MANUFACTURING MANAGEMENT SYSTEM (MMS) Investment in Advanced Manufacturing Management Systems has a strategic impact that can affect the long term competitiveness of enterprises improving the ability of firms to create new markets, introduce new products, and to react quickly and effectively to competitors. Since the available methodologies to support strategic decisions are not easily applied, firms are in the unpleasant position of evaluating strategic decisions without any practical tool that is able to estimate the value of each specific action and its consequences at the strategic level. The aim of this application software is to provide a programmable framework for the selection and configuration of Advanced Manufacturing Management Systems. In particular, a framework is proposed in which the different aspects and evaluations that are involved in long term capacity planning are properly organized. 1. Introduction In recent years a relevant change has developed in manufacturing systems technology. Such change has been mainly due to the introduction into the market of new equipment able to combine microelectronic and programmable devices within mechanical machines. The so-called Advanced Manufacturing Management Systems (AMSs) are a result of such a revolution. At the moment different architectures of Advanced Manufacturing Management Systems are available: some of these architectures are well known and tested like flexible transfer lines, flexible manufacturing cells, flexible manufacturing systems, while others are rather new and they are being studied within national and international research projects or directly proposed by machine tool builders. The problem of capacity acquisition when AMSs are considered is particularly complex for several reasons, First of all, the high investment involved makes companies very sensitive to the risk factor thus precluding the adoption of AMSs. The consequent sensitivity of the management staff could therefore cause the improper evaluation of benefits of these systems, such as scalability and flexibility. Another reason is that flexible capacity enlarges the spectrum of possible future scenarios because many alternative strategies are viable, thus making the risk evaluation more difficult. Furthermore, many advantages of these manufacturing systems are not easy to quantify and therefore they are seldom evaluated properly. The strong interaction among the components of AMSs makes it necessary to carry out evaluations considering the system as a whole. Therefore, simple rules of thumb are normally quite misleading, and appropriate and sometimes rather sophisticated evaluation methods are required. The problem is further complicated by the fact that choices must also be evaluated in both strategic and economical terms. Indeed manufacturing systems can be a good competitive weapon for the strategy of the firm if the capacity choice is coherent with the overall strategy of the firm. To be profitable in the long term, a production system must be both efficient and aligned with the company strategy. In practice even a good manufacturing plant can have problems if its production system does not conform to the company strategy. The selection of capacity is becoming more and more relevant for manufacturing companies because a good or bad decision can deeply affect the profitability of the company that invests in new capacity. In practice, since capacity has a cost, it is not possible to solve the problem simply by acquiring extra capacity to face all possible future requirements, but it is necessary somehow to weight the advantages of having enough capacity to front future needs with the cost of maintaining unused capacity. All these aspects tend to hinder the exploitation of the opportunities offered by AMSs. This is particularly true in SMEs (Small and Medium Enterprises) where structured approaches to the solution of capacity acquisition problem are not applied. The problem of capacity planning in Advanced Manufacturing Systems has been deeply investigated in the last 20 years. However, the proposed methodologies did not reach the main goal. Most of them do not propose a solution to the whole problem, but limit their focus on some well-defined sub-problems. This approach has led to solutions that are not practical to apply to real problems since they treat only a portion of the problem. A software tool like a Decision Support System (DSS) for Advanced Manufacturing Management System that is able to treat such a complex multidimensional problem is needed in order to support people involved in long term capacity decision planning. To develop this Decision Support System it is necessary first to define the whole decisional process in all its steps and details so that different sub-problems can be identified and then solved, in an integrated way, by means of specific tools. The aim of this AMMS is to provide a framework and specific tools for the selection and configuration of Advanced Manufacturing Systems in the long term capacity planning problem. This application function is organized as follows. It defines the concept of capacity in manufacturing while the basis of manufacturing strategy theory are summarily explained also. In the AMSs investigated in application are described and structures the strategic problem by means of the language modelling. 2. Manufacturing capacity Since manufacturing has become an industrial phenomenon, the problem of adequately sizing plants has always been discussed, an important work of organizations includes taking stock of the resources at one’s command and planning the fullest use of them all. Capacity in general can be defined as the set of any kind of resources that can be used to create value for the customer and, in general, the cost of capacity is lower than the value the customer pays to acquire the product or the service provided. Without capacity it is not possible to create value because at least a minimum amount of resources is necessary. Furthermore, manufacturing capacity is defined as the set of human resources and equipment that the company can use to produce goods or services to sell in the market. The dimensions of manufacturing capacity are: Type. There are in practice many manufacturing systems that differ in terms of their characteristics and several keys of classification can be used, some of them are: standard or advanced, rigid or flexible, capital intensive or not, automated or manned, etc. In practice, the characteristics of the system to stress depend on the type of analysis we want to carry out. Amount. The quantity of capacity acquired to create value to customers. Since capacity cannot be fully exploited, literature generally uses the terms theoretical or nominal amount of capacity to refer to the purchased capacity. The amount of capacity can be expressed in machine time available in a period (e.g. hours per day, hours per week, etc.) or in number of pieces per period. Knowing the production rate of products on that system it is possible to move from time to part units. The portion of capacity that is used to manufacture products is known as utilized capacity. Utilized capacity can vary day by day for several reasons, thus average and standard deviation are used to represent utilized capacity in a defined time period. Cost. The total economic value that is necessary to spend for acquiring, running, maintaining and dismissing a manufacturing system. The above characteristics synthesize the main strategic issues of manufacturing systems. The available capacity is the amount of production time the firm can effectively use to satisfy the market demand that is the amount of theoretical capacity taken from the unused portion due to any reason except lack of demand. In practice, available capacity is normally compared with what customers demand. If the available capacity is greater than the capacity used to satisfy the