F

Fabrication processed in the form of projects and steered by

means of methods of project management
▶ Manufacturing by a team (after VDI 5200).
▶ Production

Theory and Application

Facility Planning History

Facility planning developed step by step during
Hans-Peter Wiendahl1 and Peter Nyhuis2 the course of the industrial development. A rigid
1
Institut f€
ur Fabrikanlagen und Logistik, Leibniz division of work was predominant in the first
Universit€at Hannover, Garbsen, Niedersachsen, mechanical factories around the year 1900. Due
Germany to the central drive systems, using steam engines
2
Institute of Production Systems and Logistics, and transmission belts, the machines had to be
Leibniz University Hannover, Garbsen, placed together very closely. A directed flow of
Niedersachsen, Germany material according to the necessary operations
was not possible. Facility planning concentrated
on the production processes, machines, and their
Synonyms drives as well as the transport of material.
The electric motor engine, available later,
Factory design; Factory planning; Layout planning allowed for the machines to be distributed liber-
ally throughout the factory building. Therefore,
facility planning consequently focused on the
Definition optimization of the material flow and the capacity
utilization of the machines.
Facility planning is the process of planning In the 1970s–1980s, the rising diversity of
a factory from the first idea up to the start of variants initiated the lay out of the meantime
production. The process is systematic and task automated facilities toward the customers, so as
oriented, structured in phases, and carried out to be able to deliver quickly and punctually.
with methods and tools (after VDI 5200). Beside the costs, facility planning now also had
Facility planning can also concern any adjust- to consider stock, cycle times, and good schedule
ment following later in the regular operation. The performance. Since the 1990s, globalization and
process may have different causes and also con- the correlated cross-linked production brought
tain different planning cases. The tasks can be along new challenges, especially for the internal

The International Academy for Production Engineering (ed.),

CIRP Encyclopedia of Production Engineering, DOI 10.1007/978-3-642-20617-7, # CIRP 2014
F 494 Facility Planning

Vision, Mission, Strategy and Selection of Location

Staff and Processes and Material Flow and

Organization Facilities Logistics

Project Management
Employees Operating Equipment Material

Building and Property Planning

Space

Core of Facility Planning

Auxiliary Equipment and Infrastructure

Facility Planning, Fig. 1 Fields of facility planning

and external logistics, together with a decreasing production and the facility location. It is then
vertical integration. In addition, it became neces- the aim of location search to concretize the vari-
sary to adapt rapidly to new products, production ants new building, relocation or merger of several
processes, and lot size alterations, which led to locations, so far that statements about the basic
the segmented, lean, and changeable factory. feasibility, costs, and realization periods are
Nowadays, facility planning is considered an possible.
interdisciplinary task between technologists, The production processes with their main
logisticians, work designers, architects, and con- partial functions work piece production, assembly,
struction outfitters, who cooperate in a project and logistics as well as the existing facility instal-
team aiming for sustainability in economical, lations are the basis for each facility planning.
ecological, and social respect (Wiendahl et al. Operating equipment, such as machine tools,
2009; Schenk et al. 2010). devices, tools, measuring equipment, as well as
storage and transport installations, is necessary to
Theory carry out these processes. Thus, this planning field
Planning Fields primarily deals with technological issues.
Facility planning can be divided into seven The assignment of staff and their organization
planning fields and concentrates on four core is tightly linked with the processes and facility
fields, which are each directed to a characteristic installations. On the one hand, the number and
term, Fig. 1 (Wiendahl 2010). qualification level of the employees have to be
The starting point of each facility planning is determined; on the other hand, the sequence of
a strategy, which determines the long-term pro- the single production steps has to be decided.
duction concept based on a vision and mission as Thus, the staff and organizational planning
part of the corporate planning frame. This often deals with the issues of work economy and
leads to the matter of the extent of in-house organization.
Facility Planning 495 F
A final decision about the production concept extensive project management. Its task is the
cannot be made until the material flow has been structuring of the project into subprojects, the
examined and a logistics concept becomes appar- establishment and supervision of time and cost
ent. The aim is a production with low stock, plans, the draft of the project organization and the
minimum cycle times, and rapid reactions. reporting to the management as well as the orga-
Therefore, this planning field determines the nization of the production installation within the
arrangement of the facilities and considers not framework of the realization.
only the internal material flow, but also the con-
nections to suppliers and the distribution of the Planning Course
finalized products to their place of use. The con- The previously described planning fields are to be
cept of the material flow and the logistics is detailed in a structured planning process. For this,
compiled in interplay between technological, numerous suggestions have been developed, F
organizational, and logistic aspects, which are which are more or less similar and follow the
refined step by step. classic pattern of project planning. Examples
Equipment, staff, and material need a space, can be found in (Tompkins et al. 2010) and
which represents the fourth important reference (Schenk et al. 2010). The VDI guide line 5200
dimension of facility planning. The alignment divides the process of facility planning into seven
and space dimensioning of the facility subzones planning phases, Fig. 2.
are the task of the corresponding planning field. At the beginning, each facility planning
In turn, the spaces are the basis of the building requires a preparatory phase. It starts on the
concept and site planning. At this, higher-ranking management level with the target specification
aspects of the appearance, building site prepara- (phase 1). Mainly, this phase serves to specify the
tion, and later expansion possibilities are to be facility planning tasks and the basic structuring of
considered. Therefore, aspects of civil engineer- the project into work packages.
ing and architecture are at the fore in this plan- The following basis analysis (phase 2)
ning field. The form and type of the buildings, provides and evaluates the data and information
their internal and external design, their alignment relevant for the facility planning. The first step of
as well as the site development considering this planning phase is about obtaining informa-
outdoor facilities and also the scenic and urban tion. In the second step, the collected data is
integration of the facility as a whole are to be initially summarized and then evaluated with
regarded. Thus, the facility appearance becomes regard to the specified facility targets. Thus,
part of the company’s corporate identity. weaknesses of the current state are pointed out
When the entire concept has taken on clearer and the bases for the following planning steps are
outlines, the necessary auxiliary equipment, such provided, e.g., the production program, the num-
as service stations and testing facilities, the infra- ber of staff, and the rough space requirements.
structure for power and machine supply (electric In the concept planning phase (phase 3), the
power, water, pressurized air, steam), as well as fundamental solutions for the facility structure
the communication structure (local nets) have are created in an iterative approach. For this, the
to be planned. Special attention must be paid to functional and organizational units as well as
the minimization of power consumption as well their mutual correlations and their communica-
as the accumulation, storage, conditioning tion among each other are initially defined in the
and disposal of waste water and waste products. structure planning and visualized in an ideal
Finally a changeability which is appropriate function chart. Subsequently the production and
to the market environment has to be assembly structures as well as storage and trans-
realized (see ▶ Changeable Manufacturing). port concepts are to be specified based on this
Facility planning projects of a larger scale ideal function chart and on the communication
presuppose the cooperation with many internal concept. In the following step, spaces and logis-
and external partners and therefore require an tics facilities are dimensioned. Here, spaces are
F 496 Facility Planning

Facility Planning, Fig. 2 Phases of facility planning (VDI 5200)

based on the dimension of the necessary equip- transferred into scaled real layouts and
ment and staff capacities. To dimension the logis- construction drafts, considering all restrictions
tics installations, the function chart from the present. It is important to work out layout
structure planning is used. Based on this ideal variants, which are to be evaluated in detail
function chart, a spatially scaled function chart concerning their target performance under
is worked out in the first step of the ideal layout technological, logistic, and economic aspects.
planning, using the established production Within the next facility planning phase, the
spaces. Here, partial areas, such as the incoming detail planning (phase 4), all elements of the
goods department, the manufacturing, the assem- planned facility are described and graphically
bly, and the shipping area, are to be considered. displayed in detail. For this purpose, the fine
In the next step, the ideal alignment of the single layout planning initially deals with the develop-
functional areas is defined under consideration ment of machine positioning plans, the design
of material, staff, and communication flows of single workplaces, e.g., under ergonomic
(if necessary, also power and media flows). aspects, as well as with the detailed planning of
In this step, operational restrictions are usually material, information, and communication flows,
not taken into account. In the following step, the which are visualized with the aid of fine layouts.
real layout planning, the ideal layout is Modern, digital planning tools, such as the
Facility Planning 497 F
Steel Segment Stainless Segment

Stainless
Steel Carrier
Carrier

Stainless
Steel Cover
Cover

Stainless
Steel Housing
Housing

Manufacturing Sprinter

Assembly and Final Test

Shipment Area
Goods Entry

Service Segment
Testing

Storage
(Shaft, Short Shaft, Rotor)

Manufacturing Runner

Manufacturing Exotics

Office
20 m

Facility Planning, Fig. 3 Ideal rough layout of a pump factory

Digital Factory, even allow the visualization in such a way that the interruptions of the
of object movements (equipment, products, on-going production are reduced to a minimum.
staff, etc.) by means of an integrated simulation
as well as transmitting a visual impression of the Application
planned facility by using virtual 3D models. For a better understanding of the planning
After issuing the requests for approval, the detail systematics, the phase concept planning will be
planning concludes with the compilation of represented and explained using the example of
functional, respectively, detailed performance a factory producing pumps (Wiendahl et al. 2009).
descriptions. The regarded factory consists of an incoming
After the termination of the actual planning of goods department, a manufacturing area,
a facility, it is necessary to implement the plan- a service station, a testing area, an office,
ning results. This takes place in the phases real- a warehouse, an assembly area, and a shipment
ization preparation (phase 5), realization area (Fig. 3). After passing the incoming goods
supervision (phase 6), and ramp-up support department, the raw material is provided in four
(phase 7). Here, it is essential ensuring a tight manufacturing areas, which are divided into
project management concerning quality, costs, three product groups namely large-scale
and adherence to schedules, especially in the (sprinters), small-scale (runners), and single-
case of extensive projects. It is important to pre- item manufacturing (exotics). The sprinter
pare the official permit procedure at an early section is divided into one steel and one stainless
stage. After all, the relocation of the existing steel segment since the material must not be
facilities to the new location should be organized mixed. The service manufacturing station for all
 F
498

