Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Manufacturing: Definition and Brief History

Manufacturing is the process of converting raw materials or purchased components into physical products. It generally refers to all transformations that take place in a factory or plant and includes both machining and assembly operations, as well as internal transport. In almost all cases, the resulting products serve to fulfill a desired functionality that in turn determines how they are used. Products may be sold to consumers but may also serve as input for subsequent manufacturing or assembly processes; in many cases, complex final products are the result of a large sequence of manufacturing processes, performed in various stages and by different organizations in a supply chain or network. As an illustration, consider raw materials (e.g. iron ore) obtained through mining, which are converted into useable basic components (e.g. steel plates or castings) from which components are made by means of machining operations (engines and body parts for a car), which are finally used in a subsequent assembly process of a final product (a car). In this chapter, we restrict ourselves to manufacturing systems that take place at a single geographical location. Aspects of the broader concept of supply chains are discussed throughout this book, to start with the SCOR reference model presented in Chap.3.

Manufacturing as a profession is almost as old as civilization. The production of pottery in ancient times, often decorated with paintings, can be seen as a manufacturing process, carried out mostly by skilled artisans. The Phoenicians were able to build seaworthy ships more than 2500 years ago, as did the Romans and the Vikings later. The cutting of stones and the building of the pyramids indicate high levels of craftsmanship and, not to be forgotten, work organization. The production of weapons, clothes and agricultural products again shows a long history of progress and refinement. Essentially however, all energy needed for the conversion processes was delivered by human and animal forces, without much mechanical aid (although the wheel and the lever can be seen as early tools to amplify human power). Indeed, the word “manufacture” stems from the Latin “manu factus” = “made by hand”.

The first industrial revolution, lasting from about 1760 until 1820–1840, marked the start of manufacturing at an industrial scale, and hence meant the definitive transition from the classical domestic system and the craft guilds to mass production and mechanization. The invention of the steam engine by Thomas Newcomen, and its improvement by James Watt, but also the efficient application of water power and, based on these power generators, the development of machine tools were driving forces that enabled the birth of the factory system. Mass production requires physical concentration of production factors (capital, human labor, energy, machine tools and transport devices), and so the massive factories were born that colored the industrial landscape in the 19th and large parts of the 20th century. During the second industrial revolution, generally positioned between 1840 and 1870, steam-based transport (railways, inland vessels and ocean ships) and the machine tool industry developed further, enabling production and transport at a scale never witnessed before. The British textile industry was one of the early adopters of the factory system, while almost at the same time agricultural production started to benefit from mechanization. The substitution of coke for charcoal allowed the design of much larger blast furnaces for iron production, which in turn formed the basis of the massive steel industries that arose in the late 19th century. In the first decades of the 20th century, the assembly-line based manufacturing of automobiles in the Ford factories were a new milestone in industrial production. Scale, not scope, was the leading paradigm, as exemplified by the answer of Henry Ford to the question in what color the T-Ford was going to be produced: “Any color, as long as it’s black”.

Mass production and limited product diversity continued to dominate industrial production, also during the first decades after the Second World War, to fill the shortages that resulted from the preceding time period. As prosperity started to grow from 1960 onwards, consumers began to demand larger variety, leading to more complex products. In response, manufacturing industries introduced versatile machines that were able to manufacture products in many variants, albeit often at the cost of large changeover or setup times. Economies of scale and hence large batch production remained the leading philosophy.

The two oil crises of 1973 and 1979 for the first time revealed the weaknesses of the prevalent manufacturing philosophy. Raw material prices and interest rates raised sharply and industrial companies experienced a phenomenon already observed by Harris in 1913: large batch production essentially leads to high inventories, not only of final products but also of intermediate materials and parts. Factories were able to produce efficiently, but could not cope with the flexibility that a changing society and changing markets required. In addition, publications such as “Limits to Growth” by the Club of Rome (Meadows et al. 1972) stressed the depletion of natural resources and the pollution of the natural environment at an exponentially increasing rate. For the first time industry and the public started to realize that current supply chains were to become economically prohibitive, and socially unacceptable.

Another major problem concerned product quality, not so much with products delivered to the consumer but merely in the factories. Mass production without sufficient product and process quality assessment may lead to products that do not meet required standards in the final test, and hence are discarded. The result is a system that may be efficient but wastes a lot of materials. At the same time, Japanese manufacturers showed that efficient production of high quality products was possible without the burden of large stocks and wasted materials. The fact that Japan, more than any other country, lacked sufficient natural resources, may have helped leading production engineers to find unorthodox solutions when being confronted with the effects of the oil crises. But also the eastern way of viewing systems from a more holistic perspective as opposed to the fragmented western production philosophies helps to explain the success of Japanese manufacturing in the seventies and eighties of the 20th century. We return to this topic in Chap. 12.

Gradually, it was recognized that next to efficiency and quality also flexibility, i.e. the ability to manufacture large varieties of high quality products in small batches is an essential performance criterion. Fortunately, new technologies proved to be at least a partial remedy. The introduction of flexible manufacturing systems, often based on computerized (CNC) machining and robotized assembly, helped to balance efficiency and flexibility. In addition, computerized information systems such as MRP and ERP helped to synchronize production across various stages in a way that was simply not possible without the computer. At the same time, attempts were made to simplify and synchronize the processes themselves, by adopting production philosophies such as Just-in-Time, or lean and agile manufacturing that focus on rigidly removing any buffer stocks as these were primarily seen as indications of waste or slack that characterize non-synchronized production. And now, when demand for more sustainable products and processes has entered the scene, we again need to rethink our manufacturing strategies.