customer demand there is a waste quantified by the difference between available capacity and utilized capacity; this waste is also known as excess capacity. However, if the available capacity is lower than the capacity that would be necessary to fully satisfy the customer demand, there is a lack of capacity quantified by the difference between requested capacity and available capacity; this lack is also known as demand surplus. The ratio between utilized capacity and theoretical capacity is the utilization level of the plant. Among the several causes of the difference between utilized and theoretical capacity, the most frequent are: personnel scheduling, setups, maintenance and lack of demand. A quantitative analysis on the reasons for unused capacity should always be done before deciding to acquire new capacity. Manufacturing capacity is characterized by the following issues: Capacity cannot be stored. If a manufacturing system is not used in a period because of lack of demand, the related portion of capacity is wasted and cannot be utilized in the future. An alternative is to produce even if there is no demand with the purpose of storing finished goods; in this case higher inventory costs are incurred. Capacity can be changed only in discrete steps. In practice, an increase or decrease of capacity corresponds to an acquisition or a dis mission of a finite amount of resources (e.g. a machine or a human operator). Capacity can be changed with considerable lead times. In practice, ordering a new machine or a new production system can take several months. Generally capacity cannot be reduced but only expanded because it could be difficult in practice to sell used mechanical devices. Manufacturing systems have a long life cycle (usually from 5 to 20 years). Manufacturing systems have a ramp-up period in which the production level is lower than the theoretical one and all the efforts are devoted to reach the target value as soon as possible. The ramp-up period can be very critical because it can take several months, or years or in some cases the system never reaches the target production value. 3. Manufacturing strategy The capacity problem is a decision related to the overall strategy defined by the company. Strategy is a term used in business planning that refers to the overall scheme of managing and governing the future course or direction of the company. Strategy implies careful selection and application of resources for the most advantageous position, in anticipation of future events. A company strategy is a set of plans and policies by which a company tries to gain advantages over its competitors. A company strategy is defined at corporate level and must consider several issues such as research and development (R&D), sales, marketing, finance and manufacturing. From the company strategy all the function strategies are then derived and, among these, we are interested in manufacturing strategy. In particular, manufacturing strategy deals with the decisions concerning the specific role of manufacturing in order to achieve competitive advantage in the market. The role of the manufacturing in the whole company strategy pointing out the need of coherence between company strategies and implemented manufacturing tasks. Manufacturing strategy can contribute to firms’ success by supporting the implementation of the competitive strategy defined by the corporate. A company’s competitive strategy at a given time should place particular demands on its manufacturing function, and, conversely, the company’s manufacturing policy and operations should be specifically designed to fulfil the tasks demanded by strategic plans. A mismatch between company strategy and manufacturing strategy can be source of lack of competitiveness. The conceptual strategy model generally recognized In the model, mainly derived from the competitive environment influences both the company and manufacturing strategies that empirically demonstrates there is a relationship between competitive environment and competitive strategy defined at corporate level. The link between competitive strategy and manufacturing strategy has a relationship between competitive strategy and productive competence with business performance while this link is valid only for high business performers. However the relation between competitive strategy and performance is not supported by empirical evidence, the reason is that manufacturing strategy mediates between them. In conclusion, it appears that the conceptual model is valid for high performers and the link between competitive and manufacturing strategy is highly relevant, that is a competitive strategy works well when supported by coherent manufacturing tasks Let us enter into more detail on what is a manufacturing strategy. The competitive priorities are a consistent set of goals for manufacturing: Cost: production and distribution of the product at low cost. The lower the cost is the higher the profit or the possibility to operate an aggressive strategy of price competition in the market is. Delivery: reliability and speed of delivery. It is generally recognized in literature how important the level of deliveries is on the customer perception. Quality: manufacture of products with high quality and performance standards and that quality, in all its multidimensional aspects, can be used to gain competitive advantage. Flexibility: product mix and volume. The ability to change the priorities of jobs, or the machine assignments of jobs in the shop floor, or the production volume can allow the firm a competitive advantage. Innovation: capability to introduce new products or product variations effectively. The presence of innovation in the list of manufacturing competitive priorities is not generally recognized. However, innovation can be an important weapon in the market competition. After specifying the competitive priorities coherently with the company’s strategy, the manufacturing actions potentially adoptable to pursue the stated goals are classified into two categories: structural and infrastructural decision areas. The structural decision areas have generally a long term impact, are difficult to reverse and they require substantial capital investment. A brief comment for all decision areas is now reported and the reader is referred to for more details. Facilities: the company should decide on the location, the size and the focus of facilities. Process technologies: the company should decide which process technologies to adopt to manufacture products. In addition the company has to choose between acquiring or developing the chosen technology, and other strategic issues such as the degree of automation, the layout, the scalability and flexibility of the process. Capacity: the company should decide on the type of capacity to use in manufacturing, the amount and timing, that is when to acquire and how much. Vertical integration: the company should decide on the relationships with its providers and customers The infra-structural decision areas affect the people and the systems that do manufacturing work. The infra-structural decision areas are generally more tactical, linked with specific operating aspects and do not require substantial capital investment: Vendors: the company should decide on the structure and size of the network of vendors and also the relationships with them. Human resources: the company should decide how human resources shall be selected, trained and payed. Also, the company should design the job and the skill levels. System practices: the company should decide the practices to be adopted for production and material planning, management of manufacturing systems, quality, standards, etc. Organization and management: the company should decide the nature of management. For instance, employees in manufacturing can be organized by product, function, or geographical areas. It is very important that all decisions made