Facility Planning, Fig. 4 Development of a real rough layout (example pump factory)
Facility Planning
Facility Planning 499 F
Facility Planning,
Fig. 5 3D-model incl.
rough layout (example
pump factory) (Reichardt)

pump types is located between the sprinter and step consists of a merger of the site plan and the
the runner area together with the testing area. building concept with the ideal to the real layout.
The parts are produced in fixed lot sizes according For a better integration of the later users and
to prognosis and taken to the intermediate ware- to shorten the planning duration, a three-
house (make-to-stock) and from there on called dimensional visualization has proved itself.
forward by the assembly departement (make-to- Figure 5 shows the logistic and spatial concept
order). After the assembly, the pressure test of the of the pump factory with the administrative func-
finalized pumps is carried out. The pumps are tions integrated into the modular factory building
packed and provided for shipping in the shipment (Wiendahl et al. 2009, p. 478). With this concept
area. These correlations are initially displayed in a high changeability and intensive communica-
an ideal rough layout as shown in Fig. 3. tion between all employees of the order
A real rough layout is developed from the processing department are guaranteed.
ideal rough layout considering the construction
factors. In the first step, different alignment
variants of the single functional areas, consider- Cross-References
ing the required transport routes, are elaborated.
Figure 4 shows the development from the ideal to ▶ Changeable Manufacturing
the real rough layout using the example of the ▶ Factory
pump factory. ▶ Machine Tool
The known ideal rough layout from Fig. 3 is ▶ Manufacturing
represented in a more exact and scale-compliant ▶ Manufacturing System
form on the left side of picture. Production islands ▶ Production
have been installed for each component in the
manufacturing segments. They consist of an
U-shaped alignment of the machines, which are References
necessary for the according manufacturing step.
Furthermore, the restrictions, which have to be Schenk M, Wirth S, M€ uller E (2010) Factory planning
manual. Situation-driven production facility planning.
taken into account for the compilation of the real
Springer, Heidelberg
rough layout are itemized, e.g., site access, exten- Tompkins JJ et al. (2010) Facilities Planning. 4th edn,
sion possibilities, power supply, etc. The essential John Wiley & Sons
F 500 Factory

VDI-Richtlinie 5200: Fabrikplanung. Planungsvorgehen. Extended Definition

Beuth Verlag Berlin 2009 (VDI Guide Line factory A factory is a local grouping of production factors
planning. Planning procedere. VDI The Association of
German Engineers Berlin 2009) for the realization of the entire or a part of the
Wiendahl H-P (2010) Betriebsorganisation f€ur Ingenieure value chain of real goods. The main production
(Company organization for engineers), 7th edn. factors are material, facilities, work force,
Hanser, M€ unchen energy, and information.
Wiendahl H-P, Reichardt J, Nyhuis P (2009) Handbuch
Fabrikplanung. Konzept, Gestaltung und Umsetzung
wandlungsf€ahiger Produktionsst€atten (Handbook
factory planning. Concept, design and realization of Theory and Application
changeable production sites). Hanser, Munich
History
The history of factories can better be described
from the viewpoint of their main change drivers.
Factory Together with the increasing variety of the
product comes the steady permeation of techno-
George Chryssolouris1, Hans-Peter Wiendahl2, logical developments: new materials and
Loukas Rentzos1 and Sotiris Makris1 manufacturing processes, information and com-
1
Laboratory for Manufacturing Systems and munication technology (ICT), Internet applica-
Automation (LMS), Department of Mechanical tions, RFID (radio frequency identification
Engineering and Aeronautics, University of devices), and virtual reality, open up new
Patras, Patras, Greece possibilities for both the design engineer and the
2
Institut f€
ur Fabrikanlagen und Logistik, Leibniz factory planner. Furthermore, there may be
Universit€at Hannover, Garbsen, Niedersachsen, a different life cycle of factory elements such as
Germany processes, buildings, or sites compared to
a product life cycle produced in the factory,
while the time available for enterprises to react
Synonyms to market changes decreases.
Developments in the last decade include
Manufacturing plant; Production site concepts such as lean production (Womack
et al. 1990), business reengineering (Hammer
and Champy 1993), and the fractal factory
Definition (Warnecke 1998):
• Products and their production programs were
Factory, manufacturing systems, and production broken down into components, modules, and
are closely related terms and often are not subsystems, and key competencies concen-
distinguishable, neither in literature nor in indus- trated on them. In-house manufacturing was
try. In the following essay, manufacturing and then drastically reduced, while orders for
production will be seen as functions for the gen- required items were shifted to external
eration of products (Chryssolouris 2006), suppliers. In addition, the workforce was
whereas the factory represents the physical and considerably reduced as well.
logical means of performing production and • The entire procurement logic was restructured
manufacturing processes. An industrial in order to be differentiated and accelerated.
manufacturing site includes the buildings, the Supplies were delivered directly to the site and
manufacturing facilities, and the ground on permanent relationships with the suppliers of
which they are located (CIRP Dictionary of modules and systems were developed.
Production Engineering 2004). A further example was the allocation of
Factory 501 F
Factory,
Fig. 1 Influencing
Planning Processes Supply
parameters on a factory
and Control Market

Markets
Facilities
Factory Products

Finance/
Material Flow
Economy

Resource Law F
Efficiency Regulations
Work Force

a complete spectrum of the so-called C-parts, Influencing Parameters

namely, parts that are only worth 5–10 % of A factory is the subject of many influencing
the total product value but make up to 80 % of parameters which determine their configuration
the number of parts, to a logistics service and explain why any factory is unique (Fig. 1)
provider. (Wiendahl et al. 2009).
• The direct value-adding area of the The products to be manufactured and their
manufacturing and assembly was restructured markets, the work force available, the local laws
into segments and was decentralized. Based and regulations, and the proposed processes
on the 1960s group technology (Mitrofanow determine typically the necessary facilities, the
1960) and the 1970s/1980s manufacturing material flow from the suppliers through the
cells, the concepts of modular factories factory to the customers, and the approach to
(Wildemann 1998) and fractal factories planning and controlling the orders. Special
(Warnecke 1998) were created. The general attention must be given to the efficient use of
idea behind it was that part or component material and energy.
groups, requiring similar manufacturing or
assembly technology, be produced as one Factory Components
unit for a market segment with specific Factory resources could be summarized under the
demands regarding delivery times and reli- heading facilities, organization, and employees
ability. Functions such as material and tool (Wiendahl et al. 2009).
planning, scheduling, servicing, and mainte- Facilities are grouped into core and support
nance were integrated into such a unit. functions. Core functions transform the raw
• Factories can be integrated into a network of material into final components (manufacturing)
enterprises: Companies join together into and products (assembly) and, thus, create an
a virtual enterprise, appearing from the outside added value. Support functions are necessary
like a large enterprise and offering all of the but do not add value. These are logistic facilities
services from “one place.” Such a concept (mainly storage and transport equipment) and
allows small and midsize enterprises to suc- ICT (Information and Communications Technol-
cessfully bid and develop large projects at low ogy) Systems. Their work schedule determines
overhead costs. the output of the factory. The dominating
F 502 Factory

2 2

6 – 10 m
1 2 1 2
20 – 100 m
2
1

7 – 20 m
6–7m

2
1 1

10 – 20 m 20 – 100 m

flat buildings hall buildings combination forms

e.g. - office e.g. - refinery
- store/production

2 1.2
2
- high shelf store
4.5 m
3m

1
1 1 horizontal
12 – 15 m 18 – 25 m
2 vertical

storey buildings special buildings expansion directions

Factory, Fig. 2 Industrial buildings (Reichardt and Wiendahl 2009)

facilities are those for manufacturing, assembly, Buildings are – as already mentioned – for
and logistics since they mostly determine due to many reasons necessary for the performance of
their size, weight, and energy consumption the the processes and the protection of the employees
factory content. and products from outer disturbances such as
Employees are grouped according to their qual- weather conditions and unauthorized people.
ification level, which in turn is determined by the The buildings are statically structured, the outer
technical level of the products, processes, and shell, the media routes, and the building
equipment regarding complexity, precision, and equipment for air conditioning, lighting, IT, and
variation. The working conditions, namely, ergo- energy provision. Typical forms of production
nomic, room climate, lighting, and safety, take building are depicted in Fig. 2. The buildings
care of the human needs and health. Finally, the are usually one storey, but if the land price is
wage system (pay by hour or by performance) and very high or light weight, products are to be
the incentive structure, in the long term, are essen- manufactured there, then multi-storey buildings
tial for the quality of the output. are also utilized.
The organization determines, on the one hand, The site very much determines the material
the hierarchy in which the factory is structured supply into and from the factory, as well as
and corresponds to the layout levels. On the other factors, namely, climate, wages, and taxes.
other hand, the business processes have to be The selection is a process on its own and is
documented, often in quality handbooks. Part of dominated by long-term aspects.
the organization is finally the job assignment of
factory tasks, at a certain qualification level, Virtual Factory
of the employees and the grouping into wage The incorporation of virtual reality (VR) tech-
classes. nology for the digital simulation of the modern
Factory Design 503 F
operations, at factory level, has been manufacturing: history, perspectives, and outlook.
progressing rapidly in the last decade. It is Proc Inst Mech Eng B J Eng Manufact 222(5):451–462
Chryssolouris G, Papakostas N, Mavrikios D (2008b)
mainly used for virtually planning the layout A perspective on manufacturing strategy: produce
and processes of a factory, simulating tasks or more with less. CIRP J Manuf Sci Technol 1:45–52
evaluating alternative designs prior to the CIRP Dictionary of Production Engineering (2004)
start of production (Chryssolouris et al. 2008a). Manufacturing systems, vol 3, 1st edn. Springer, Berlin
Hammer M, Champy J (1993) Reengineering the corpo-
Using such solutions, the industry has ration: a manifesto for business revolution, 1st edn.
implemented virtual factory systems that allow Harper Business Books, New York. German edition:
greater reuse of engineering data, better control Business reengineering: Die radikalkur f€ ur das
of late engineering changes in the design cycle, unternehmen, so erneuern sie ihre firma, 7th edn.
Frankfurt am Main, 2003
and more sophisticated simulations of NC Jovane F, Yoshikawa H, Alting L, Boër CR, Westkamper E,
machining processes and factory-floor layouts Williams D, Tseng M, Seliger G, Paci AM (2008) The F
(Chryssolouris 2006). incoming global technological and industrial revolution
towards competitive sustainable manufacturing. CIRP
Ann Manuf Technol 57(2):641–659
Factory of the Future Mitrofanow SP (1960) Wissenschaftliche grundlagen der
The factory of the future is visualized as a factory gruppentechnologie [Scientific fundaments of group
with enough intelligence to adapt with flexibility technology], 2nd edn. VEB Verlag Technik, Berlin
to various market trends and demands or unex- (in German)
Ohno T (1988) Toyota production system: beyond large-
pected events, while maximizing the production scale production [English translation]. Productivity
performance and minimizing the environmental Press, New York
footprint of its operation, by utilizing advanced Reichardt J, Wiendahl H-P (2009) Changeable factory
ICT solutions and sustainable methods and buildings – an architectural view. In: Elmaraghy H
(ed) Changeable and reconfigurable manufacturing
strategies (Jovane et al. 2008). It will also have to systems. Springer, London, pp 389–401
invest heavily on the knowledge existing in the Warnecke H-J (ed) (1998) Aufbruch zum Fraktalen
system, especially the part that has to do with unternehmen – praxisbeispiele f€ ur neues denken
humans. The technologies to be developed and und handeln [On the move to the fractal enterprise –
examples for novel thinking and action]. Springer,
used for the factory of the future are ICT-based Berlin (in German)
systems that aim to increase the performance and Wiendahl H-P, Reichardt J, Nyhuis P (2009) Handbuch
the adaptability of the production, while consider- fabrikplanung. Konzept, gestaltung und umsetzung
ing in real time the environmental aspects and the wandlungsf€ahiger produktionsst€atten [Handbook
factory planning. Concept, design and realization of
quality of services (Chryssolouris et al. 2008b). changeable production sites]. Hanser, Munich
(in German)
Wildemann H (1998) Die modulare fabrik – kundennahe
Cross-References produktion durch fertigungssegmentierung [the mod-
ular factory – product segmentation as an approach to
customer-oriented manufacturing], 5th edn. TCW
▶ Changeable Manufacturing transfer-centrum f€ ur produktions-logistik und
▶ Machine Tool technologie-management [TCW Transfer Center for
▶ Manufacturing Manufacturing Logistics and Technology] GmbH,
M€ unchen
▶ Manufacturing System Womack JP, Jones DT, Roos D (1990) The machine that
▶ Production changed the world: based on the Massachusetts
▶ System institute of technology 5-million dollar 5-year study
on the future of the automobile. Rawson Associates,
New York
References

