Design of Flexible Production Systems

In the last decade, the production of mechanical components to be assembled in final products produced in high volumes (e.g. cars, mopeds, industrial vehicles, etc.) has undergone deep changes due to the overall modifications in the way companies compete. Companies must consider competitive factors such as short lead times, tight product tolerances, frequent market changes and cost reduction. Anyway, companies often have to define production objectives as trade-offs among these critical factors since it can be difficult to improve all of them.

Even if system flexibility is often considered a fundamental requirement for firms, it is not always a desirable characteristic of a system because it requires relevant investment cost which can jeopardize the profitability of the firm. Dedicated systems are not able to adapt to changes of the product characteristics while flexible systems offer more flexibility than what is needed, thus increasing investment and operative costs. Production contexts characterized by mid to high demand volume of well identified families of products in continuous evolution do not require the highest level of flexibility; therefore, manufacturing system flexibility must be rationalized and it is necessary to find out the best trade-off between productivity and flexibility by designing manufacturing systems endowed with the right level of flexibility required by the production problem. This new class of production systems can be named Focused Flexibility Manufacturing Systems-FFMSs.

The flexibility degree in FFMSs is related to their ability to cope with volume, mix and technological changes, and it must take into account both present and future changes. The required level of system flexibility impacts on the architecture of the system and the explicit design of flexibility often leads to hybrid systems, i.e. automated integrated systems in which parts can be processed by both general purpose and dedicated machines. This is a key issue of FFMSs and results from the matching of flexibility and productivity that respectively characterize FMSs and Dedicated Manufacturing Systems (DMSs).

The market share of the EU in the machine tool sector is 44%; the introduction of focused flexibility would be particularly important for machine tool builders whose competitive advantage is based on the ability of customizing their systems on the basis of needs of their customers. In fact, even if current production contexts frequently present situations which would fit well with the FFMS approach, tradition and know-how of machine tool builders play a crucial role. Firms often agree with the focused flexibility vision, nevertheless they decide not to pay the risk and efforts related to the design of this new system architecture. This is due also to the lack of well-structured design approaches which can help machine tool builders to configure innovative systems. Therefore, the FFMS topic is studied through the book chapters following a shared mission: "To define methodologies and tools to design production systems with a minimum level of flexibility needed to face, during their lifecycle, the product and process evolution both in the technological and demand aspects. The goal is to find out the optimal trade-off between flexibility and productivity".

The book framework follows the architecture which has been developed to address the FFMS Design problem. This architecture is both broad and detailed, since it pays attention to all the relevant levels in a firm hierarchy which are involved in the system design. Moreover, the architecture is innovative because it models both the point of view of the machine tool builder and the point of view of the system user.

The architecture starts analyzing Manufacturing Strategy issues and generating the possible demand scenario to be faced. Technological aspects play a key role while solving process plan problems for the products in the part family. Strategic and technological data becomes input when a machine tool builder performs system configuration. The resulting system configurations are possible solutions that a system user considers when planning its system capacity.

All the steps of the architecture are deeply studied, developing methods and tools to address each subproblem. Particular attention is paid to the methodologies adopted to face the different subproblems: mathematical programming, stochastic programming, simulation techniques and inverse kinematics have been used.

The whole architecture provides a general approach to implement the right degree of flexibility and it allows to study how different aspects and decisions taken in a firm impact on each other. The work presented in the book is innovative because it gives links among different research fields, such as Manufacturing Strategy, Process Plan, System Design, Capacity Planning and Performance Evaluation; moreover, it helps to formalize and rationalize a critical area such as manufacturing system flexibility.

The addressed problem is relevant at an academic level but, also, at an industrial level. A great deal of industrial sectors need to address the problem of designing systems with the right degree of flexibility; for instance, automotive, white goods, electrical and electronic goods industries, etc.

Attention to industrial issues is confirmed by empirical studies and real case analyses which are presented within the book chapters.