in different areas are coherent and together contribute to reaching the defined competitive priorities. Indeed, the success of a company depends on the coherence of its strategy with the competitive environment and the level of integration of its strategies and decisions. The firms that do not maintain consistency between the pursued competitive priority and the manufacturing decisions they implement do not achieve superior business performance. The selection of capacity is one of the strategic decisions of a firm’s manufacturing strategy that has direct consequences on all the competitive priorities defined in the manufacturing strategy. First of all, the capacity choice deeply affects production costs. Indeed, different manufacturing systems have different costs because they may differ in the personnel involved, cost of devices, consumption of power and tools, reliability of equipment, etc. It is also important to take into account the timing of the investment. If a 100% increase of the customer demand is forecasted in the immediate future it is necessary to have the additional capacity necessary as soon as possible to front the market expansion. If the increase of demand is forecasted to occur 5 years in the future, it is hopeful that the firm waits to expand its capacity unless large wastes are incurred. Therefore, a firm can fundamentally adopt two different policies: lead or follow the customer demand. If there is a capacity demand surplus, i.e. the firm follows the market demand, the capacity utilization will be high but there is also a risk to lose customers due to long delivery lead times. If there is an excess capacity, the firm anticipates the market demand, the system utilization will be low but it will easier to maintain high delivery reliability and flexibility. However, market demand is uncertain and it may occur that the capacity expansion of the firm is not followed by the increase of market demand thus causing capacity wastes. In other words there is a trade-off between utilization and delivery reliability. Also flexibility is affected by the capacity choice because an excess of capacity allows the firm more flexibility to react to changes in market demand. Depending on the type of equipment selected it will change the ability of the firm to modify the production mix; for instance a rigid machine forces the firm to run large lot sizes in order to avoid expensive set-up times. Furthermore, the type of capacity can influence the quality of products; for instance different machine tools reach different precisions and therefore products with different quality levels. Also innovation can be improved by properly selecting the type of capacity. If a firm has only dedicated systems in the shop floor, the frequency of the introduction of new products will probably be small because the launch of a new product involves the re-configuration, often very expensive, of the whole system. 4. Advanced Manufacturing Systems Advanced Manufacturing Technology (AMT) has been subject of investigation since late the ’70s when computer numerically controlled (CNC) machine tools started to be widely adopted on shop floors. AMT covers a large area of nontraditional technologies that firms can use to maintain or improve their competitiveness. In practice production systems such as CNC machine tools, automated flow lines, cellular manufacturing systems, flexible manufacturing systems, or design tools such as CAM (Computer Aided Machining), CAPP (Computer Aided Process Planning), but also management tools such as MRP (Material Resource Planning) and ERP (Enterprise Resource Planning) are considered advanced manufacturing technology. In this book, for simplicity of exposition, the authors deal with only a portion of the large set of production systems: chip removal manufacturing systems that is systems having turning, or milling, or drilling, or grinding, or all those processes that obtain the finished part by cutting material (therefore deforming, casting and assembling are not considered). Furthermore, in this book the attention is restricted to those systems with high degree of automation and large amount of capital involved. The reason is that completely automated systems are considered complex and it is difficult in practice to evaluate their performance, to manage and design them. Therefore the need of having adequate decision models for this class of systems seems to be evident. In addition, this necessity grows if these systems require large amounts of capital because a wrong choice could compromise the profitability of the investment and, in some cases, the survival of the companies in the market. In particular, systems like standalone machine tools are not considered, even if they are expensive (e.g. machines are CNC type), because they are easy to evaluate in terms of production rate and utilization and adequate models are already available. Two classes of systems are taken into consideration throughout this book: Dedicated Manufacturing Systems (DMS) and Flexible Manufacturing Systems (FMS). These classes of systems are described in the following sub-sections. 4.1 Dedicated Manufacturing Systems Dedicated Manufacturing Systems are those systems that are conceived, designed and managed appositely on the needs of a product or a very restricted family of products. The main characteristics of these systems are: Rigid equipment. The equipment is designed to satisfy the needs of the product, or the restricted family of products, to which the whole system is dedicated. Therefore the machines and devices such as transporters, grippers, etc. are designed to accomplish a very limited set of operations that cannot be normally expanded unless large costs are incurred. Stations in transfer lines are a typical example of equipment rigidity. Normally machine movements are not numerically controlled by a computer but mechanically by means of cams or other mechanical devices. High production rates. The equipment is dedicated and normally designed to minimize processing times. In order to cut processing times, one or more operations can be performed in parallel. As a consequence machines are generally fast allowing the system to reach higher production rates compared with other ones (e.g. Flexible Manufacturing Systems). Low skills. The skills needed to run the system are normally low since human jobs are reduced to loading and unloading parts and maintenance. Easy management. Given the limited number of products a DMS processes, and the simplicity of flows in the system, the scheduling of resources is quite easy. Low investment. The equipment is rigid and everything is designed to accomplish only the operations that are necessary to manufacture the products to which the DMS is dedicated. Therefore the investment cost of the system is not large if compared with that of more flexible systems with CNC machines. Excess capacity. The amount of capacity unused because of lack of demand cannot be used to manufacture different products. For this reason the residual value of the investment is very small. There are fundamentally two categories of DMS in practice: dedicated machines and dedicated flow lines. Dedicated machines are those machines appositely designed to perform efficiently the product process cycle; these machines are generally conceived and developed in the firm because a high knowledge and experience is necessary on the process. Dedicated machines work in stand-alone mode and in general are completely automated except for the loading and unloading of parts; therefore they are also simple to manage not requiring any sophisticated tool and for this reason they are not considered in this book. We deal with Dedicated manufacturing flow lines that are an important and wide spread type of DMS. This type of manufacturing systems is described in detail in Chapter 4. 