Chryssolouris G (2006) Manufacturing systems: theory

and practice, 2nd edn. Springer, New York
Factory Design
Chryssolouris G, Mavrikios D, Papakostas N, Mourtzis D,
Michalos G, Georgoulias K (2008a) Digital ▶ Facility Planning
F 504 Factory Planning

processes and is aimed to fabricate boreholes of

Factory Planning high quality to serve engineering fits. The
reaming process especially improves the surface
▶ Facility Planning roughness, diameter, and form tolerance of
a borehole, whereas the hole location will not
necessarily be effected. In preparation of the
reaming process, the holes have to be predrilled
Feed Drives or core drilled. The process can be divided
into round and profile reaming.
▶ Computer Numerical Control

Theory and Application

Feedback Control System History

Fine finishing of holes was introduced to ensure
▶ Servo System fits to mating of two mechanical components as
either clearance, interference, or transition fit
(ISO 286-1, ISO 286-2, ISO-R286 and ANSI
B4.2-1978). Besides the multi-flute fine finishing
Fine Boring tools were known for a long time, the guiding
principles for single-blade reamers were inspired
▶ Fine Finishing of Holes by the concept of deep hole drilling application
with auto-guiding functionality already
established in the seventeenth century to produce
canon and gun shafts of high quality with
Fine Ceramic (Japan) improved roundness and straightness. The
research investigations in the twe7ntieth century
▶ Ceramic Cutting Tools helped in understanding the working mechanisms
of the deep hole drilling process and thus the
usage of guiding pads. In the 1970s, the fine
boring tool or single-blade reamer was developed
Fine Finishing of Holes (Kress 1975). The single-blade reamer became
a tool for accurate and economic bore machining
Christoph Gey in mass production (Koppka 2009).
Kennametal Shared Services GmbH,
F€urth, Germany
Kinematics

Synonyms The kinematics of reaming is comparable to

the drilling process that incorporates a rotational
Fine boring; Precision machining of holes; movement and a linear movement in
Reaming axial direction of the tool (Fig. 1). Reaming
is basically a counter bore operation with thin
chips. The target is to produce high precise
Definition holes in diameter with excellent surface and
form quality.
Fine finishing of holes or reaming process The reaming process is distinguished among
belongs to the high-precision machining three phases starting with the entering of the tool
Fine Finishing of Holes 505 F
Fine Finishing of Holes, Table 1 Cutting parameters –
recommended feed rates are between 0.1 and 0.25 mm per
tooth
Cutting speed in mm*min
1
Cutting grade
Workpiece material Cermet Carbide
Steel 100–210 20–120
Cast iron 130–210 30–110
Stainless steel 10–20

Classification F
Basically, reamers are divided into hand
reamers and machine reamers. More important
for industrial applications are the machine
reamers.
Three different groups of reamers exist:
1. Multi-Flute Reamers, Nonadjustable (Fig. 2)
These reamers have more than one cutting
edge. The reaming diameter is created during
the manufacturing process of the reamer by
a grinding operation and cannot be changed
Fine Finishing of Holes, Fig. 1 Reaming process
afterward. The guidance of the reamer during
the cutting process is guaranteed by the
in the hole. The second phase describes the machine spindle. The targeted hole tolerance
reaming of the entire bore and the third phase is IT7 and better.
the exiting out of the hole. 2. Multi-Flute Reamers, Adjustable (Fig. 3)
The machining process is mainly determined These reamers have also more than one
by the cutting geometry in the front of the reamer. flute and have an incorporated expansion
The geometry is mostly responsible for the mechanism. This mechanism was developed
diameter, hole geometry, surface quality, and to compensate the tool wear during the cutting
chip evacuation. process. The diameter of the tool can be
expanded by an internal cone to readjust the
requested tool diameter. The expansion
Tool Materials mechanism works in the area of elastic
deformation of the tool body material.
As cutting materials solid HSS, cemented car- The expanding mechanism is always of high
bide, and cermets are used. In addition brazed precision. An adjustment of the tool diameter
cemented carbide and cermets as well as PCD, within 2 mm is possible. The targeted hole
PCBN tipped insert versions on solid HSS or tolerance is IT6 and better.
cemented carbide base bodies are common 3. Single-Blade Reamers with Guiding Pads,
(Table 1). Adjustable (Fig. 4)
Guide pads usually consist of cemented In addition to one cutting insert, these
carbide, cermets, or PCD. reamers have guiding pads. Guiding pads
Most of the cutting materials are coated. support the reamer into the hole. The cutting
The coating increases the tool life and prevents insert is always fine adjustable. The adjust-
the tool wear. ment function allows adjusting the tool
F 506 Fine Finishing of Holes

Fine Finishing of Holes,

Fig. 2 Multi-flute reamers

holes require straight-fluted reamers and through

holes require helical-fluted reamers to do the
optimal chip evacuation. Other hole criteria are
straightness, roundness, cylindricity, and
length of the hole. These are the basis to select
a single-blade reamer or a multi-blade reamer

Fine Finishing of Holes, Fig. 3 Multi-flute reamer

adjustable Single-Blade Reaming

The major benefit of padded reamers is that

extremely high ratios of length to diameter can
be achieved due to the guide pads supporting
the tool as it moves down the bore. In effect,
these tools support themselves on the bore they
have just cut, so the deeper the bore the better
support.

Multi-Blade Reaming

Multi-blade reamers incorporate a high number

of cutting edges, mainly limited by diameter of
Fine Finishing of Holes, Fig. 4 Single-blade reamer the tool and necessary space for the chip gashes.
with guiding pads Based on the increased numbers of teeth,
these tools run with a much higher feed rate
diameter and the clearance of the cutting per revolution. The advantage is the increased
insert. The purpose of these reamers is to productivity.
machine highly precise holes with tolerances
around IT5. Special designs of these tools
have more cutting insert in order to increase Determining the Hole Tolerance
the productivity.
In most of the cases the permissible hole
tolerance is described by the ISO tolerance
Tool Selection system, e.g., IT7 range. Reamers are ground
within the desired range of tolerance depending
The hole type, blind or through hole, the required on the application. Modern adjustable reamers
hole diameter tolerance, and surface specification can be adjusted by the user within micron ranges
mainly determined the tool selection. Blind on the diameter.
Fine Grinding 507 F
Reaming Difficulties

Deviations from the required hole quality can

be introduced by various process affecting factors
Fine Finishing of Holes, Fig. 5 Reamer with steerable
like machine tool stiffness, work piece clamping,
holder
material and geometry, cutting tool stiffness,
geometry, manufacturing tolerance of the cutting
tool itself, coolant conditions, or process with the largest market being automotive.
parameters. It is also important that the reamers Typical automotive applications are as follows:
run as accurately as possible, as any run-out Dowel holes
errors at the cutting edge could result in oversize Cam bores
bores, poor surface finish, or poor geometry Crank bores
Valve guide
F
(taper, straightness, etc.). To eliminate this
potential error, the use of steerable holders is Tappet bores
recommended on machining centers. By means Cylinder bores
of radial and axial adjusting screws, errors in both Gearbox bearing bores
the axial and radial directions can be completely Taper bores in steering knuckles
avoided. If reaming on a lathe, the use of “floating Hydraulic and pneumatic valve blocks require
holders” is recommended. Small alignment the use of high precision reamers in the production
errors between the turret centerline and chuck/ of the spool bores. The mating spool is also manu-
hole centerline create, in effect, run-out errors. factured to extremely close tolerances, resulting in
The result is the same as in a machining center, an almost perfect fit between spool and bore.
oversize bores, poor surface finish, and geometry.
Modular Reamers with steerable holder allow the Cross-References
perfect alignment of turret centerline to chuck
centerline, resulting in good bores (Fig. 5). ▶ Cutting Fluid
▶ Cutting, Fundamentals

Coolants
References
Coolants can have a large effect on the outcome
Koppka F (2009) A contribution to the maximization of
of a reaming application. Coolants have two productivity and workpiece quality of the reaming
functions: (1) to wash the chips away from the process by analyzing its static and dynamic behaviour:
cutting edge and (2) to “coat” the tool in an analysis with focus on automotive powertrain pro-
duction Dissertation Aachen: Shaker Verlag
a lubricant to stop microparticles welding them-
Kress D (1975) Reiben mit hohen schnittgeschwin-
selves to the reamer cutting edge. This buildup on digkeiten: untersuchung der möglichkeit beim reiben
the cutting edge stops the reamer from cutting die schnittgeschwindigkeit an der erheblich höheren
correctly and usually results in poor surface finish schnittgeschwindigkeiten anderer spanender verfahren
anzupasse [Reaming with high cutting speed: investi-
in the bore. In padded reamers, the lubricant in
gations on adapting the cutting velocity to reasonable
the coolant removes the friction between the pads higher speeds of other cutting processes]. Vogel,
and the bore. W€urzburg, (in German)s

Reaming Applications
Fine Grinding
Reaming applications can be found in all
industries needing to produce mechanical fits, ▶ Honing
F 508 Finite Element Method

characterizing the problem over the whole

Finite Element Method domain. These eventually lead to algebraic equa-
tion sets to be solved with the aid of digital
A. Erman Tekkaya1 and Celal Soyarslan2 computers in the form of
1
Institute of Forming Technology and
Lightweight Construction, Technical University ½Kfug ¼ fFg (1)
of Dortmund, Dortmund, Germany
2
Institut f€
ur Umformtechnik und Leichtbau, In above [K] is named as the stiffness matrix,
Technische Universit€at Dortmund, Dortmund, whereas {F} denotes the load vector. This form of
Germany the equation set, where the unknown vector {u}
can be explicitly found, is only handled for linear
analysis. The nonlinearities associated with the
Synonyms nature of the manufacturing processes usually
end up with a nonlinear set of equations with
Rayleigh-Ritz-Galerkin methods
½KðfugÞfug ¼ fFg (2)