2 Manufacturing Systems: Fundamentals (Basic)

The raison d’être of a manufacturing system is to make products that add value or that provide a desired functionality to their buyers. Hence, it is essential to know what potential customers value and how one should structure both the manufacturing processes as well as manufacturing logistics to optimally respond to these perceived market demands. Demand sensing and forecasting but also marketing are disciplines entitled to explore potential markets; aspects of these will be discussed in Chap. 6. Here, we concentrate on a characterization of both product/market combinations and manufacturing systems design that guide strategic choices on how to operate to respond to specific demand.

2.1 Product/Market Typology

Knowledge of a company’s customer base for a particular set of products is essential to decide upon the organization of the necessary production processes. Product positioning strategies have been proposed by many authors, here we follow a characterization suggested by Fogarty et al. (1991), see also Zijm (2000). An important notion is that of the Customer Order Decoupling Point (CODP) which splits a chain of steps needed to manufacture a final product into forecast driven and customer order driven activities, c.f. Fig. 5.1.

Fig. 5.1
figure 1

Four CODP’s representing four product positioning strategies

Full size image

  • Make and assemble to stock (MATS). This is the typical manufacturing philosophy for the majority of consumer products such as electronic equipment, food and drugs. In most cases, there is no direct relation between the manufacturer and the market; typically, the wholesaler and the retail sector are responsible for sales and services. Nevertheless, responsibility for a product’s performance resides with the manufacturer. The fact that legal obligations with respect to product safety have multiplied in the last decades may indeed force a company to initiate sometimes massive product call-back actions to not only prevent accidents but also to save its reputation.

  • Make to stock, assemble to order (MTS/ATO). When a large variety of different products is built up from a limited number of components, it makes sense to produce components to stock but to perform the final assembly based on customer orders (catalogue products). In this way, one avoids high final product inventories, but the price to be paid is of course the fact that customer service is no longer immediate. The manufacturing of cars and trucks is a good example of an MTS/ATO system. A fierce competition may force companies to reduce final assembly and delivery times as much as possible, often enabled by a far-reaching degree of automation in the final assembly process.

  • Make to order (MTO). Companies facing a high diversity of end-items in small quantities (small batch manufacturing) where the diversity originates already at the component level, typically operate in a Make-to-order mode. Most metal working (machine) factories belong to this category; their customers are typically OEM’s (Original Equipment Manufacturers) but also public utility organizations, not so much the end-consumer. Materials to be used are still universal and often procured based on forecasts.

  • Engineer to order (ETO). An engineer-to-order company typically designs and engineers products based upon a functional specification of the customer, and in close co-operation with the latter. Highly specialized equipment is typically produced in an ETO mode. Customers may be exploration industries in e.g. the oil and gas industry, ship owners and high-tech manufacturers. Only when agreement on the design is reached, the company starts to purchase materials, next manufactures parts and components and finally assembles, tests and installs the product.

It should be said that, as a result of the thorough penetration of computer integrated manufacturing, but also through the introduction of product modularity, borders between some or the categories above are blurring. When building a product from standard modules, its final constitution may take place at the retailer’s shop or even at a customer’s home. The Swedish company IKEA is a fine example of the latter, offering an almost unlimited customer choice by designing standard modules for furniture that can be arbitrarily combined. The plug and play philosophy in the computer industry also is a good example of the same product philosophy (interestingly, however, a company like Apple was able to build an impressive market position, while not allowing its products to integrate with those of other computer manufacturers and software suppliers).

A related phenomenon is that of mass-customization, i.e. the possibility to adapt a high volume consumer product to specific customer wishes. Examples can be observed in many sectors, ranging from household appliances to fashion. The basic strategy in these sectors is that of postponement, similar to the make to stock, assemble to order philosophy; producing a product or its constituting parts with all the functionality desired, but leaving it to the customer to select the final combination of parts, or to choose a personalized outlook (e.g. a print on a shirt, to be delivered by that customer).

And finally, the advance of new production technologies may cause products to shift from one category to another one. One of the most prominent examples is additive manufacturing, better known as 3D printing, which is discussed in detail in Chap. 23. An early predecessor of 3D printing is stereo lithography which has been in use for more than 30 years as a technique for rapid prototyping but the current range of additive manufacturing techniques is much wider, and allows for the application of a much broader range of materials. It allows for the manufacturing of rather complex parts and products, but is still relatively expensive, needs much energy and is relatively slow, and therefore primarily suitable for low volume or one-of-a-kind manufacturing.

2.2 Manufacturing Process Typology

A second major manufacturing typology pertains to the processes that are needed to build products. As with products, the borders between various types of manufacturing systems are blurring, primarily as a result of the deep penetration of automation and robotics on the manufacturing floor. Nevertheless, the following typology has proved to be useful when translating a manufacturing strategy into concrete production requirements.

Continuous production. Continuous flow production is generally reserved for manufacturing in the process industries and for bulk materials. Oil refineries, the production of a wide range of chemical products but also food processing lines are examples of continuous (as opposed to discrete) production. Products are often liquids that can be packed in any amount desired; the packaging lines, although clearly ending with discrete products, are still generally considered as part of the continuous production system (e.g. a bottle filling line).