4.2 Flexible Manufacturing Systems CECIMO (Commit Europeenne de Cooperation des Industries de la Machine Outil) defines an FMS as an automated manufacturing system capable, with a minimal human action, of producing any part type belonging to a pre-defined family; these systems are generally adopted for the production at small or medium volumes, in variable lot sizes that differ also in their composition. The system flexibility is generally limited to the family of part types on which the system is conceived. The FMS has devices for planning the manufacturing, scheduling the resources and saving the production data. As the above definition points out, the main characteristics of FMSs are: Flexible equipment. The equipment is flexible enough to satisfy the needs of all the products belonging to the family. Indeed all the machines are CNC type and can be programmed to perform a large number of operations. In practice, it is only needed the ability to write a simple computer program to code the process cycle of a product into instructions that the numerical control of the machine can read, understand and operate to execute them. Low production rates. Machines have generally a spindle for executing operations in a sequential way. As a consequence machines are generally slow in comparison with machines of DMS. Recent innovations such as high spindle machines and linear motors are rapidly spreading in FMSs thus reducing processing times and inactive rapid movements respectively. Medium/High skills. The skills needed to run the system requires a minimum knowledge in programming and managing CNC machines. Complex management. The management of FMSs is complicated by the large number of products. Indeed, for each product it is necessary to schedule properly machines, fixtures and tools. Large investment. Machines are flexible and require large investments. Therefore, the investment cost of the system is very large if compared with that of dedicated systems. Excess capacity. The amount of capacity unused because of lack of demand can be used to manufacture different products. The residual value of the investment takes into account this issue. 5. A framework for capacity problems Many factors have to be taken into account when a capacity investment decision is analyzed: firm’s strategy, uncertainty of markets, competitors’ strategy, available system architectures, types of technologies, etc. Investment in AMSs is like an umbrella covering different sub problems that have to be analyzed and solved before making the final decision which system to buy”. These sub-problems are not independent since they are related one another and their relationships are not simple to formalize and to quantify. Some of the different sub-problems the firm has to solve when an investment in AMSs is analyzed are in the following. Market. The firm has to decide where to concentrate its efforts: niche or broad market. This decision is taken at corporate level and is generally already available when the capacity problem is analyzed. This subject is out of the scope of this book and it is not taken into consideration. Products. The firm has to decide which products will sell in the selected market. This decision is made at the corporate level which defines the market segment and the macro-characteristics of the products with which the firm wants to compete. Therefore the new potential products to launch in the market are known. At the manufacturing strategy level of detail the final choice will deal with the selection of product codes to launch in the market. The first decisional level of product selection is out of the scope of this book and it is not taken into consideration. Service level. The firm has to establish the level of service provided in the market. A high level of service typically involves large efforts, however a low level of service may be the cause of a loss of customers that, unsatisfied, change their supplier. This decision is part of the manufacturing strategy and it will be investigated in Chapter 2. Technology. The firm has to decide which technology is most appropriate for manufacturing the products to market in the future. The choice of technology can be fundamental in the market strategic position of the firm. An innovative process technology developed internally to the firm can put the firm in a leadership position. On the contrary, a standard technology process can be acquired by any competitor and cannot be a competitive weapon in the market. The firm has to establish if and how much production capacity can be acquired from subcontractors. The decision has to consider all the possible future consequences that can derive from this choice. Indeed, the outsourcing of a product can be loss of knowledge and skills and can decrease the innovation level of the firm in the long term. Also the contractual power with outsourcers is critical because it dynamically changes depending on the particular relationships that are defined and modified between seller and buyer. Guidelines of the outsourcing strategy are generally decided at the corporate level while details on outsourcers and the quantitative levels of externalization are part of the manufacturing strategy and will be faced in the company. Flexibility. The firm has to decide the levels of flexibility the manufacturing capacity should have. The more flexible the acquired capacity is the faster and cheaper the firm’s reaction to any changes in the market is. This decision is part of the manufacturing strategy and will be faced. System architecture. The firm has to decide the type of production systems. Indeed, given a type of technology selected at higher level in the decisional process, the firm has to choose among several potentially adoptable alternatives to manufacture products. This decision is part of the manufacturing strategy Resources. The firm has to decide on the specific type and number of machines, carriers, fixtures, tools, etc. to use in the new system. In other words the firm has to decide on the detailed configuration of the manufacturing system, eventually supported by the builder of the production system. This decision is part of the manufacturing strategy and will be faced. A correct evaluation of the investment in AMSs should consider all the factors in an integrated and global risk-approach that analyzes the investment from different points of view. Frequently it occurs that an action to improve a specific key-factor of the firm can have a negative impact on other key-factors; for instance, an increase of flexibility often causes an increase of costs incurred by the firm. Therefore, it is necessary to quantify the impact that each single decision has on the whole problem in order to solve the numerous trade-offs that normally characterize strategic problems. Taking as a reference the manufacturing strategy model described above, the firm has to evaluate the impact that each alternative AMS has on the competitive priorities the firm defines at manufacturing strategy level. Also dependencies with the other decisional areas are very important because an incoherence between the value of each capacity choice depends also on the type of selected technology, facility position, current knowledge, etc. The strategic problem of planning the manufacturing capacity in Advanced Manufacturing Systems is described by means of the IDEF formalism where inputs, outputs, controls and mechanisms are encoded using the ICOM approach. 