Definition where an isolated representation of {u} is not

possible. These kinds of implicit equations are
A numerical method used to approximate the usually solved using iterative schemes, e.g.,
solution of boundary- and initial-value problems Newton–Raphson method. The finite element
characterized by partial differential equations. methods admitting equations of type 1 are
named as linear, whereas those resulting in
equations of type 2 are named as nonlinear finite
Theory element methods. A detailed mathematical
treatment of the method can be found in Oden
The finite element method is a systematic pro- and Reddy (1976).
cedure of approximating continuous functions Typical steps of the finite element analysis
as discrete models. This discretization involves involve:
finite number of points and subdomains in the • Discretization of the problem domain into
problem’s domain. The values of the given finite elements and nodes
function are held at the points, so-called nodes. • Formation of element equations by local
The non-overlapping subdomains, so-called approximation
finite elements, are connected together at nodes • Assembly of element equations to arrive
on their boundaries and hold piecewise and local structural equations
approximations of the function, which are • Modification and solution of emerging
uniquely defined in terms of values held at equations with boundary conditions
their nodes. The collection of discretized ele- • Interpretation of results
ments and nodes is called the mesh and the The nodal functional values can be various
process of its construction is called meshing. depending on the problem solved. Although they
A typical finite element partition of a two- do not have to have a physical meaning in general,
dimensional domain with triangular finite ele- in manufacturing engineering applications, these
ments is given in Fig. 1. may refer to the displacements, velocities,
The boundary- and initial-value problems are pressures, or displacement gradients, e.g., in
formulated in a weak form (or integral form) metal forming applications, temperature, e.g., in
which demands weaker continuity conditions, so heat treatment applications.
that the contributions of each finite element to the The number of nodes used in the discretization
global integrals sum up to give an integral defines the number of unknowns. At the
Finite Element Method 509 F
a b F1 c
F2
Ωd Node 2
Ω
(X2, Y2)
i Node 1

Ωk (X1, Y1)
y

(X3, Y3)
Node 3
x

Finite Element Method, Fig. 1 A 2D problem domain O, its finite element partition Od into triangular elements, and F
regarding force and displacement boundary conditions

industrial-level applications, the size of the Special Elements

model or its refined discretization may result in • Contact elements
large sets of simultaneous algebraic equations to • Tying elements
be solved. Thus development of numerical Once the problem domain and element types
methods after robust and efficient solution strat- are given, it comes to discretization of the problem
egies constitutes an active branch of research. domain. Efficient and automated discretization
Approximating the function locally, generally process for construction of the finite element
by polynomials, over the element gives possibil- meshing for random domains is important.
ity to a disjoint treatment of finite elements by
isolating them from the mesh no matter what their
connectivity or location in the problem domain is. Applications
Accordingly, leaving the nodal values of the local
approximation arbitrary, an independent con- Its robustness and flexibility gives rise to the appli-
struction of a catalog of finite elements is possible cability of the method to quite general domains,
for one-, two-, or three-dimensional domains. e.g., very complicated geometries, which is one of
One possible classification of the elements the main reasons behind penetration of the method
among the many is the following (Fig. 2). to all areas of engineering and applied sciences.
Today finite element method is used in many
Continuum Elements industrial and applied fields involving:
• 1D • Structural mechanics
• 2D • Thermodynamics
– Plane stress • Electrical and magnetic systems
– Plane strain • Fluid dynamics
– Axi-symmetric • Acoustics
• 3D Nearly in all branches of manufacturing
industry, application of the finite element
Structural Elements analysis is due. Together with various references,
• Spring elements these involve:
• Dashpot elements • Casting; see, e.g., Lewis and Ravindran
• Beam elements (2000).
• Plate elements • Forming of metals, rubbers, polymers, and their
• Membrane elements composites; see, e.g., Tekkaya (1998a, b),
• Shell elements Wagoner and Chenot (1997).
F 510 Finite Element Method

Finite Element Method, 3

Fig. 2 Certain 2D and 3D 3
y y
continuum elements 6
5
1 1
2
2 4

x x
6
y 4 y 4
3 3

7 5

1 2 1 2
8
x x
7
6

y 8
4 3 5
y
4 3

2
1
x 1 2
z x
z

• Cutting, machining, and grinding; see, – Reduction of die costs

e.g., Childs et al. (2000). – Press downsizing
• Heat treatment; see, e.g., Şimşir and G€ur (2010). – Increase of reliability
• Welding; see, e.g., Lindgren (2001). • Increase of product quality
• Bonding; see, e.g., De Hoff et al. (1995). – Optimal selection of the workpiece material
• Electronics assembly and packaging; see, – Production of more complicated parts
e.g., Agwai et al. (2009). – Know-how accumulation for new materials
• Micro- and nanofabrication; see, e.g., Luttge – Process repeatability
(2011). – Optimization by variants
The industrial goals of the forming simulation Although today the analysis of manufacturing
with finite element method can be summarized in processes with finite element methods reached
three main groups: a level that allows the accurate estimation of
• Time reduction process parameters, the successful application
– Early checking of producibility of still requires the following:
workpieces • The existence of a well-defined physical
– Reduction of the development times problem, for which a numerical analysis can
– Reduction of the tryout times provide a solution
– Quick response to modification wishes • The correct idealization of this physical
• Cost reduction problem (simplifications, assumptions,
– Cheaper products detection of governing physical phenomena)
Finite Element Method 511 F
• The correct spatial discretization of the Finite Element Method, Table 1 Comparison of
idealized problem (type of elements, topology implicit and explicit approaches
of element mesh, density of element mesh) Solution approach Static implicit Dynamic explicit
• The setting of correct boundary conditions Physics Natural Artificial
(friction, heat transfer, machines, dies) formulation formulation
• The use of correct material laws and parame- Computational speed Low High (with mass
scaling)
ters (flow curve, anisotropy, failure)
Storage requirements High Low
• The selection of correct numerical parameters
Robustness May diverge High
(penalty factors, convergence limits, Accuracy control Convergence No control
increment sizes, remeshing criterion) check
• The economic analysis (reasonable computa- Wrinkling Bifurcation By-product
tional times, reasonable modeling times, problem F
reasonable storage requirements) Spring back Feasible Not feasible
• The correct interpretation of the numerical
results
in forming space filling three-dimensional
discretizations are given in Table 2.
Metal Forming Basically, there are two possible approaches:
using tetrahedral or hexahedral elements. Due to
Finite element method is used extensively in the Brezzi-Babuska condition, the tetrahedral
the analysis of manufacturing processes. elements must be at least quadratic. The basic
Forming applications constitute a leading role advantages of the usage of tetrahedral elements
in development of the nonlinear finite element are the simplicity to automate the meshing
analysis. The nonlinearity sources stem from process and the flexibility to, for instance, locally
geometric nonlinearities due to large displace- refine the existing mesh. On the other hand,
ments and large strains, material nonlinearities serious drawbacks of tetrahedral elements are
due to nonlinear relation between stress and the increasing complexity of the contact handling
strain tensor for metallic materials on the course (at least quadratic elements!), the high number of
of loading within the deformation range, degrees of freedom resulting, and the poor ability
and finally contact nonlinearities due to of these elements to withstand severe distortions
evolving boundary conditions with tool and (tetrahedral elements are in a view degenerated
workpiece interactions. Majority of the metal hexahedral elements). Hexahedral elements can
forming processes are quasi-static processes be linear. Therefore, they do not exhibit the
where the emerging equations are solved within difficulties in the contact handling and, further-
a static-implicit procedure. Nevertheless, due to more, they can endure much severer distortions
various advantages dynamic-explicit schemes as compared to tetrahedral elements. The
are also used especially in sheet metal forming meshing of these elements can be performed in
community. Two approaches are compared in two distinct ways: Incompatible meshes, such as
Table 1. used very extensively in sheet metal forming
In the following the selection of element processes in a two-dimensional version, can be
technologies in finite element analysis of forming automatically generated simply. They can be also
processes is given. In bulk forming only locally refined easily. The available element
continuum elements are used. In addition to ini- types for the finite element simulation of sheet
tial domain meshing, remeshing is required for metal forming processes are given in Table 3.
especially the simulation of certain bulk forming The high surface to volume ratio of the
processes for handling of severely distorted workpiece in sheet metal forming gives the
elements in the three-dimensional space. From opportunity of using structural elements, which
a systematic point of view, the possible methods make use of the plane shape of the sheet Wang
F 512 Finite Element Method

Finite Element Method, Table 2 A systematic view of mesh generation for three-dimensional domains
Element type Tetrahedral Hexahedral
Least element order Quadratic (Babuska-Brezzi) Linear
Remeshing – Compatible Incompatible
Automation Simple Difficult Simple
Local refinement Simple Difficult Simple
Contact handling Difficult Simple Simple
Number of DOF High Low Low
Distortion tolerance Low High High

Finite Element Method, Table 3 Element types used in sheet metal forming simulations

Membrane Bending Transverse shear stiffness Transverse

Element type Stiffness Stiffness Kirchhoff Mindlin Strain variation
Membrane Yes No No No
Plate No Yes No Yes No
Shell (thin) Yes Yes No Yes No
Shell (thick) Yes Yes No Yes Yes
Continuum Yes Yes Yes Yes

and Budiansky (1978). Hence, there are four dif- Example Applications for Metal Forming
ferent types of elements available for the analysis
of sheet metal forming processes. In sheet metal In order to demonstrate the use of finite element
forming processes for which bending occurs over method in metal forming practice, two examples
a bending radius which is larger than roughly ten are selected. The first example belongs to the
times the sheet thickness, membrane elements sheet metal forming and the second one the bulk
can be used for modeling. Generally, however, metal forming.
if deep-drawing type of deformation dominates, An application of dynamic-explicit finite
experience has shown that membrane elements element method in the prediction wrinkling of
are not appropriate for modeling. Therefore, the metallic sheets during deep-drawing process is
most common element used for deep-drawing- given in Fig. 3. As seen in the figure, in correla-
dominated forming processes is the shell ele- tion with the experimental findings, steel sheets
ment. These elements can be divided into two experience a relatively early wrinkling as com-
groups: shell elements, which can (thick) and pared to the aluminum sheets during the forming
cannot (thin) consider a transverse pressure process.
appropriately. Transverse pressure is essential if As an example for a bulk metal forming
blank holder actions on the general deformation process with thermomechanical coupling, we
behavior shall be considered accurately. Contin- present the forming process under induction
uum elements are seldom used for sheet metal heating. For a more detailed treatment, the reader
modeling. On the other hand, in sheet forming is referred to Okman et al. 2006. For hollow
processes such as hydroforming or blanking, con- shafts made of material X5CrNi189 with initial
tinuum elements can be appropriate. The approx- dimensions as D0 ¼ 30 mm, Di ¼ 22 mm
imation functions of these elements can be lower L0 ¼ 60 mm Li/L0 ¼ 0.75, the numerical pre-
and higher order. Besides the element edges can dictions and their comparison with the experi-
be curved or straight. ments are given in Fig. 4 for heating time 27 s.
Finite Element Method 513 F
Thickness <mm>