Mixed model flow and assembly lines. This category encompasses many assembly processes as can be found in the automotive industry, in consumer electronics, as well as parts manufacturing systems that are based on a fixed, repetitive, sequence of process steps, basically identical for all products. In the past, often a distinction was made between dedicated and mixed model flow production, where the former refers to a line entirely dedicated to the manufacture of a single product. As a result of far reaching automation and robotics, but certainly also due to the rise of demand variability, dedicated flow lines hardly exist anymore (and are in fact unaffordable). Still, mixed model flow and assembly lines are typically suitable for the manufacture or assembly of products that are sold in sufficiently large volumes, with limited basic product variety.

Job shop manufacturing. Job shops are characterized by a highly functional process structure in which machines are grouped according to specific processes, such as milling, drilling, turning or grinding in a machine shop. Each product or small batch of products may have its own routing through the shop, hence the system can in principle handle a large variety of different products. Job shops are typically suited for small product quantities. Routes and machine processes in a job shop are typically determined in a preceding step, process planning, which also yields detailed machine instructions. In the final quarter of the 20th century, with the introduction of Computer Integrated Manufacturing Systems (CNC workstations, Flexible Manufacturing Cells) the ability to manufacture even larger product varieties increased due to the development of advanced NC-programming tools.

Group-technology based or cellular manufacturing systems. In such systems, products are grouped based on similarity in production characteristics and consecutive processing steps will take place in the same cell (consisting of machines, tools and a small number of workers). Advantages typically stem from reduced materials handling and the fact that a team of workers is entirely responsible for, and very knowledgeable on, a specific set of products. As such, these systems form an interesting compromise between flow lines and job shops.

On site manufacturing. Examples of on-site manufacturing include the realization of complex infrastructural works (bridges, tunnels) or the completion of a major industrial facility, works that are often organized as a separate project. These processes are characterized by the fact that equipment needed to realize the product is transferred to the product’s site, instead of the other way around, although a lot of components needed may be manufactured in a factory. Also here, one may observe a shift from one category to another one. In house building for instance, we observe the application of pre-fab constructions in particular in the lower price segments. An interesting other example concerns ship building where we may see the construction of modular ship segments at remote yards which subsequently are brought together, followed by assembly and further finishing operations on site.

2.3 Manufacturing Performance Measures

The primary objective of a manufacturing organization is to ensure that the output of their processes meets market demand in volume, against agreed quality standards and at competitive prices. It goes without saying that the overall costs and benefits should leave room for a sufficient margin to guarantee continuity and to satisfy internal and external stakeholders (including but not limited to shareholders). To that end, a well-thought strategy and a close collaboration of marketing and sales, quality control, financial planning and control, human resource management and production and materials planning is essential. We will not treat all these functions in detail here but limit ourselves to a sketch of main performance criteria and an overview of organizational functions in the context of manufacturing management. In Chap. 12, we will discuss a number of Manufacturing Planning and Control Systems in more detail, while in Chap. 19 algorithms for decision support of these systems will be presented.

As we already saw in the brief historical sketch at the beginning of this chapter, the objectives of a manufacturing company have shifted over time. Still, for many organizations, cost efficiency is a key performance index.

Efficiency is a qualification on how much output is produced given the availability of certain amounts of resources (equipment, materials and manpower). The higher the output per unit of resource input, the more efficient the system is. Generally, we distinguish between technical efficiency (output quantities in relation to input quantities in a technical production sense) and economic efficiency (which is merely based on input and output prices). In the latter case, the cost to produce a given number of products should be as low as possible in terms of tariffs charged for the use of needed production factors (equipment, utilities, materials and manpower). As we will see later, this one-dimensional view on production costs may lead to significant problems at other parts in the supply chain.

Already in the first decades of the preceding century, management started to realize that a strict focus on efficiency bears the risk of producing large quantities of products of which a significant portion turned out to be useless due to bad quality. That observation led to the inclusion of a second major objective: quality.

Quality defines whether a produced product meets pre-defined technical and functional specifications. Statistical methods, as developed at Bell Telephone Company (Shewhart 1931), helped to systematically check product quality. Even more important was the early shift to methods of statistical process control, to check whether working methods and machine parameters are tuned such that bad quality production is prevented, and to adjust parameters in case of deviations. The focus on quality, advocated by industrial engineers like Edwards Deming (cf. Deming 1986), appeared to be one of the major factors behind the success of Japanese production systems, next to the third major objective: flexibility.

Flexibility concerns the ease with which a production system changes between various product variants, or the ability to produce several variants of a product (almost) simultaneously, as well as its adaptability to (temporarily) changing volumes.

The term flexible manufacturing systems is sometimes used explicitly to denote systems that consist of computer numerically controlled (CNC) machines, equipped with tool magazines and automatically tool-changing devices, together with automated material handling systems (AGV’s, pallet conveyor systems). However, flexibility is a much broader concept, sometimes the word agility is used to denote systems that are able to quickly adapt to changing requirements in terms of product volume and mix.

The fourth and final objective often results once the first three objectives are fulfilled but nevertheless needs special mention: speed.