5.1 Planning Production The purpose of activity in Planning production capacity in Advanced Manufacturing Systems is to define the detailed configuration of Advanced Manufacturing Systems in the planning horizon. The viewpoint adopted in the diagram is that of decision-makers. Decision-makers are the managers that solve the capacity problem. Starting from input information regarding system architectures (i.e. type of production systems that are currently used and potential ones that could be acquired as additional capacity resources) and products (i.e. technical and economic data related to those products that are currently manufactured and potential ones that could be manufactured by the firm in the future), the outputs of the activity Production Planning are the definition of the detailed configuration of AMSs to be adopted in the different periods of future planned horizon and the selection of the product codes that will be produced by the firm per period. In particular, the defined plan is a timing of the estimated capacity in AMSs that will be required in the future planning horizon. The planning horizon is broken down into periods of three or six months depending on the level of detail of the analysis. An example of the main output of the activity is the minimum and maximum values of needed internal and external capacity are tabled for each product, on each manufacturing system, for every time period in the planning horizon. More precisely, capacity planning requires a large set of information as explained in the following. The necessary input information for the proposed capacity planning models is: • Current manufacturing structure: information describing the whole set of hardware and management resources currently used to manufacture the products that are sold in the market. In particular with hardware we refer to production systems and equipment, while for management we refer to general practices that are necessary for the production such as production planning, quality control, etc. • Current products: information on products that are currently produced and marketed by the company. This information contains both technological (i.e. technical drawing, process cycle,) and management information (i.e. forecasted market demand, production cost, price,) • Potential product families: information on the new products potentially marketable in the future. This set of products is decided higher up at the corporate level, however the final choice can be made only after a complete and detailed product profitability analysis is carried out, that is after the manufacturing system to use for these products has been selected and consequently also estimates on future production costs become more reliable. Product input contains both technological and management information: features, process cycle, forecasted market demand, etc. • Current position: definition of the actual firm’s market position, if there is any (e.g. the company could also be a new comer in the market). Capacity planning is controlled by higher decisions made at the corporate level or structural characteristics of the market in which the firm operates in: o Competitive strategy: the whole strategy pursued by the company at corporate level. All competitive strategies are variants of generic strategies characterized by a choice between differentiation and delivered cost, i.e. the product price. This choice should be completed with the information of the market focus: niche or broad market? This information is necessary in order to plan capital investments coherently with the corporate policies. For instance, the aggressive policy of increasing market share deeply affects the capacity decision. o Competitive environment: this constraint follows the description of the market in which the firm will operate in the future. A model of competitors, customers, type of market, etc., represents the reference environment to be considered in decision making. • Budget: the profile of the budget available for investments in AMS in the long term. Also this constraint is decided at the corporate level because it involves the analysis of the firm’s financial position. • Outsourcing strategy: it is the decision, made at the corporate level, dealing with which products should be outsourced and fixing a qualitative level of externalization. The main outputs of a capacity planning problem are: • Capacity plan: the decisional process leads to a plan of all the internal and external capacity that is necessary to have in the planned horizon. • Products to market: the decisional process leads to the final selection of product codes the firm will produce and market in the planned horizon. In order to obtain the above outputs it is necessary to use the following mechanisms: • Decision models: the decision process is supported by formalized models and tools that aid decision-makers in structuring the problem and in quantitatively evaluating, in terms of benefits and cost, the value of each alternative solution. • Process & System data base: decision-makers normally use technological information on product process cycles and production systems potentially adoptable in shop floor. We assume that this information is already available in a database or it is provided by technicians. According to the description of the capacity problem, the decision is hierarchical and thus it is necessary to make first some important strategic decisions such as the quantitative level of provided service, the flexibility needs that future capacity should have, etc. After the main strategic variables have been fixed, it is possible to evaluate the production capacity that is necessary to have in order to reach the defined strategic objectives. At this step, a more detailed investigation about the required production capacity is needed: this means to evaluate the make or buy sub-problem in order to define a rough internal” production capacity level per period. Starting from this information, alternative system configurations can be proposed and a performance evaluation of each one (in economic and productive terms) is required to select the best ones for each time period of the planning horizon. This hierarchical decision process is described by means of the IDEF0 modelling language in the following subsections. 5.2 Planning Production Level diagram The goal of a Planning Production Level diagram is to have a more detailed definition of the overall architecture for the decision of capacity planning in Advanced Manufacturing Systems. In a Planning Production level diagram the different decisional steps of the whole problem are shown pointing out their interactions in terms of information. Activity Planning Production has been hierarchically decomposed into 4 sub-activities that are now described in detail. For each sub-activity of a Planning Production level diagram a decision model that supports decision-makers in the long term capacity planning problem is proposed. Interactions with decision-makers are also specified in the comments of the diagram. The first two activities deal more specifically with the strategic aspects involved in the investment in AMS, i.e. selection of investment amount, accepted risk level, production mix and system type, while the last two activities face the problem of the detailed AMS configuration, that means the generation of alternatives of production systems, the evaluation of their performance and finally the choice of the best ones to use in the final capacity plan. All these activities are strictly related by information and decision flows. The Planning Production level diagram is shown and its functions are explained below. • Planning at strategic level. The purpose of this activity is to design strategic variables involved in the capacity acquisition problem when the company strategy, the competitive scenario and the competitive position are properly defined by decision-makers. Indeed, when firm plans the production capacity in AMSs in the next planning horizon, it becomes necessary to revise its manufacturing strategy on the basis of the estimated investment cost of new production systems to be acquired. This involves, among the others things, to select products that could be manufactured in the future. However, at strategic level other decisions are considered. Furthermore, at strategic level the type of production system has to be preliminarily defined since it can have long-term