1.180
1.160
1.140
1.120
1.100
1.080
1.060
1.040
1.020
Aluminum 90 mm Steel 90 mm 1.000 F
10000 N BHF 6517 N BHF 0.980
min. thickness: 0.944 mm min. thickness: 0.960 mm 0.960
max. thickness: 1.226 mm max. thickness: 1.153 mm

Finite Element Method, Fig. 3 Simulation results for deep drawing of aluminum and steel blanks

Folding
Buckling

Finite Element Method, Wavy shape of

Fig. 4 Comparison of the
numerical results with the the side walls
experiments for induction
heating

As the figure depicts many features of the actual the number of divisions. Formal devising of the
experiments, such as buckling, folding, and wavy method in solution of engineering problems dates
shape of the side walls, can be successfully back to 1943 where an assembly of triangular
predicted for this coupled thermomechanical elements on which linear warping function is
problem. defined is used to represent domain of St. Venant
torsion problem solved by principle of minimum
potential energy. A complete historical account
Early History can be found in Oden (1971). Coining of the term
finite elements goes back to 1956, emerging as
The basic concept relies on the ancient geometri- a method to conduct stress analysis of aircraft
cian’s approximation of circle by a polygon structures. The finite element method has been
where accuracy of approximation depends on established basically through the efforts of
F 514 Fitting

Courant (1942), Argyris (1954), Turner et al. Oden JT (1971) Finite elements of nonlinear continua.
(1956), Clough (1960), and Zienkiewicz and McGraw-Hill, New York
Oden JT, Reddy JN (1976) An introduction to the mathe-
Cheung (1965) in the years 1940 through 1965. matical theory of finite elements. Wiley, New York
Use of finite element analysis in metal forming Okman O, Özmen M, Huwiler H, Tekkaya AE (2006) Free
processes started with the pioneering studies forming of locally heated specimens. Int J Mach Tool
of Lee and Kobayashi (1973) which utilized rigid Manu 47(7–8):1197–1205
Şimşir C, G€
ur CH (2010) Simulation of quenching process
plasticity. Among pioneering studies on large strain for prediction of temperature, microstructure and
elastoplastic formulations, one can refer to residual stresses. J Mech Eng 56(2):93–103
McMeeking and Rice (1975) and Nagtegaal (1982). Tekkaya AE (1998a) State-of-the-art of simulation in
metal forming. In: 6th SheMet conference, University
of Twente, Enschede/The Netherlands, 6–8 Apr 1998,
vol I, pp 53–66
References Tekkaya AE (1998b) Status and developments in the
simulation of forming processes. WIRE 48(1):31–36
Agwai A, G€ uven I, Madenci E (2009) Damage prediction Turner MJ, Clough RW, Martin HC, Topp LJ (1956)
for electronic package drop test using finite element Stiffness and deflection analysis of complex structures.
method and peridynamic theory. In: Electronic com- J Aeronaut Sci 23(9):805–823
ponents and technology conference, 2009. ECTC Wagoner RH, Chenot J-L (1997) Fundamentals of metal
2009. 59th, 565–569 forming. Wiley, New York
Argyris JH (1954) Energy theorems and structural Wang N-M, Budiansky B (1978) Analysis of sheet metal
analysis: a generalized discourse with applications on stamping by a finite-element method. Trans ASME
energy principles of structural analysis including the J Appl Mech 45:73–82
effects of temperature and non-linear stress–strain Zienkiewicz OC, Cheung YK (1965) The finite element
relations part I. Aircr Eng Aerosp Technol method in continuum and structural mechanics.
26(10):347–356 McGraw-Hill, New York
Childs THC, Maekawa K, Obikawa T, Yamane Y (2000)
Metal machining: theory and applications. Arnold,
London
Clough RW (1960) The finite element method in plane Fitting
stress analysis. In: Proceedings of ASCE 2nd confer-
ence on electronic computation, Conference papers
American Society of Civil Engineers 2nd conference ▶ Assembly
on electronic computation, 8–9 Sept 1960, Pittsburgh
Courant R (1942) Variational methods for the solution of
problems of equilibrium and vibrations. Bull Amer
Math Soc 49:1–23
De Hoff PH, Anusavice KJ, Wang Z (1995) Five-Axis
Three-dimensional finite element analysis of the
shear bond test. Dent Mater 11(2):126–131 ▶ Machine Tool
Lee CH, Kobayashi S (1973) New solution to rigid-plastic
deformation problems using matrix methods. Trans
ASME J Eng Ind 95(3):865–873
Lewis RW, Ravindran K (2000) Finite element simulation
of metal casting. Int J Numer Methods Eng Five-Axis Tool Path Generation
47(1–3):29–59 (Special Issue: Richard H. Gallagher
Memorial Issue)
Lindgren L-E (2001) Finite element modeling and Bert Lauwers
simulation of welding. Part I: increased complexity. Production Engineering, Machine Design and
J Therm Stress 24(2):141–192 Automation (PMA), KU Leuven, Heverlee,
Luttge R (2011) Microfabrication for industrial
Belgium
applications. William Andrew, Oxford
McMeeking RM, Rice JR (1975) Finite-element
formulations for problems of large elastic–plastic
deformation. Int J Solids Struct 10:601–616 Definition
Nagtegaal JC (1982) On the implementation of inelastic
constitutive equations with special reference to large
deformation problems. Comp Meth Appl Mech Eng In order to produce parts with a complex-shaped
33(1–3):469–484 geometry, milling machines have to be
Five-Axis Tool Path Generation 515 F

CAM NC Post-
CAD CLData processor
Tool path G-code Machining
Part design file
generation G-code

Five-Axis Tool Path Generation, Fig. 1 Process flow

programmed in a way that milling tool Theory and Application

movements result in removing all necessary F
excessive material. Starting from a CAD Since the input is a CAD file, and the output is
model, Fig. 1 illustrates the main phases of pre- a machine-/controller-independent tool path data,
paring the NC-programs for NC machining. tool path generation is mainly responsible for the
Based on the CAD model, the CAM software calculation of the tool trajectory.
generates a tool path, a description of tool move-
ments in the part coordinate system, depending Theory
on the part geometry and settings defined by the CAD Model Representation
CAM engineer. The output of the CAM system The choice for a certain surface representation
is a CLDATA-file (controller independent) system will have far-reaching consequences for
which in a next step is converted by an the capabilities and performance of the tool path
NC-postprocessor to an NC-program. The latter generation algorithm. In general, there are three
describes the tool movements within a machine different surface representation techniques:
coordinate system. The postprocessor is also polynomial representation, faceted models, and
responsible for converting commands into a Z-map models. For five-axis tool path generation,
format that is supported by the exact controller only the two first are the most important and
of a machine at which the tool path will be described below.
executed. The output of the postprocessor is a Polynomial surface models: most CAD
file of instructions (commonly named G-code systems are using parametrical functions to
instructions) that identify tool movements, represent surfaces. In this form, surface points
tool changes, feed speed, status of coolant, and are described by a parametric function p(u,v)
other parameters. Optionally, generated G-code where p is a point in 3D space and u and v are
may be verified by verification software. It location parameters of a point in a 2D reference
simulates the machine axis movements and plane. When a coordinate system U ¼ (Q, x, y, z)
behavior as if the machine executed an NC- is chosen in 3D space, then the mapping is
program. This technique allows validating uniquely defined by p(u, v) ¼ (x(u, v),
the quality of the machined surface and y(u, v) z(u, v)). This is illustrated in Fig. 2. Lines
detecting undercuts and overcuts that may on the surface with a constant value for u or v are
appear during machining. Finally, the called the isoparametric lines of the surface.
NC-program is executed on a machine resulting CAD systems are using partial polynomial
in the produced part. functions and a set of control points P to define a
The tool path generation step is thus the parametric free-form surface. The form of this
second step in the design-to-part process flow. free-form surface is determined by weighting
The input for this step is a CAD file, while the the control points with these partial polynomial
output of this step in the process flow is the functions. In general, the surface is not going
machine-independent tool path data (called through these control points. Depending on
CLDATA-file or Cutter Location DATA-file). the used polynomial functions, these free-form
F 516 Five-Axis Tool Path Generation

Five-Axis Tool Path V

Generation, d
Fig. 2 Parametric V=C
representation of a surface
(Dejonghe 2001)
U=b
Vp p
p
C V=d
U=a
z
a Up b U

Q
y
x

z
P0,4 z
a b
P4,4

P0,1

P0,0

P1,0
y y

x x

P4,0

Five-Axis Tool Path Generation, Fig. 3 Parametric free-form CAD surface (a) Control points (b) Resulting surface
(Dejonghe 2001)

surfaces are called Bézier, B-spline, or NURBS engineering a workpiece is often converted in an
surfaces. This is illustrated in Fig. 3. STL-model which then is directly used for tool
Faceted models: a faceted model in its most path generation.
general form is a discrete nonparametric
representation, where the surface is represented Tool Shape and Trajectory Definition
by a set of faces. The most used form is the one In order to curb a great variety of all possible tool
where triangles are used for representing the shapes and sizes, the industry focuses on the
surface. One of the most used triangle-based standardized geometric definition of
facet model formats in CAD/CAM technology a generalized cutting tool that is based on the
is probably the STL-format. Each triangle is APT definition. The most comprehensive shape
represented by its three vertices (Fig. 4). The of an APT cutter may consist of the bottom and
STL-format is also very popular for copying side lines and fillets connecting the lines, as
reverse engineered workpieces on an NC-milling shown in Fig. 5. However, in practical applica-
machine. The conversion of a cloud of points into tions, the range of used tool shapes is narrower.
a polynomial surface model is cumbersome and The most used tool shapes, also in five-axis
time consuming. Therefore, the result of reverse machining, are shown in Fig. 6.
Five-Axis Tool Path Generation 517 F
Five-Axis Tool Path
Generation,
Fig. 4 Faceted model of
a surface (Dejonghe 2001)

Tool axis φ

R Cutter radius
r Cutter corner radius
R β Bottom angle
r
Tool tip φ Taper angle r
(CL point) r