Tullio Tolio is Full Professor of "Manufacturing and Production Systems". He teaches "Manufacturing", "Integrated Production Systems" and "Reconfigurable Manufacturing Systems" at Politecnico di Milano, 2nd Faculty of Engineering (Faculty of System Engineering). He is the head of the course on "Management of Research" offered to all Ph.D. students of the Politecnico di Milano.

He has carried out research activities at the Laboratory for Manufacturing and Productivity (LMP) of the Massachusetts Institute of Technology (MIT). He has been scientific responsible for the Politecnico di Milano of the Brite Euram project MOD-FLEX-PROD (EU - 4th framework Programme of the EU)and of the IST project DRIVE (EU - 5th framework Programme) funded by the European Community, and of the project SPI7 funded by MURST (former Italian Ministry of University and Research). He is national coordinator of the project "Methodologies and tools to design production systems with focused flexibility" funded by MIUR (Italian Ministry of University and Research), scientific responsible for the Politecnico di Milano of the project FIRB01 "Software frameworks and technologies for the development and maintenance of open-source distributed simulation code, oriented to the manufacturing field" and of the Network of Excellence VRLKCiP "Virtual Research Lab for a Knowledge Community in Production" (EU - 6th framework Programme). He has published more than 100 papers in international journals and international conferences. He is Fellow of the CIRP (the International Academy for Production Engineering) and member of AITEM (Associazione Italiana di Tecnologia Meccanica -Italian association for manufacturing. Regarding management activities, he is the Delegate of the Rector of Politecnico di Milano on "Quality assurance in education" and Head of the Division "Tecnologie Meccaniche e Produzione (Manufacturing)" of the Department of Mechanical Engineering of Politecnico di Milano; in the years 2000-2003 he was Head of the Ph.D. program in "Manufacturing and Production Systems" of Politecnico di Milano.

His research topics cover the areas of design and management of integrated production systems. The research activities encompass innovative and traditional system architectures in different sectors (machine tools production, production of mechanical components, services). The main research activities are summarized in the following points.

Configuration of Manufacturing systems

The design of manufacturing systems is a strategical activity for manufacturing firms due to the high investments involved in system acquisition and to the impact that the design decisions have on the future performance of the firm. In particular in this research area research has been carried out on the problem of selecting the proper systems architecture (transfer lines, flexible manufacturing systems, manufacturing cells, modular manufacturing systems) and machine architecture, on the problem of defining the right timing for production capacity acquisition, on the problem of system dimensioning and quality control system design. Research efforts have been also devoted to the problem of automatic generation of simulation models to be adopted within the configuration modules.

Short term management of Manufacturing systems

To fully exploit the system capacity and capability it is necessary to adopt system management tools. In this area studies have been carried out on the loading problem in Flexible Manufacturing Systems (FMSs) and on the scheduling of manufacturing systems.

Performance evaluation of Manufacturing systems

Both to design and to manage a system it is necessary to asses its performance. In this area, research activities have been concentrated on the development of new methodologies and tools to evaluate system performance with the aim on one hand of reducing the burden of modelling and running experiments typical of simulation techniques and on the other hand of eliminating the restrictive assumptions typical of analytical tools. In this area a new decomposition technique has been developed which can model systems composed of machines subject to different types of failures. Moreover new solutions have been proposed to model systems with loop configurations and flexible lines producing different products.

Process Planning

Part programs required to machine mechanical components on numerically controlled CNC machines are normally defined as a fixed sequence of operations and are generated without considering the actual load on the machines. This approach results in a reduced exploitation of system capacity. An alternative way of tackling the process planning problem is to generate non-linear process plans which include alternatives in the operation execution order and/or in the types of operations required. In this area research activities have been devoted to the automatic generation of non-linear process plans and on the use on non-linear process plans both in the short term production planning and in the selection of the resources during the system configuration phase.