impacts; in fact, the introduction of a new technology can deeply affect a firm in the change management phase or in searching for new people with different skills. Therefore, the system architectures that seem to be the most promising at strategic level are indicated to the downstream decision-makers in the capacity planning problem. More in details, input for activity A1 consists of the following set of information: o Current manufacturing structure: production systems owned by firms and described in terms of type, production rate, cost and availability. Details on system practices are not necessary at this level of the problem. This information arrives from the manufacturing and accounting areas. o Current products: process cycle of products currently manufactured by the firm, historical production volumes, historical demand, forecasting on the average demand value in the long term, internal and eventually external production costs. This information arrives from manufacturing and accounting areas. o Potential product families: rough process cycle of potential products the firm may manufacture in the long term, forecasting on the average demand value in the long term, estimates on internal, and eventually external, production costs. The indication of the products derives from corporate decisions while the detailed information arrives from R&D, marketing and manufacturing areas. o Current position: definition of the actual firm’s market position. Current position is characterized by the market share, or the growth rate for each actual product, etc. This information comes from the marketing area. The constraints imposed by the company strategy and the global context are described in more details: • Competitive strategy: the market strategy the company selects to pursue such as cost leadership, differentiation but also risk attitude of management, financial strategy of the company, etc. This information arrives from the corporate level and is qualitative. • Competitive environment: information regarding the market in which the company intends to compete, that is the market uncertainty level, the market competition rate, the market innovation rate, the market concentration rate and so forth. This information derives from the marketing area. • Outsourcing strategy: information about the policy decided at the corporate level and main supplier characteristics such as location, reliability, outsourcing prices and so on. In a few words the outsourcing conditions describe the market supplier network where the company usually does business. This is very important for defining capacity acquisition strategies because market strategy. This information arrives from the corporate level and the manufacturing area. • Budget: negative cash flows available in the planned horizon for the investment in additional manufacturing capacity. A reasonable assumption is that portions of budget that are not invested in a time period can be used in the following ones. This information arrives from the corporate level. The output will consist of the following indications: • Aggregate long term capacity: amount of production capacity required to produce the potential production mix at the established service level. This information can be useful to decision-makers that evaluate the output of activity A1 and can decide to introduce some changes in the problem definition. • Service level: definition of the minimum level of satisfaction of the market demand that is acceptable to achieve the strategic goals. This information is necessary to define an optimal capacity planning in the long term and represents a constraint for activities A2 and A4. • Outsourcing level: detailed indications about make or buy” strategies of the firm. In particular for each product a range of admissible levels of externalization is defined. This information is necessary to limit the outsourcing coherently with the company’s strategic decisions already taken at the corporate level and represents a constraint for activities A2 and A4. • Potential production mix: preliminary selection on the types of products the firm could manufacture in the planning horizon specifying long term volumes for each product. This information is an input for all downstream activities. • Rough investment estimates: preliminary estimate on the investment cost that is necessary to acquire additional capacity. This information can be useful to decision-makers who evaluate the output of activity A1 and can decide to introduce changes in the problem definition. • Types and amount of AMS: indications on the type of manufacturing system architectures potentially profitable to work the production mix. As already written in the previous sections, the models in this book will deal with Dedicated Manufacturing Systems and Flexible Manufacturing Systems. It is possible that the same product can be manufactured profitably on both types of AMS, in this case the final choice will be made downstream this activity after more refined analysis, or the firm decides to adopt, if possible, both systems to get more flexibility. This information is an input for activities A2 and A3. To produce its outputs, activity A1 uses: • Mathematical Programming: standard mathematical programming techniques are used to define a first capacity planning that is necessary to have for making the strategic decisions above described. • Expert Systems: an expert system with fuzzy rules is proposed to make decisions at this level. Indeed the fuzzy approach seems proper to model the vagueness characterizing the strategic design input variables. • Process & System data base: at this level aggregate and rough technical data are used to define and evaluate the production rate of the alternative AMS. What is important at this level is the total product processing time at a machine and the aggregate costs of the system such as investment and main variable and fixed costs. Other useful information includes the lead time of the AMS that is the time between the ordering of an AMS and the beginning of its running. • Defining the capacity profile. After the main strategic decisions have been made, the next step is to decide the production capacity that has to be acquired during the whole planning horizon. Indeed, the analysis developed at the previous step can suggest only an aggregate long-term production capacity level without taking into account dynamic features such as the market demand volatility or the possibility of delaying the investment during the planning horizon (Lim and Kim, 1998; Brandimarte and Villa, 1995; Dangl, 1999). The ability of the system to meet the demand can be assessed considering both capacity of owned production systems and corrective actions: for instance, demand peak can be smoothed out by acquiring some extra-capacity from a subcontractor or by stocking in advance some production items. Of course, these actions cannot be always taken, and even if they can, a careful optimization of their use is needed, as they have a cost. However, information about the possibility of using subcontractors (such as costs, reliability and accuracy of each sub-contractor, etc.) and storage (storage capacity, inventory costs, and so on) is needed. The purpose of activity A2 is to define the capacity timing introducing the multi-period and dynamic point of view. Starting from the designed strategic variables, A2 defines how to use the make or buy strategy, indicating how much capacity the firm has to acquire (see Figure 1.7) per period and how to proceed with the externalization, if there is any, suggesting more refined values. From activity A1, activity A2 gets as inputs: • Potential product mix: the technical characteristics of products are necessary to plan capacity in the long term. • Types and amount of AMS: the choice between rigid or flexible capacity affects the definition of the amount of internal capacity, as they have different