β d d d

Five-Axis Tool Path Generation, Fig. 5 General Flat-end Ball-nose Fillet

description of a cutting tool (Plakhotnik 2012)
Five-Axis Tool Path Generation, Fig. 6 APT cutter
shapes (Plakhotnik 2012)
Current industrial practices are to represent
five-axis tool paths, a spatial trajectory of the
tool, as a sequence of intermediate tool positions, path generation algorithms use two possible
also known as tool postures, because it is seldom methods. In a first method, the tool path genera-
possible to derive a closed-form analytical tion algorithm uses a local coordinate system
expression of these trajectories. Each tool posture (LCS) which has an origin in the point where
is described by a coordinate of a cutter location a tool touches a surface. This point is called
(CL) point, a common name of the end point of the cutter contact (CC) point, as shown in
the APT cutter (x, y, z), and a unit vector of the Fig. 7b. The LCS aligns tool geometry to local
tool orientation (i,j,k), as shown in Fig. 7a. surface properties (normal, mean curvature
These coordinates (tool position (x, y, z) and directions). Without LCS, a CL point would
tool orientation (i, j, k)) are represented in the define tool location within a global frame
workpiece coordinate system (WCS). Switching neglecting any information about surface
from one WCS to another causes invariant trans- geometry. The axes of LCS are directed
formations and, therefore, does not influence as follows:
five-axis tool path generation. Most five-axis • FL is a feed direction and denotes X axis of the
tool path generation algorithms do not use LCS.
WCSs (x, y, z, i, j, k) to compute tool postures. • NL is a surface normal and denotes Z axis of
In order to ease computation of tool postures, tool the LCS.
F 518 Five-Axis Tool Path Generation

a Tool orientation b
(i, j, k)

CL point TL
(x, y, z)
NL CL point

Z
Y
FL
Feed
FL x NL

X
CC point

Tool posture Local Coordinate System

Five-Axis Tool Path Generation, Fig. 7 Tool position (Dejonghe 2001)

a b

α NL NL
β

TL TL

Five-Axis Tool Path ϕ

Generation, Fig. 8 Tool
posture angles (a) Angles:
lead α and tilt b (b) Angles: FL FL
inclination φ and screw θ θ
FL × NL FL × NL
(Dejonghe 2001)

• (FL  NL) is a product of a feed direction and tool orientation vector TL onto the plane
a surface normal and denotes Y axis of the defined by the normal NL and feed direction
LCS. FL vectors. Tilt angle b is the angle between
In a second method, five-axis tool path the normal vector NL and the projection of the
generation algorithms usually do not use (i, j, k) tool orientation vector TL onto the
notation for tool orientation. It is replaced by a plane defined by the normal vector NL and
notation that includes two angles which the (FL  NL ) axis.
can define tool orientation within the LCS. • Inclination and screw angles (Fig. 8b).
These angles are easier to use to align tool Inclination angle f is the angle between the
shape to the local surface properties. There are normal vector NL and the tool orientation
two ways to define these angles: vector TL. Screw angle f is the angle between
• Lead and tilt angles (Fig. 8a). Lead angle a, the feed direction FL and the projection of
also called Sturz angle, is the angle between the tool orientation vector TL onto the
the normal vector NL and the projection of the plane defined by the feed direction FL and the
Five-Axis Tool Path Generation 519 F

F
Zig-zag Spiral Contour

Five-Axis Tool Path Generation, Fig. 9 Milling patterns (Plakhotnik 2012)

(FL  NL) axis. In a case when the tool orien- only one class of surfaces, called developable sur-
tation vector TL is within the FL–NL plane, tilt faces (which can be flattened to a plane without
and screw angles become equal to zero, and distortion), can be assigned with 2D patterns. As
lead angle is equal to inclination angle. a result, a big group of methods to generate pat-
In a case when the tool orientation vector TL is terns on a complex surface has been developed.
within the FL–NL plane, tilt and screw angles The most common strategies to generate
become equal to zero, and lead angle is equal patterns are listed below:
to inclination angle. • Drive surface method. This method is based
on the use of a so-called drive surface. In this
Milling Strategies and Tool Path Trajectory method, patterns are generated on relatively
Patterns simple surfaces, like planes or spheres, and,
The key description of a tool path is a tool path then, projected onto the part surface, as shown
pattern. It represents a layout of tool motion in Fig. 10.
across the surface to be machined. Some of • Pattern generation in the surface parametric
the patterns can be seamlessly recognized from domain. This method is based on generation
their shapes. Figure 9 illustrates several patterns of patterns in the (u, v) parametric coordinates
which are often used in milling. Other patterns do of a given surface. A 2D pattern is arranged
not inherit a particular recognizable geometrical within the (u, v) domain of a surface and,
representation, but their shapes can target various eventually, mapped back to the surface, as
objectives like obtaining particular surface shown in Fig. 11. A typical pattern of this
quality after machining, following paths of method is a zigzag path (isoparametric path)
small variation in surface normals along because, usually, a surface patch is mapped
the pattern, or following iso-curvature paths (¼ onto the square (u, v) domain with the edge
paths with equal curvature). Different existing length equal to 1 which is very suitable for
approaches to generate tool paths applied to the generating zigzag paths. Besides the zigzag
five-axis milling are described in this section. pattern, other patters are used as well. For
Historically, tool path patterns came into instance, Fig. 12 illustrates an example in
existence to fulfill the needs of 2D milling. which a non-square (u, v) domain is covered
Afterwards, these 2D patterns were exploited for with space-filling curves.
simultaneous five-axis milling of complex-shaped • Cutting plane method. In this method, tool
surfaces. It was impossible to map straightfor- path curves are defined by intersecting
wardly 2D patterns onto an arbitrary surface; a group of parallel planes (cutting planes)
F 520 Five-Axis Tool Path Generation

with the surface, as shown in Fig. 13. This

method is also called drive surface, iso-planar,
and APT methods (Jerard 1998).
• Machining strip-based methods. Some tool
path generation methods consider surface
quality after machining as an important input
parameter for the generation of tool trajectory
patterns. They incorporate the computation of
scallop height, an accepted measure of surface
quality which is shown in Fig. 14 as a distance
from a designed surface to the cusp of material
left unmachined between two adjacent tool
path tracks. These methods challenge
Five-Axis Tool Path Generation, Fig. 10 Pattern from
generation of tool paths that result in scallop
a guide surface (Plakhotnik 2012) height that is smaller than an allowable value.

mapping
of a track

Five-Axis Tool Path

Generation, Fig. 11 (u,v) Surface model
domain to surface
(Plakhotnik 2012) 0 1 u

a 1 b

−10
0.8 z
−20

0.6
40
v
0.4 20

y 0

0.2 −20

−40 40
0 20
0 −20
0 0.2 0.4 0.6 0.8 1 −40 x
u
Space-filling curves of the (u,v) domain) Space-filling curves on the surface

Five-Axis Tool Path Generation, Fig. 12 Curvilinear space-filling curve (Makhanov 2008)
Five-Axis Tool Path Generation 521 F
machining
strip
width

nominal
scallop surface
height

Five-Axis Tool Path Generation, Fig. 14 Scallop

cusps (Plakhotnik 2012)

F
into isophote regions, regions where the normal
vector varies in a certain range from a given
reference vector. Figure 16 shows several
isophote regions, bound by the iso-inclination
curves, on a surface that can be used as guide-
ways for tool path curves. If a tool is oriented
normally to a machined surface, the isophote tool
paths tend to reduce variation of the tool orienta-
Five-Axis Tool Path Generation, Fig. 13 Cutting tion vector.
planes A second method is the iso-conic partitioning
method (Nan Wang 2008) that resembles the
Figure 15 shows tool paths that have been isophote method, but it generates isolines, called
generated by two different approaches. The equi-space contour lines, on a surface that refer
isoparametric tool path shown in Fig. 15a is to small variation of the machine rotary axes.
generated along isoparametric curves in the Figure 17 shows curves that are equidistant in
parametric domain so that the scallop height regard of machine rotations. Similar, these
between successive cutter passes does not exceed isolines are not tool path curves, but some of
a certain value. The iso-scallop tool path shown them can be used as guide ways for tool paths.
in Fig. 15b is also generated in the parametric
domain, but it does not follow isoparametric Setting Tool Inclination
curves, and tool path distribution ensures In five-axis tool path generation, the tool inclina-
constant scallop height between successive cutter tion can be constant (e.g., constant lead angle) or
passes. The machining strip concept was used in optimized in each point of the surface to be
the research (Lee and Lee 2002) that led to machined. Especially when end mills or toroidal
a method that ensures generation of the widest tools are used, the tool inclination can be
machining strips. optimized for maximal machine strip width. In
• Methods based on isolines of surface topology other words, the tool inclination can be set so the
properties. A number of methods assess curvature of the tool in the tool contact point
different properties of surface geometry. Two perfectly fits with the curvature of the surface in
methods are described below. a direction perpendicular to the feed direction
The first method is the isophote method (Fig. 18) (Dejonghe 2001). This can be expressed
(Zhonglin Han 2001) that segments surfaces in the following equation:

Surface curvature ð? feed directionÞ

Sinðinclination angleÞ ¼
Tool RadiusðRÞ
F 522 Five-Axis Tool Path Generation

Five-Axis Tool Path

Generation, Fig. 15 Tool
paths by different tool
path planning approaches
(Lee 1998)

An iso-inclination curve gouging (because the optimal calculation is only

based on the surface properties in the tool contact
point). This gouging and also other collisions can
often be solved by increasing the inclination angle
or by introducing an additional tilt and/or screw
C1 V
Ci+1 C1 n angle.
qi
V If an additional change of tool angle cannot
qi+1 n fully solve the occurring gouging/collision, tool
u Σ (q1.qi+1) path generation algorithms will lift the tool, mostly
P(u,v)
along the surface normal, away from the work-
v
piece. This will of course result in additional rest
z
y material which has to be removed in a following
o x An iso-phote operation (using other tools and strategies).