Speed is a requirement put to many manufacturing systems and denotes how fast a system is able to respond to external demand. Clearly, fast delivery can also be realized by producing items to stock from which demand is satisfied but for unique or highly specific products that is simple not an option.

These four performance measures are all meant to audit the internal processes in a manufacturing company in the first place. Of course, they are implicitly reflecting also market demands (in particular quality and speed) but they are not explicitly dealing with the value that products and services of the company provide to the market. Indeed, for that reason, many authors have criticized the one-sided internal focus of these performance indices and have advocated a more holistic view on customer value management. We will return to the concept in several subsequent chapters.

3 Case Study: El-O-Matic

The firm EL-O-Matic was founded in 1973 in the small town of Borne in the Netherlands and started with the design and sales of pneumatic aluminum actuators. In 1981 the company moved to the city of Hengelo. In 1990, it employed about 100 people in its main production facility in Hengelo, while small production facilities and/or sales offices were located in the UK, Germany, the United States and India. Starting with an annual sales of 11,000 pneumatic actuators in 1981, about 110,000 pneumatic and 10,000 electric actuators were sold in 1990. For 1995, a further increase to 177,000 pneumatic and 30,000 electric actuators was expected (of which 25% in the US and 10% in the Far East). As early as 1988, El-O-Matic was therefore rethinking its manufacturing strategy.

Regarding its products, a product family structure could be discerned. El-O-Matic distinguished 12 basic types (sizes) of actuators. Each basic type can be delivered in a large number of variants. The most important parts of the actuator were the housing, two pistons, two end covers and the drive shaft. The diversity of actuators (the variants) is to a large extent determined by the actuator housings since these parts have to fit on the specific equipment of the customer.

The conventional planning procedure operated as follows. Each month, a sales forecast covering the next 13 weeks (one quarter) was presented at product family level. A product family corresponds to the basic actuator types mentioned, while furthermore a distinction was made between products manufactured for the UK and for the Continent, leading to 24 different product families. The company needed a total lead time of 9 weeks, divided into three weeks for the procurement of raw materials (basic housings, shaft and covers, identical within one product family), two weeks for some basic parts manufacturing operations (variant independent), and four weeks for the variant-specific operations, some finishing operations and the assembly of the actuators. El-O-Matic promised customers a four-week delivery lead time (not including shipment and installation). The relatively low diversity during the first five weeks enabled the company to use forecasts to decide upon initial production quantities; only the last four weeks’ production (consisting of variant-specific operations) was based on customer orders.

Facing the rapid product volume and mix increase in the late 1980s the company realized additional investments were needed in particular in metal-cutting capacity, and after careful consideration decided to install a flexible manufacturing cell to perform a number of variant-specific processing steps on the actuator housings. Also, it experienced a severe pressure on its delivery lead times, and so the firm decided to install a Flexible Manufacturing System (FMS) for performing a number of variant-specific processing steps on the actuator housings. The FMS consisted of three CNC machining centers, linked by a pallet transport vehicle and an integrated pallet buffer system. Each machine was equipped with a tool magazine with an automatic tool changing device, while a tool robot provided the connection between a central tool store and the tool magazines. Housings were loaded and, upon completion of the FMS manufacturing cycle, unloaded again at an I/O station, equipped with special fixtures and clamping devices.

The installation of the FMS enabled El-O-Matic to shorten their overall lead times drastically. The forecast-driven part of the production process still consisted of the procurement of raw materials and only a few pre-processing steps for which three weeks in total was reserved. The latter production phases (all metal-cutting operations, including those performed at the FMS), some finishing operations and final assembly took only three more weeks, yielding now a total lead time of six weeks. Since also the tapping of the screw thread is now performed at the MFS, together with all variant-specific operations, the number of basic housing types and therefore the number of product families reduced to 12. During the final stage, still some 150 variants were manufactured and assembled, based on customer orders.

4 Manufacturing Organization (Advanced)

In Fig. 5.2 an overview is presented of the most important functions and their mutual relations in a manufacturing organization, ranging from long-term (on top) to short-term (below), where we restrict ourselves to those functions that have a direct impact on manufacturing. For that reason, the marketing function is not separately discussed since in particular its branding function goes beyond manufacturing; aspects of marketing that directly influence demand are included in Long Range Sales Planning and Demand Management. The same holds for Human Resource Management, where manpower planning aspects are included in Facilities and Resources Planning. Below, we discuss a number of functions in more detail.

Fig. 5.2
figure 2

Functions and their relations in a manufacturing organization

Full size image

Long Range Forecasting and Sales Planning. Long range forecasting aims at the prediction of a market as a whole and is partly a method of expert judgement. A qualitative method to estimate future market volumes is e.g. the Delphi Method. With respect to quantitative forecasting methods, we distinguish between causal models and time series models. For the prediction of market volumes in the long run, causal models are usually exploited, for instance (multiple) regression models based upon earlier observed relations between various identified causal factors and realized sales volume for similar products (Makridakis et al. 1998). Subsequently, sales planning concerns the decision on a target market share for product ranges on a highly aggregate level, based on the market analysis mentioned above, as well as on an assessment of the power of competitors. Also, price setting constitutes an important instrument in gaining a particular market share; for that a well-thought and quickly manufacturable product design may result in a significant competitive advantages.