investment and operating costs. From activity A1, activity A2 receives as constraints: • Outsourcing level: external capacity planning is limited by the decision made in activity A1. In such a way the strategic directions on outsourcing provided by the corporate flow through the decisional process. • Service level: it is necessary, in order to avoid unsatisfactory or trivial solutions, to know the minimum level of demand fulfilment the firm should guarantee as decided in activity A1. • Budget: the budget available limits the set of feasible solutions. • Requirements on capacity variability: decision-makers can introduce more constraints during the planning horizon. For instance, it could be dangerous to double the capacity of the firm because serious operative problems could be incurred. Whatever the reason is, the decision-maker can introduce this type of constraint on the capacity of the firm. Because of the level of detail, this constraint is not present in the context diagram of Figure 1.4. The output of activity A2 consists essentially of: • Risk evaluation: a preliminary evaluation of the risk in terms of variability of cash flows. This information is useful for decision-makers that analyze the output of the activity. • Internal capacity: information on the amount of capacity that is necessary to have in the future. It is represented as a time varying range of required capacity for each type of internal resource (i.e. DMS and FMS). This range of effective production capacity per period is used by the downstream activity A3 to select configurations feasible to the plan. • External capacity: information on the amount of capacity that will be probably externalized in the future. It is rep resented as a time varying range of required capacity that it may be externalized. This range of external production capacity per period is used in activity A4 to select the best capacity plan. The essential mechanism is based on: • Stochastic optimization: optimization methods are used to select the time manufacturing capacity taking into consideration the uncertainty of the market demand and the different decision times in the planning horizon. • Process & System data base: same information as in activity A1. • Identifying the AMSs alternatives. Function A3 is the activity that has the objective of accurately defining the potential configurations of AMSs during the planning horizon specifying all the allowable changes that can be introduced into the system to react to future market evolutions. These identified configurations are only potential because they are a preliminary selection of the production systems to be adopted in the future; the final selection will be done by the downstream module A4. In order to define a preliminary set of detailed configurations, it is necessary to consider the range of capacity established at higher level by activity A2. The example in Figure 1.8 shows the capacity profile the AMS has to respect. In such a way the sets of systems that do not fit with the range of capacity provided by activity A2 are discarded thus decreasing the number of potentially adoptable solutions. The detailed alternative investment plans, which are the outputs of the activity, are modelled as possible paths in a graph in which nodes represent detailed configurations (i.e. type and number of machines, carriers, fixtures, etc.) and arcs represent feasible transitions for moving from a specific configuration to another one (i.e. in the case market demand increases, a production system can be enlarged by adding a new machine). The example in Figure 1.9 shows the identified alternative AMS configurations based on the internal capacity profile of Figure 1.8. The inputs of activity A3 are: • Potential product mix: detailed information on the potential set of products. At this level the analysis is more refined and the information on the single processing operations is necessary in order to correctly estimate machine processing times. This information arrives, enriched with more details, from activity A1. • Types and amount of AMSs: set of manufacturing systems to dimension by allocating their resources: type and number of machines, carriers, tools and buffers. This is the same information that arrives from activity A1 to A2. Activity A3 is controlled by: • Internal capacity: the estimated needed internal capacity range expressed in number of pieces for each product. The capacity range depends on the time period. This information arrives from activity A2. • Budget: the budget constraint inserted from decision makers. The outputs of activity A3 are: • Feasible AMSs: feasible configurations to adopt in the planning horizon. The information on these configurations (i.e. the nodes of the graph) is very detailed because it specifies the type of systems with all their resources such as machine tools, buffers, part carriers, tool carriers, fixtures, load/unload stations. This information is used by activity A4. • Feasible transitions: future allowable changes in configurations (i.e. the arcs of the graph) that can be introduced by the firm in the future. Costs and times to implement transitions on configurations are also provided as outputs of the activity. This information is used by activity A4. Mechanisms used by activity A3 are essentially: • [Performance evaluation tools: analytical methods are used to evaluate the performance of configured manufacturing systems. In particular, simple and static equations model in an approximate way the behavior of AMS in a preliminary analysis, while queuing theory is used to dynamically evaluate the behavior of manufacturing systems. • Configuration rules: set of technological rules that allows the proper selection of system devices coherently with the operations of potential products. • Process & system database: detailed information on system devices: speed of machines, working cube, movement times, etc. Given the type of problem and the long planning horizon, these tools must take into account uncertainty. In particular, within activity A3 two different performance evaluation modules are used, the first one based on approximate analytical techniques for cases where uncertainty can be expressed in stochastic terms (see Chapter 4) and the second one for cases where uncertainty must be evaluated in fuzzy terms (see Chapter6). [A4] Searching for the capacity acquisition plan. The goal of activity A4 is to find out the most profitable capacity plans in the planning horizon on the basis of detailed alternative configurations defined by activity A3. To do this it is necessary to simulate” the described market environment selected at strategic level in order to estimate the value of the different capacity plans. Optimal plans have to consider outsourcing policies defined by activity A2; indeed a simulation of the market will allow calculating profitability of the outsourcing level of a specific product instead of acquiring more internal capacity. Furthermore, optimal plans have to fit with all the constraints established at strategic levels such as the profile of budget, or the maximum acceptable risk. Therefore, final plans of internal and external capacity are provided to decision-makers. Again, decision-makers will have the possibility of interacting with this module; for instance, a request from the user could be the evaluation of a particular plan that has in his mind, or to recalculate optimal plans after the introduction of more strict constraints. The representation by fuzzy set theory may be useful to represent market condition variability and its relative level of uncertainty. This is possible both in a long period view, in which it is very difficult to represent the probabilistic values, and in a short period view, where, for example, it is necessary to define the demand of a new product. In this case, in fact, no historical data are available to infer correctly the probability