Five-Axis Tool Path Generation, Fig. 16 Isophote Applications

method (Zhonglin 2001) Today many industries are gaining advantages of
using five-axis milling machines. These machine
Optimal Gouge and Collision-Free Tool Postures tools are actively used for the production of
Another important aspect within tool path gener- ordinary and precision components in the aero-
ation algorithms is the generation of optimal space, diesel engine, automotive, and heavy indus-
gouge and collision-free tool postures. Gouging tries. A number of other industries, such as
is defined as the intersection of the tool with the medical, textile, oil, and glass industries, are
part surface to be machined (Fig. 19a). Collision exploring five-axis milling technological options
is defined an intersection of other (non-cutting as well. Since about a decade ago, a five-axis mill-
tool parts) with the workpiece (Fig. 19b) or even ing machine has spilled from its niche to the market
other collisions (e.g., between moving machine of machine tools and became common machinery
components) that occur during machining. in the mainstream metal working industry.
Advanced five-axis tool path generation There is no specific application or task that is
algorithms have functionality to calculate gouge assigned for five-axis machining. Components of
and collision-free tool postures. Gouging and different sizes and shapes are produced in five
collision avoidance algorithms are based on the axes, but the most renown “five-axis” parts remain
following procedures: stamping dies for the manufacturing of car body
panels, large airspace structures like wing bodies,
Additional Inclination of the Tool submarine and ship propellers, and blades for tur-
It can happen that the most optimal tool inclina- bines and compressors. Some examples are illus-
tion (as calculated in the previous section) gives trated in Figs. 20–22 (Lefebvre 2007).
Five-Axis Tool Path Generation 523 F
Five-Axis Tool Path
Generation,
Fig. 17 Equi-space
method (Nan Wang 2008)
55
50
45

Z
40
35
30
120
100 −40
−30
80 −20
−10 F
60 0
X 10 Y
40 20
30
20 40
50
0 60

Five-Axis Tool Path

Generation,
Fig. 18 Optimal setting of
the tool inclination to the
curvature of the surface

a b

collision

overcut CL point gouging CC point

Five-Axis Tool Path
Surface error due to overcut Tool bottom and flank interfering surface
Generation,
Fig. 19 Five-axis issues geometry
F 524 Flexibility

References

Dejonghe P (2001) An integrated approach for tool path

planning and generation for multi-axis milling.
KU Leuven, Leuven
Jerard BC (1998) Sculptured surface machining: theory
and applications. Kluwer, Dordrecht
Lee C-J, Lee Y-S (2002) A machining potential field
approach to tool path generation for multi-axis sculptured
surface machining. Comput Aided Des 34(5):357–371
Lee Y-S (1998) Non-isoparametric tool path planning by
machining strip evaluation for 5-axis sculptured
surface machining. Comput Aided Des 30(7):559–570
Lefebvre PP (2007) Development of an operation
planning system for multi axes milling of complex
shapes. KU Leuven, Leuven
Makhanov WA (2008) Curvilinear space-filling curves for
five-axis machining. Comput Aided Des 40(3):350–367
Five-Axis Tool Path Generation, Fig. 20 Turbine
Nan Wang KT (2008) Five-axis tool path generation for
(PowerMEMS project – KU Leuven)
a flat-end tool based on iso-conic partitioning. Comput
Aided Des 40(12):1067–1079
Plakhotnik D (2012) Tool path planning and optimization
for five-axis flat-end milling considering machine
kinematics. KU Leuven, Leuven
Zhonglin Han DC-J (2001) Isophote-based ruled surface
approximation of free-form surfaces and its applica-
tion in nc machining. Int J Prod Res 39(9):1911–1930

Flexibility

▶ Productivity

Five-Axis Tool Path Generation, Fig. 21 5-axis

milling dental application (micromachining. com) Flexible Manufacturing System

Hoda ElMaraghy1 and Alessandra Caggiano2

1
Canada Research Chair in Manufacturing
Systems, Industrial and Manufacturing Systems
Engineering, Intelligent Manufacturing Systems
(IMS) Centre, University of Windsor, ON,
Canada
2
Fraunhofer Joint Laboratory of Excellence on
Advanced Production Technology, Department
of Chemical, Materials and Production
Engineering, University of Naples Federico II,
Naples, Italy

Synonyms

Five-Axis Tool Path Generation, Fig. 22 Car Adaptable production system; Adjustable
prototype modelling (Siemens) manufacturing system; FMS
Flexible Manufacturing System 525 F
Definition 1960s, and the first physical FMS was established
around 1967 in the United States (Groover 2000).
Flexible manufacturing system (FMS) is an inte- Flexible manufacturing systems were intro-
grated group of processing CNC machines and duced in response to a new demand for more
material-handling equipment under computer con- variety and for greater responsiveness to changes
trol for the automatic processing of palletized parts. in products, production technology, and markets.
It is implemented for the mid-volume, mid-variety The objective of an FMS is to carry out on
production and represents a compromise between a single system the fabrication of several types
the high flexibility of versatile job shops and the of parts that can change over time (ElMaraghy
high production rate of a dedicated mass produc- 2007), with small conversion time, at the required
tion system (e.g., transfer lines). It is capable of volume and quality, with the aim of achieving
producing limited number of preplanned part fam- productivity and flexibility at the same time. F
ilies and utilizes similarities between members of An FMS is based on the concept of group tech-
a parts family using group technology. The FMS is nology, as it is designed to produce parts within
most suited for the mid-variety, mid-volume pro- a specific range of styles, sizes, and processes.
duction range. FMS exploits the benefits of the Thus, FMS flexibility is limited to the production
Economy of Scope while achieving the efficiencies of a single part family or a limited range of part
of the Economy of Scale (ElMaraghy 2006). families. A part family is defined as a number of
The reason the FMS is called flexible is that it is parts with similar dimensions, geometric features,
capable of processing a variety of different part and tolerances or processing requirements or both.
variants belonging to the same family simulta- By taking advantage of these similarities between
neously at the various workstations, and the mix parts in design and/or manufacture to achieve
of part styles and quantities of production can be economy of scope, FMS can address changes in
adjusted in response to changing demand patterns. work orders, production schedules, part-programs,
The ability to process many different parts within and tooling (Groover 2007).
FMS with minimum engineering effort and change-
over time requires both physical and logical enab- Flexibility
lers (ElMaraghy 2006 and Wiendahl et al. 2007): Flexible manufacturing systems are based on the
Hard (physical) enablers: concept of flexibility which can be defined as
• Programmable machines and equipment the capacity of a system to adjust itself in
• Sensors for monitoring feedback response to changing requirements without sig-
• Modular design of machines and adaptable nificant expense in terms of time, effort, cost, or
HW equipment performance (Toni and Tonchia 1998).
Soft (logical) enablers: The ability to process many different parts
• Group technology – A prerequisite for FMS within FMS with minimum engineering effort
implementation and changeover time requires both physical and
• Adaptable process planning, setup, fixtures, logical enablers (ElMaraghy 2006).
and tools planning There are different types of flexibility:
• Adaptable scheduling, flexible routing/ • Machine flexibility: Various operations
sequencing performed without set-up change
• Real-time control, decision making, and fault • Production flexibility: Number of all part
recovery types that can be produced without adding
major capital equipment
Theory and Application • Mix flexibility: Ability to change product mix
without affecting the production quantity
Introduction • Product flexibility: Ease (time and cost) of
The concept of flexible manufacturing system introducing new products into an existing
(FMS) was conceived in Britain in the early product mix
F 526 Flexible Manufacturing System

• Routing flexibility: Number of feasible each product so that the system has all the
alternate routes of all part types/number of required information for transforming raw parts
part types into finished parts (Matta and Semeraro 2005).
• Volume flexibility: The ability to vary produc-
tion volume profitably within production Hardware Components
capacity FMS main hardware components include work-
• Expansion flexibility: Ease (effort and cost) of stations, material-handling systems, and a central
augmenting capacity and capability when control computer; moreover, tooling, fixturing,
needed (approaching reconfiguration) and pallets are required. Human labor is used
No ▶ manufacturing system can be primarily in load /unload stations and some
completely flexible, since it cannot produce an assembly tasks. However, operators are used in
infinite range of parts or products. There are FMS much less than in traditional production
limits to the flexibility that can be incorporated systems.
into an FMS: However, to qualify as being
flexible, a manufacturing system should satisfy Workstations Computer-controlled machines
specific criteria. Tolio and Valente (2006) (CNC machine tools, machining centers, assem-
proposed a solution by having only part of the bly machines, robots, etc., load/unload stations,
manufacturing system being flexible and called it and auxiliary stations (painting, washing, heat
“focused flexibility.” treating, inspecting (CMM), etc.). In order to
In literature, the following tests of flexibility deal with a variety of product and process
for a manufacturing system have been identified requirements, system automation must be pro-
(Groover 2007): grammable and easily adaptable and versatile
1. Part variety test. It concerns the system capac- machines should be employed (Chryssolouris
ity to process different part styles in a non-batch 2006). For this reason, ▶ Computer Numerical
mode. Control (CNC) ▶ machine tools and robotic
2. Schedule change test. The system should be workstations represent most of the equipment in
able to readily accept changes in production these systems, as their movements during the
schedule, as well as variations in either part machining operations are locally controlled by a
mix or production quantities. computer. These machine tools are easily repro-
3. Error recovery test. The system’s ability to grammable and they are provided with tool
recover from equipment faults and break- changer and tool storage systems to accommodate
downs quickly and easily to avoid production a variety of tooling.
interruptions.
4. New part test. New part designs should be Load/Unload Stations Load/unload stations are
introduced into the existing product mix employed to clamp/fixture parts on pallets before
without too much effort. entering the system, and remove finished parts
In order to be classified as an FMS, after their process cycle has been completed.
a manufacturing system should stand at least the These stations can employ human operators or
first three tests. robots to accomplish the task.