Medium term sales planning in Make and Assemble to Stock (MATS) systems usually exploits time-based forecasting methods. This is quite different from the practice in Make to Order (MTO) and Engineer to Order (ETO) companies, where production plans need to be aligned with customer contracts that generally cover longer time periods (cf. Fig. 5.1). in addition, more and more MTO and ETO companies gain significant additional incomes from after-sales service contracts. In the long run, sales in an MTO or ETO environment heavily depends on specific customer relations for which general forecasting methods have less value. Primary contract winners are often those companies that are also frontrunners in technological product and process design.

Product Design and Process Engineering. It is well-known that the far majority of costs made during manufacturing are determined by decisions made during the product and process design phase (Ulrich and Eppinger 1995). A well-known approach is for example Concurrent Engineering in which a new product range and the required processes are designed simultaneously, in order to prevent the design of superior products that eventually turn out to be too costly or, even worse, of which the time to design the corresponding manufacturing system, and hence the time to market, becomes too long because of engineering or quality problems in the start-up phase. A general guideline underlying Concurrent Engineering is also to postpone design decisions as long as possible, in order to increase flexibility during the corresponding technology selection processes. In addition, approaches such as Design for Manufacture and Design for Assembly have proven to be extremely valuable, cf. Boothroyd et al. (1994). Also, modularity of products and a high level of standardization of parts and components may help to reduce inventory investments considerably. Of the more technically oriented methods to quickly evaluate a proposed design, we mention rapid prototyping, see e.g. Kalpakjian (1992); interestingly, the most well-known rapid prototyping technique, stereo lithography, is an early example of 3D printing, which recently has developed in a far more mature technology. Another important development concerns feature-based design; here features represent basic physical elements, i.e. a combination of material, physical shapes and tolerance measures. Various CAD (Computer Aided Design) systems exploit elementary and compound features as their basic building blocks (Rembold et al. 1993).

Facility and Resources Planning. Both technological product/process design and commercial planning serve as input to the planning and possible acquisition of the facilities and resources that are needed. In this phase required resources, including manpower, machines and auxiliary equipment, are specified to enable the planned sales volumes to be realized. With respect to the selection of processes and equipment, decisions are fundamentally related to scale. If a product family is expected to run for a sufficiently long time in large volumes, it makes sense to install dedicated machines or e.g. a specialized assembly line. When volumes are only moderate, the product mix increases or product life cycles are relatively short, it makes sense to invest in more generic equipment. Most ETO and MTO systems exploit universal resources and make increasingly use of CNC workstations which can handle a large variety of (coded) work processes.

Here, we also briefly pay attention to the selection of a process layout. Clearly, for high volumes to be assembled the classical (mixed-model) assembly line is a natural choice. On the other hand, the assembly of high capital goods (ships, aircraft) is usually done at site, with components and parts gradually moved to the spot. For parts manufacturing often a functional departmental structure is selected, in which case a smart layout planning is needed to minimize the costs of materials handling between and within these departments (e.g. Francis et al. 1992; Tompkins et al. 1996). Still, in a rigid functional layout, transportation times and the sizes of batches may be significant, hence leading to high work-in-process inventories and overall long manufacturing lead times. For that reason, many authors are advocating a more product-focused layout, of which a cellular manufacturing system is a profound example (Wemmerlov and Hyer 1989). Burbidge (1990) suggests the application of production flow analysis as a means to arrive at a more group-technological process layout.

Demand Management and Aggregate Capacity Planning. The implementation of this function clearly depends on the logistic product/market structure of the company. In medium to large volume manufacturing and assemble-to-stock products, the most wide-spread method is time-based forecasting of which the various versions of exponential smoothing are well-known and much applied examples. Trends and seasonal fluctuations are easily included in such methods. For a detailed account on time-based forecasting methods the reader is referred to Box and Jenkins (1970), see also Makridakis et al. (1998). Next, these demand forecasts are translated into prospective orders and finally order acceptance. Aggregate capacity planning in MATS systems involves the synchronization of production requirements with available resource capacities. Also the planning of additional shifts during certain periods and the decision to temporarily subcontract the production of certain components, may be a part of aggregate capacity planning, for which Linear Programming Models are an excellent tool, see e.g. Hax and Candea (1984), Winston (1994). Next, orders are accepted on a routine basis.

In an MTO or ETO environment, the order acceptance function is generally more complicated, and includes the specification of functional and technical requirements, quality definitions, a delivery time and a price. Now, the role of aggregate capacity planning as a vital part of order acceptance is in particular to quote realistic customer order due dates, based on estimated internal lead times. In ETO companies, aggregate capacity planning does not only relate to the manufacturing divisions, but also to the design and engineering department. In all cases, a clear insight into the relations between the available resource capacities, a possible workload and the resulting manufacturing lead times is essential in order to determine sound inventory policies and to generate realistic customer order delivery times. We pay more attention to techniques for aggregate capacity planning in Chap. 19.

Financial Management. Financial management in a broad sense regards all other functions but its main purpose is to keep the organization financially healthy (Wouters et al. 2012). Naturally, it assesses whether investments are allowable and determines the payback period (or break-even point) of any investment in equipment and resources. Another essential function is to provide the manufacturing organization with sufficient working capital (either internally or externally funded) to run the business smoothly. The amount of capital tied up in material and parts and products inventories limits the liquidity, i.e. working capital, of the organization, hence requires a close interaction with materials and inventory management. The same holds for purchasing; in particular long term contracts with suppliers should be approved by the finance department. Clearly, financial management is also in the lead when drawing up annual budget plans (look ahead) as well as the financial annual report (look back), and the interaction with external financial stakeholders such as banks, insurance companies and the external controllers.