distributions. Under activity A4 innovative algorithms to solve a Travel Salesmen Problem on fuzzy networks are devised. The model must consider not only the uncertainty in the input parameters provided by the configuration activity, but also all the evolutions that are consequences of the different adopted strategies. The inputs of the module are made up of the following information: [A4]-I1 Feasible AMSs: the nodes of the graph, i.e. feasible configurations per period, their performance and optimal areas of exploitation, from activity A3. [A4]-I2 Feasible transitions: the list of the feasible transactions between the configurations of each successive sub-periods with the related cost and time, from activity A3. [A4]-I3 Potential product mix: the whole set of information about products. This information arrives from activity A1. Constraints to the activity A4 are constituted by: [A4]-C1 Competitive environment: description of the environment in which the firm operates. This information, directly from decision-makers, is used to generate simulations for testing the feasible AMSs. [A4]-C2 External capacity: possible outsourced quantities for each product. In the case in which a system cannot satisfy at the desired service level the market demand the firm can outsource production in the quantities specified by this control. This control arrives from activity A2. [A4]-C3 Service level: the level of market demand satisfaction. If this constraint is not satisfied a plan is considered unfeasible. This control arrives from activity A1. [A4]-O1 Capacity plan: the detailed capacity acquisition plan in the planning horizon. The information concerns the optimal evolution path among the many possible paths on the graph of the alternative configurations, in other words which type and how many resources for each type have to be acquired in every time period and which capacity size will be the recourse to externalization. [A4]-O2 Products to market: the final choice of products, among the potential ones in input that are selected for the production in each time period, i.e. the products that can be profitably marketed. The mechanisms used in activity A4 are: [A4]-M1 Fuzzy-DEVS formalism: given the vagueness of information, some parameters must be defined in fuzzy terms for each elementary period. Therefore, a model to represent production system dynamics under a fuzzy market representation is developed: to achieve this goal, Fuzzy-DEVS formalism [Anglani et al., 2000] is used. [A4]-M2 Graphs optimization: heuristics for finding the optimal path in the graph defined by activity A3. The models described as mechanisms in the proposed framework will be described in the following chapters. These models implemented in software tools constitute SW modules which, integrated in a common software architecture, will constitute the specific packages of a Decision Support System (DSS) to long term capacity planning in AMSs: Strategy planner Module to aid decision-makers in taking strategic decisions such as market segment, market share, growth rate, products to market, etc. This module will be based on an expert system incorporating top management rules. Vagueness of market will be modelled by means of fuzzy set theory. An aggregate capacity planning is also modelled. For details see models described in Chapter 2. Risk planner Module to aid decision-maker in defining the profile of the estimated capacity that is necessary to pursue strategic goals. This module will be based on stochastic and dynamic optimization to find the required time capacity in the planning horizon. Once required capacity has been defined, another module will be able to aid decision-makers in selecting between internal (e.g. firm’s shop floor) and external (e.g. outsourcing) resources. For details see the models described in Chapter 3. Configurator Module to aid decision-makers in defining the detailed design of production system. This module will use a performance evaluator tool to find good solutions. For details see models described in Chapter 4 and Chapter 6. Capacity selector Module to aid decision-makers in selecting which” AMSs have to be acquired and when” they will be acquired. This module will select also the type of capacity (i.e. internal or external) and will be based on optimal path’s search in graphs where both arcs and nodes are weighted by costs. For details see models described in Chapter 5. Decision models have been tested on a real case in the automotive metal-component sector. In this market the fierce competition leads firms to increase the flexibility of their facilities in order to react to the frequent market changes (Koren et al., 1997,Matta et al., 2000,Matta et al., 2001). The reason for this change in turn is motivated by the fact that automotive suppliers tend to increase the range of products to attract the consumer, launch new models of car and decrease the time to market; in other words more attractive products in shorter intervals they propose in the market and more competitive they are. Each component manufacturer tends to produce well defined types of products (e.g. outlet manifolds) that are supplied to different car manufacturers, therefore the market is composed by fewer and fewer focused suppliers. The whole market of final goods is subject to uncertainty: each single final product can be a success or a failure and the same is for components of which is made. Given the fact that stock reduction and just in time policies are normally adopted, the producer of components must follow, even in the short term, the fluctuation in the demand. Also the weak contractual power of producers of components reduces the profit per part. Car component suppliers suffer this trend. They have to face frequent changes in product demand, changes in mix, modifications on existing products and introduction of new products selecting the best production system in terms of profitableness. Existing production systems do not match with the above market trends. Traditionally DMS have been adopted for the production of a small family of part types (one or few part types) requested by the market in high volume (Matta et al., 2001). Since DMS scalability is low they are normally sized to reach from the beginning the maximum market demand the firm forecasts to satisfy in the future. But in many situations DMS do not operate at full capacity due to the lack of demand. Analyzed transfer lines operating in the sector of automotive components were saturated 53 % on average (Matta et al., 2000). In this case DMS profitableness is very low because the potential capacity of the system is not exploited. On the contrary FMS have been adopted for the production of a large part mix in small quantities. FMS are conceived to react to all the possible changes of the market, therefore their flexibility is too large and expensive for the needs of the firms (Perrone and Diega, 1999). In many cases car component suppliers partially exploit the flexibility offered by these systems, given the fact that it is rare that their part mix changes completely. Investment to acquire FMS is very high and it considerably affects the cost per part unit produced. Flexibility, customized to the potential future changes the products may undergo, would be fully exploited by the car component suppliers that would not buy unneeded flexibility. The production system would be designed with the desired level of flexibility so that it can face efficiently the future changes of the part families during their life cycle (Matta et al., 2000,Perrone and Diega, 1999). The solution is to have rapid adaptive machines to industrialize new parts in short times, to react to limited changes in demand and part features, and finally to produce with low cost per part (Perrone and Diega, 1999). All the proposed models have been validated on the data collected from an enterprise competing in the component automotive sector.