inc., new york, 1990 groover, m. p., automation, production systems, and computer integrated manufacturing, prentice hall of india, new delhi, 1989 gumaer, r, “beyond erp and mrp ii”, iie solutions, vol. 28, iss. 9, sep., 1996, pp. 32-35 gupta, i and p. datta, “the nuts and bolts of strategy”, the strategist, business standard, 21 march, 1995 hayes, r. h and r. w. schmenner, “how should you organize manufacturing?”, harvard business review, january/february, 1978, pp. 107-108 kohli, v, “corporate viagra”, businessworld, 7-21 august, 1998, pp. 18-26 korgaonker, m.g., just in time manufacturing, macmillan india limited, madras, 1992 koziol, d. s, “how the constraint theory improved a job-shop operation”, management accounting, vol. 69, no. 11, may, 1988, pp. 44-49 maccarthy, b. l. and j. liu, “a new classification scheme for flexible manufacturing systems”, int. j. of production research, vol. 31, no. 2, 1993, pp. 299-309 miller, e. j and a. k. rice, systems of organization, tavistock publications, 1967 narasimhan, s.l.; d.w. mcleavey and p.j. billington, production planning and inventory control, 2nd ed., prentice hall of india, new delhi, 1997 newman, w and v. sridharan, “manufacturing planning and control: is there one definitive answer ?”, production and inventory management journal, vol. 33, iss,

9

25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.

1, 1992, pp. 50-54 padukone, h and s. subba rao, “global status of jit – implications for developing countries”, productivity, vol. 34, no. 3, oct-dec, 1993, pp. 418-429 ptak, c. a., “mrp, mrp ii, opt, jit and cim – succession, evolution, or necessary combination”, production and inventory management journal, vol. 32, no. 2, 2nd qtr, 1991, pp. 7-11 ramchandran, j, “an architecture for a process control costing system” in measures for manufacturing excellence edited by robert s. kaplan, harvard business school press, 1990, pp. 193-222 rehg, j. a., computer-integrated manufacturing, prentice hall, inc., 1994 schonberger, r. j., japanese manufacturing techniques, the free press, 1982 shingo, s, a study of the toyota production system from an industrial engineering viewpoint, productivity press, 1989 shivnan, j; j. richard and j browne, “production and inventory management techniques – a systems perspective”, modern production management systems, elsevier science publishers b.v., 1987, pp. 347-362 skinner, w, “manufacturing – missing link in corporate strategy”, harvard business review, may/june 1969, pp. 136-145 sridharan, r, “turning lean and mean”, business today, 7 oct., 1998, pp. 101-105 suresh, k, “manufacturing excellence in tvs: an indian example” in strategic manufacturing for competitive advantage edited by a. lahiri and v. r. dhanuka, oxford and ibh publishing, 1998, pp. 87-110 thorne, k and m. smith, “synchronous manufacturing: back to basics”, management accounting – london, vol. 75, iss. 11, dec., 1997, pp. 58-60 weston, r, “erp users find competitive advantages”, computerworld, vol. 32, iss. 3, jan., 1998, pp. 9 wheelwright, s. c., “reflecting corporate strategy in manufacturing decisions”, business horizons, february, 1978, pp. 61

problem for discussion : production planning (source : applied mathematical programming by stephen p. bradley, arnoldo c. hax, thomas l.magnanti, addison-wesley publishing company, 1977, pp 229-230.)

10

a tire manufacturing company has agreed to deliver the following during the next 3 months: month nylon fiberglass june july august total

4000 8000 3000

1000 5000 5000

15000

11000

the company has two presses, a wheeling machine and a regal machine, and appropriate molds that can be used to produce these tires, with the following production hours available in the upcoming months: machine june july august

wheeling machine

700

regal 1500

300 1000

400 300

the production rates for each m/c & tire combination, in terms of hours per tire, are as follows: wheeling machine nylon fiberglass

0.15 0.12

regal machine 0.16 0.14

the variable costs of producing tires are rs.5.00/ operating hour, regardless of which machine is being used or which tire is being produced. there is also an inventory-carrying charge of rs.0.10 per tire per month. material costs for the nylon and fiberglass tires are rs.3.10 and rs. 3.90 per tire, respectively. finishing, packaging and shipping costs are rs.0.23 per tire. prices have been set at rs.7.00 per nylon tire and rs.9.00 per fiberglass tire. the manager is interested in determining the production plan that will meet the delivery requirements and maximize the profit.

11