A topic that has received a lot of attention throughout the years is the determination of manufacturing costs. Obviously, costs of personnel and investments in machines, tools and infrastructure (i.e. both direct and indirect costs) are well-known but the annual costs clearly also depend on the depreciation rates and terms. Classically, the summation of these costs led for instance to a machine hour tariff and hence to the cost price of a product, but quite soon it was recognized that such a calculation does not provide a solid basis for e.g. make or buy decisions. Alternatives such as Activity Based Costing are better suitable to assess what activities contribute most to a product’s cost, or what are the main cost drivers. Based upon that analysis and on the company’s market position, sales prices are determined which form the basis for the company’s revenue as well as taxes to be paid.

Process Planning. The function of Process Planning is to specify all the technical information needed before a production order, a job or part of a job can be executed. Usually we distinguish between macro and micro process planning, where the first concerns all decisions at a shop level, while the latter deals with the detailed machine and tool instructions. The way in which a production order is split into a number of potential production jobs, to be loaded on the various resource groups, is a macro-process planning decision, and the same holds for the specification of resource requirements (machines, operators and auxiliary equipment) and product routings. The determination of cutting patterns, and of tool speed and feed rates at machines are micro process planning decisions, captured in NC programs that subsequently instruct machine tools.

Within an MATS environment, process planning is already performed during the process design phase and hence hardly plays a separate role. The same holds for an MTS/ATO system but for MTO and ETO environments the situation becomes quite different. In order to speed up the process planning activities, a selection is often made of possible processes, machining methods and tool combinations stored in a database, after which a CAPP (Computer Aided Process Planning) system automatically generates the NC programs. These CAPP systems in turn are often based on the use of process planning features (not to be confused with the design features discussed earlier) that specify basic material processing patterns (e.g. bending, material removal, welding patterns). The combination of many process planning features yields a complete machine instruction (Kusiak 1990).

It is important to realize that in principle much freedom exists in the selection of machining methods and hence of routings and process plans. Currently, having more advanced CAPP systems available, process planning in many metal working factories is performed only a couple of days (and sometimes less) before actually processing a job. Consequently, it makes sense to take into account the actual work load on the shop floor when developing process plans for a new order, for instance with the aim to balance the load among various workstations. One way to significantly increase the loading flexibility on the shop floor is by developing several alternative job routings (cf. Zijm 1995).

Master Production Planning. Within the MRP terminology this function is often called Master Production Scheduling but it is important to note that this is only a schedule in time, which critically depends on the availability of all resources needed. A Master Production Plan is defined at end-item level for MATS systems but for MTS/ATO systems it generally operates at the level of parts or components, after which it is followed by a customer-order driven Final Assembly Schedule. To explain why that is needed, the concept of a Bill of Materials (product structure specification) is helpful, see also Orlicky (1975) and Chap. 12 of this volume. Each end-item demand is translated (exploded in the Material Requirements Planning terminology) via the Bill of Materials in dependent demand for lower level items that have to be produced in advance (using so-called off-set lead times), or should already be available in intermediate stocks. The problem however is that within an MTS/ATO system, customer order lead times are generally short, and hence the manufacturing of parts and components should be completed well in advance of knowing their exact demand. In particular when a large variety of end-items can be assembled from a relatively limited number of components, an MPP at end-item level would be unmanageable while an MPP at parts level, based upon aggregate demand, is perfectly useful. We will come back to Material Requirements Planning and alternatives in Chap. 12 in more detail. In Chap. 19, we discuss in detail the relation between resource capacity profiles and off-set lead times.

Inventory Management and Materials Planning. Inventory management plays an essential role at both an aggregate and detailed level (Silver et al. 2017). When smoothing aggregate production plans, inventories naturally arise in case of a temporary foreseen shortage of resource capacity, making it necessary to manufacture some products or parts well in advance of their perceived demand. This is the capacity smoothing function of inventories. A second source of inventories is the production in batches, often due to reasons of economies of scale. As we will see in Chap. 12, the well-known Economic Production Quantity essentially balances the costs of the start of a production run (set-up costs) against inventory costs; if set-up costs are high it makes sense to produce a large batch which may be stored to gradually fulfill external demand while meanwhile the resources are used to manufacture other products. A third reason for holding inventories is to buffer against demand uncertainty or normal demand fluctuations; in order to be able to satisfy external demand, so-called safety stocks may be inevitable. Finally, obsolescence may occur, leading to excess part or product inventories, which may have to be disposed of or at best sold against discount prices.

The Materials Planning function translates demand at MPP level towards lower level item demand, as discussed above, taking into account available inventories of lower level items. Materials Planning and Inventory Management together perform the materials supply function to each department or resource group, and hence represents an essential input to Job Planning and Resource Group Loading, as well as to Purchase Management.

Job Planning and Resource Group Loading. Once customer orders or replenishment orders have been accepted and macro process plans have been determined, jobs can be constructed at the resource group level. Basically, a job can be seen as the restriction of an order to a specific department, work cell or resource group (often called production units, cf. Bertrand et al. 1990). However, several customer order related jobs may be combined into a composite job (batching) or one large job may be split into several smaller jobs, e.g. to balance the load among several work cells or to speed up work (lot splitting). In addition, the availability of parts or components in stock may also alter the lot size of a job (this is called netting in the Materials Requirements Planning terminology, see Chap. 12). It is important to notice that jobs are the operational entities to be controlled at the shop floor, starting with their release and ending with their formal completion. The simultaneous loading of various resource groups aims at matching the required and available capacity within each group, by considering effective resource group capacities as well as routing constraints of jobs between the groups, but without specifying in detail routing and precedence constraints of a job within a group. Planning is based on either customer order delivery dates, or inventory runout-times, and in turn defines internal release and due dates for each separate job.

Purchase Management. This function takes care of the procurement of all components and materials that are purchased from external suppliers. It receives instructions from inventory management while the allocation of production jobs to time windows naturally depends on the availability of these externally procured materials. It is a strategically important function as well, typically the Purchase Department establishes contracts with a large number of external suppliers, specifies service level agreements in which supplier delivery lead times, fill rates and prices are set, and carefully watches delivery reliability (Schnor and Wallace 1986). It closely interacts with materials and inventory management and often has to act within strict budget limits. OEM’s of in particular capital intensive goods often primarily take the role of a system integrator with sometimes more than 60% of their total turnover consisting of purchased parts. An extensive treatment of purchasing and procurement is given in Chap. 4.

Shop Floor Scheduling and Shop Floor Control. This is the level where the detailed scheduling of jobs on all workstations in a resource group takes place. The goal is to meet the internal due dates set at the higher Job Planning and Resource Group Loading level. Hence, at this level we typically deal with the sequencing of job-operations on individual workstations, but generally not with lot sizing aspects (these have been covered at the Job Planning level already). Shop Floor Control deals with the monitoring and diagnostics of all operations, reporting on quality aspects, and signaling major disruptions that may require a rescheduling or replanning phase. We pay more attention to shop floor scheduling in Chap. 19, see also Schutten (1998), or Pinedo and Chao (1999).

Quality Control. Quality Control regards most processes at an operational level. First of all, it has an important role in checking incoming (purchased) goods on their conformance to agreed specifications. Such a check is mostly based on samples but a detection of non-conforming products may eventually lead to a full control. Today, many companies satisfy a range of ISO standards which has among others the advantage that a bad quality material or part is easily traced back to the time period it was produced and the machine that processed it, allowing the supplier to trace other parts produced in the same batch. Naturally, Quality Control is also ultimately responsible for the quality of all products leaving the factory but many manufacturing systems also have in-line quality control checks in order to feed any problem back to the shop floor as early as possible. More general, since long a shift can be observed from product to process control, for instance by statistical process control techniques but today often by continuously monitoring machine and tool conditions (tolerances, stand time of tools), enabled by smart sensors, in order to prevent damage instead of repair afterwards (if possible at all). See Ishikawa and Lu (1985) or Taguchi (1986) for an overview of Quality Control philosophies and techniques. In the same spirit, we observe a shift from corrective via preventive to predictive maintenance, enabled by advanced condition monitoring techniques of machines and tools but also indicated by first signals of quality problems of parts or products delivered.

This concludes the description of a manufacturing organization structure. We have decided to cover also a number of technological functions that are less often found in managerial textbooks. In the forthcoming manufacturing chapters we will primarily focus on the capacity planning and materials management functions at the tactical and operational level, not so much on the general organizational issues. In these chapters, we have deliberately decided not to separate material requirements planning and resource group loading, as is still done in many textbooks. We believe such a separation is not only artificial, but in fact the source of many problems. In Chap. 19, we describe how to integrate capacity and materials planning, based upon a workload control concept which highlights the tight relationship between effective capacity (throughput) and lead time management. First we take a look at current and future developments of manufacturing systems.

5 Future Manufacturing Systems (State-of-the-Art)

After several decades of declining interest in manufacturing in both Europe and the US, its significance as one of the few ways to create wealth (Gershwin 1994) has been rediscovered. Both technological and societal developments are responsible for paradigm shifts in manufacturing system design and engineering. Below, we discuss the most important phenomena that have changed, or are currently changing, the landscape.

New Materials and Manufacturing Technologies

Research into new and lightweight (bio-)materials, polymer technology, bio-engineering and nanotechnology has opened exciting possibilities to design an entire range of new products. These products find their applications in both complex industrial assets (e.g. lithographic systems to be used in new generations of semiconductor manufacturing equipment, membranes for separation at a molecular scale), medical instruments (e.g. lab-on-a-chip devices for quick infection or contagion detection) and personal life (e.g. ICT-inclusive wearables). Technologies like precision machining and additive manufacturing (3D-printing) are further steps towards mass-customization. 3D printing in particular is believed to have a major impact on small batch and one-of-a-kind manufacturing as well as in spare parts supply for maintenance, see Chap. 23 for a detailed discussion on its perceived merits. In all cases, the tuning of products to the specific need of the user is greatly facilitated by the versatility of 3D printing as an innovative manufacturing technology.

Automation and Robotics, Internet of Things, Digital Manufacturing

The application of robotics to replace or assist human labor in manufacturing and assembly has already been visible for a long time e.g. in automotive assembly lines, in precision machining, in Automatic Storage and Retrieval Systems (AS/RS) in warehouses, and more recently in the development of unmanned vehicles for both passenger and freight transport, and drones with so far primarily transport and surveying tasks. Their fast development has been made possible by the design of new generations of (micro-)sensors and actuators that enable high precision positioning of automatic devices. One step further is the Internet of Things in which, based on sensor information, devices automatically signal other devices that action has to be taken, e.g. to provide new materials or parts to the manufacturing floor, or to start a detailed equipment inspection after an indication of malfunctioning performance, based upon automatic condition monitoring. More general, engineers envision manufacturing systems in which both materials and machinery are able to communicate with each other and find solutions based on decentralized and autonomous decision making using state-of-the-art algorithms, often based on Artificial Intelligence. This opens the world of digital or smart manufacturing, sometimes also referred to as Industry 4.0, in which a high level of integration of functions of both resources and assets through automatic communication and actuation is expected. In Chap. 12 we pay more attention to recent developments in digital and cloud manufacturing and the advance of cyber physical systems.

It goes without saying that the rapid advances in digital manufacturing require highly skilled workers trained in a variety of technical disciplines but moreover in interdisciplinary thinking. Already now, we observe a shortage of engineers in both basic and applied disciplines such as nanotechnology, bio-engineering, artificial intelligence to name but a few. To fill this gap is a major challenge for both industry and educational institutes at all levels alike.

Circular Economy and Closed Loop Supply Chains, Sustainability

Already since the first appearance of “Limits to Growth”, published by the Club of Rome (Meadows et al 1972) it is recognized (albeit not by everyone) that exponential growth may lead to the unavoidable depletion of natural resources. In addition, many industrial processes and products used have a profound negative impact on our natural environment, through emission of hazardous materials (CO2, NOx, particulate matter), noise and stench, water pollution and infrastructural problems (e.g. congestion). The latter can be remedied by the design of cleaner engines, e.g. electric, hybrid or LNG-powered vehicles for city distribution and local passenger transport, but also through a better utilization of existing equipment by means of smart scheduling. To stop the depletion of natural resources requires a drastic change from a fossil-fuel based towards a renewable resources based economy, as exemplified by solar and wind energy and in particular the re-use of materials or components from disposed end-of-life products. The emergence of the circular economy philosophy essentially describes a system in which waste is turned into fuel or resources for future production, as discussed e.g. by McDonough and Braungart (2002), and by an already longer existing research stream focusing on Closed Loop Supply Chains. The idea is simply to collect, or to set up a return flow process for, disposed products and assets that are disassembled after which components and/or materials are reused or recycled to basic materials for future use. Chapter 16 of this volume is entirely devoted to the design and analysis of Closed Loop Supply Chains.

Reshoring Manufacturing, Social Responsibility

From the sixth decennium of the preceding century onwards we have witnessed a growing disappearance of manufacturing and assembly to so-called low wages countries, in particular in the Far East (e.g. India, China, Korea) and more recently also to Eastern European countries. Initially, most outsourced production concerned the fabrication of low-value and often low-tech products. Increasingly, and as a result of massive investments in higher education, workers in the Far East have demonstrated to be able to manufacture high-tech products as well. Nevertheless, in the last decade we have witnessed a gradual reshoring of production back to Europe and the US. There are a number of reasons for that. First, wages in particular in Eastern-China have raised significantly compared to only 20 years ago. Second, the amount of manual labor of in particular high-tech products has decreased dramatically, making the low-wage advantage (if still existing) less dominant. Third: the logistics costs of transporting products overseas have increased significantly (partly due to high fuel prices but also due to international regulations regarding safety, security and environmental concern), making again far away production less attractive. And finally, there is a growing consciousness under both the consumer and industrial entrepreneurs of what is called Corporate Social Responsibility (partly as a result of stakeholder involvement), in favor of local-for-local production. Some authors use the word “glocal” as a shortcut for “global when needed, local when possible”. Although future supply chains are expected to remain global to a large extent, the trend to avoid useless or low value transport also from an environmental perspective will continue to help diminishing the ecological footprint of manufacturing and logistics.

6 Further Reading

Some information in Sect. 5.1 is based on the first chapter of Hopp and Spearman (2000) who review the history of manufacturing, while in addition emphasizing the differences between production philosophies in East and West. A far more complete history is presented by Chandler (1977), although from a strong American point of view. Section 5.1 draws heavily on Zijm and Klumpp (2016) in which trends and developments in the broader field of supply chain management are reviewed. The product/market typology framework sketched in Sect. 5.2 follows Fogarty et al. (1991), see also Zijm (2000) and Silver et al. (2017). The case study in Sect. 5.3 stems from Zijm (1996). References for the various functions in a manufacturing organization are given throughout the text in Sect. 5.4 while furthermore Zijm (2000) served as a basis for this section. An excellent introduction to Manufacturing Planning and Control Systems is also provided by Vollmann et al. (1997). The relation between capacity and lead time management is discussed in detail in Hopp and Spearman (2000). Silver et al. (2017) provide an extensive overview of inventory management models and techniques. A must-read on the essence of Quality Control is Deming (1986). Further details on future manufacturing systems can be found in Chap. 12 which focuses on a manufacturing planning and control systems, including Enterprise Resource Planning, Just in Time Manufacturing, workload control and a further discussion on Digital Manufacturing.