FMS Architecture and Components Material-Handling Systems Several different

Flexible manufacturing systems are characterized types of computer-controlled material-handling
by an integral architecture where hardware and systems are employed to move parts between
software compose an efficient and reliable unit. workstations automatically. Suitable handling
FMS equipment and system are programma- systems for FMS include conveyors, automated
ble. A process cycle defining all the technological guided vehicles (AGV), tow-carts, stacker
information (e.g., type of operations, tools, feed cranes, and industrial ▶ robots. Shuttles are
movements, working speeds, etc.) is available for mechanisms for transferring work pieces
Flexible Manufacturing System 527 F
between stations and MHS, which are also used a significant role in system management, opera-
as limited in-process buffer. tion, and control. Tasks performed by human
The selection of the type of material-handling workers include:
system is a function of several factors. • Loading and unloading of parts
• First, the material-handling system must be • Supplying materials, tools, and fixtures
able to hold the part, its fixture, and pallet as • Maintaining and repairing equipment
needed. Non-rotational parts are often moved • Programming CNC machines and robots
on pallet fixtures designed for the particular • Managing the overall system
handling system and the fixtures are designed
to accommodate the various part family geom- Computer Control System
etries. Rotational parts are often handled by FMSs incorporate a distributed computer system
robots, if their weight is within the robot that is interfaced to the workstations, material- F
carrying capacity. Large and heavy parts handling systems, and other hardware compo-
require other handling systems such as roller nents. It consists of microcomputers controlling
conveyors, guided vehicles, or track-driven the individual machines and components and
vehicle systems. a central computer that coordinates the activities
• Second, the number of machines in the system for the overall operation of the system. Tasks
and the layout should be taken into consider- performed by an FMS computer control system
ation, as the material-handling system should include (Groover 2000):
be able to reach all the workstations. • Workstation control
• Third, the material-handling system must be • Distribution of control instructions to
capable of moving parts from one machine to workstations
another in a timely manner; otherwise, the • Production control
system will be unproductive. • MHS traffic control
• Work piece monitoring
Buffers These are the places where parts wait • Tools control (tool location, tool life
for the availability of system’s resources. monitoring)
In-process buffers are minimized in FMS by • Performance monitoring and reporting
design and central automated storage and • Diagnostics
retrieval systems are commonly used for storing FMS software consists of several modules asso-
incoming and finished parts. ciated with the various functions performed by the
manufacturing system (part programming,
Tooling The tooling used in an FMS must be machines control, material-handling system con-
capable of supporting a variety of products or parts. trol, tool management). Data and commands are
transmitted from the central computer to the single
Fixtures and Pallets Fixtures are often dedi- machines and other hardware components, and data
cated to specific products, but in recent years, on execution and performance are sent back from
work has been focused on creating flexible the FMS modules to the central computer. In order
fixtures able to support a variety of components. to support data transmission, local area network
As parts within an FMS are frequently moved (LAN) technology is commonly employed in FMS.
from one station to another on fixtures, these
should be designed so that the material-handling Types of FMS
system can easily handle them. Fixtures provide Flexible manufacturing systems can be classified
stability to parts during processing and allow according to the number of machines in the
automatic clamping of parts on pallets. system as illustrated below:
• Single machine cell
Human Operators In FMS, human operators • Flexible manufacturing cell
do not carry out processing, but they play • Flexible manufacturing system
F 528 Flexible Manufacturing System

Flexible Manufacturing
System, Fig. 1 Single and
multiple machine FMS

Another criterion for FMS classification FMS Planning, Design, and Operation
is based on the layout that can be one of the The FMS planning process involves
following as illustrated in Fig. 1: a number of steps. First of all, the analysis of
• In-line layout the parts that will be produced by the system
• Loop layout should be carried out according to the following
• Ladder layout factors:
• Network layout that allows branching for • Application of group technology to form
flexible routing part families, which is a prerequisite for
• Robot-centered cell layout implementing FMS
Flexible Manufacturing System 529 F
• Processing requirements, process plans and utilization, deterministic models can be used.
and precedence consideration, and possible Deterministic models do not allow the evaluation
alternate resources and routing (Azab and of operating characteristics such as queues and
ElMaraghy 2007) other dynamic behaviors that can impair the FMS
• Characteristics and features of the work parts performance.
and machine capability requirements Queuing models, based on the mathematical
• Production volume and foreseen variation in theory of queues, can be used to describe some of
market demands the dynamics not accounted for in deterministic
Once the part family, production volumes, and approaches.
other part issues have been defined, the FMS In the later stages of design, discrete event
design process can be carried out by identifying simulation offers a more accurate method for
the following features: modeling an FMS by taking into considerations F
• Types of workstations and their capabilities aspects such as layout configuration, number of
• Variations in process routings and FMS machines, tools and fixtures, size of buffers, part
layout routing heuristics, and production scheduling
• Material-handling system and its characteristics rules.
• Work-in-process and storage capacity
• Tools, tool magazines, tool storage, and FMS Applications and Benefits
maintenance which is normally central Flexible manufacturing systems typically address
• Fixtures and pallets mid-volume and mid-variety production. If
In order to meet production requirements and a product is required in high quantities with no
achieve operational objectives related to profit, style variations, then a dedicated mass production
quality, and customer satisfaction, the FMS manufacturing system is most appropriate.
existing resources must be optimized by dealing For products characterized by low volume and
with the following operational problems: high variety, then job shops with versatile
• Scheduling and dispatching priorities and machines or even manual methods would be
rules more appropriate (ElMaraghy 2007).
• Machine loading, system balancing, and takt Historically, most of the applications of FMS
time have been in milling and drilling operations (non-
• Part routing including route flexibility/alter- rotational parts), and more recently in turning
nate routes (rotational parts), through the employment of
• Part grouping into one or few preplanned parts CNC machining centers. Other implementations
families of FMS include assembly, inspection, sheet-
• Tool management policies metal processing (punching, shearing, bending,
• Pallets and fixtures allocation and forming), and forging.
FMS reduces or eliminates waste by
FMS Analysis ensuring that non-value added tasks such as
A number of quantitative analysis techniques load/unload, parts clamping and fixturing and
have been presented in literature for FMS set-up, gauging, tool changes and maintenance
analysis (Matta and Semeraro 2005) and Groover occur outside the processing cycle, hence
(2007). FMS analysis techniques include deter- achieving higher productivity. Its automated
ministic and queuing models, discrete events features, monitoring and feedback adaptability
simulation, and other approaches including and fault recovery strategies allow un-manned
heuristics. utilization of second and third shifts which
To obtain early estimates of system perfor- increases the utilization of its expensive
mance metrics such as production rate, capacity, equipment.
F 530 Flexible Manufacturing Systems

Compared to conventional manufacturing sys- perspectives and future challenges. Springer Science,
tems, the benefits of FMS implementation include: Boston, pp 35–44
Groover MP (2000) Automation, production systems, and
• Increased machine utilization (as high as 90 %) computer integrated manufacturing, 2nd edn. Prentice
• Fewer machines required Hall, Englewood Cliffs, NJ
• Reduction in necessary ▶ factory floor space Groover MP (2007) Fundamentals of materials
• Greater responsiveness to changes manufacturing: materials, processes and systems,
3rd edn. Wiley, New York
• Reduced inventory requirements Matta A, Semeraro Q (2005) Design of advanced
• Reduced work-in-process (WIP) manufacturing systems. Springer, The Netherlands
• Lower lead times Tolio T, Valente A (2006) An approach to design the
• Reduced direct labor requirements and higher flexibility degree in flexible manufacturing systems.
In: 16th international conference on flexible automa-
labor ▶ productivity tion and intelligent manufacturing, University of
• Opportunity for un-attended ▶ production Limerick, Ireland, June 2006, pp 26–28
Some drawbacks of FMS include: complex Toni DE, Tonchia S (1998) Manufacturing flexibility:
management of large number of products, signif- a literature review. Int J Prod Res 36(6):1587–1617
Wiendahl H-P, ElMaraghy HA, Nyhuis P, Zaeh M,
icant capital investment, and sophisticated Wiendahl H-H, Duffie N, Kolakowski M (2007)
technology to be managed. Since flexible Changeable manufacturing: classification, design,
manufacturing systems are designed to fulfill the operation. CIRP Ann 56(2):783–809, Keynote Paper
processing requirements for all members of the
parts family, the “Flexibility required for antici-
pated changes is built-in a priori. Hence, FMSs are Flexible Manufacturing Systems
relatively costly and some of their capabilities are
often under-utilized as not all part variants are ▶ Holonic Manufacturing Systems
processed at the same time” (ElMaraghy 2006).

Cross-References Flow Forming

▶ Computer Numerical Control ▶ Shear Forming

▶ Factory
▶ Machine Tool
▶ Manufacturing System
▶ Production
Flow of Materials
▶ Productivity
▶ Robot ▶ Logistics

References
Flow Stress, Flow Curve
Azab A, ElMaraghy HA (2007) Mathematical modeling
for reconfigurable process planning. CIRP Ann Alan Bramley
56(1):467–472 Mechanical Engineering, University of Bath,
Chryssolouris G (2006) Manufacturing systems: theory
and practice. Springer, New York
Bath, UK
ElMaraghy HA (2006) Flexible and reconfigurable
manufacturing systems paradigms. Int J Flexible
Manufact Syst (IJFMS) 17(4):261–276, Special issue Definition
on reconfigurable manufacturing systems
ElMaraghy HA (2007) Reconfigurable process plans for
responsive manufacturing systems. In: Cunha PF, Relationship between true stress and strain for
Maropoulos PG (eds) Digital enterprise technology: a given material undergoing plastic deformation.
Flow Stress, Flow Curve 531 F
Related Terms capacity of the machine tool being used.
This mechanical resistance of the workpiece
Stress–strain curves; work/strain hardening curve material is defined by its flow curve which can
be established independently by testing. Thus, in
modelling of a forming process, the flow curve
Theory and Application of the workpiece material is an essential
prerequisite and for this purpose the use of empir-
Introduction/Context ical stress–strain curves and constitutive relations
In metal-forming technology a major parameter are used.
is the load or force required to perform the The flow curve is best explained by
operation. This is dependent on the geometrical considering a simple uniaxial tensile test on
configuration, friction at the tool–workpiece a typical piece of metallic material, mild steel at F
interface, and the mechanical resistance to plastic room temperature. This is depicted in Fig. 1.
flow of the workpiece material and in turn A typical force–extension curve from such
determines the required force and energy a test is shown in Fig. 2.

Diameter d0
Area A0

F F

10

Gauge length

Flow Stress, Flow Curve, Fig. 1 A tensile test specimen

40

35

30
Tensile force/ kN

25

20

15

10

Flow Stress, Flow Curve,

Fig. 2 Force–extension 0
0 5 10 15 20
relation in simple tension
for a mild steel Extension /mm
F 532 Flow Stress, Flow Curve

Original Specimen

Specimen taken just after point B

Specimen at Point C

Flow Stress, Flow Curve, Fig. 3 Tensile specimen at various stages in a tensile test

The Elastic Region and Yielding This is known as the nominal stress of
engineering stress.
The initial part of the curve, OA in Fig. 1, is the The yield stress (defining the onset of plastic
characteristic linear-elastic behavior of the deformation) ¼ force at yield point/original
material, i.e., the extension that occurs is fully cross-sectional area.
reversible and the relationship between the force The extension is converted to strain, defined
and the extension is linear. At an atomic level the as the extension, Dl, divided by the original
bonds between the atoms of the crystal structure length, lo.
are just flexing. The extension in this region is Thus, strain e ¼ Dl/lo which is referred to as
however very small and can only be measured the engineering strain.
using special extensometers. This linearity ceases The relationship is then known as a flow
at point A and the material starts to behave curve (Fig. 4).
irreversibly, i.e., permanent or plastic deforma-
tion occurs. This phenomenon is known as
yielding. In this region the atoms take up new The Plastic Region
position relative to each other by the mechanism
of dislocation activation. The yield stress defines the onset of plastic flow,
After initial yielding the load continues to rise, and the specimen becomes permanently
reaching a peak at point B, thereafter falling until deformed so that if the load is removed,
the specimen fractures at point C. This is depicted the specimen partially recovers its dimensions
in Fig. 3. elastically but is longer than the original
In metal forming the deformation, by specimen. This permanent extension is accompa-
definition, is plastic and so the yielding behavior nied by a reduction in the lateral dimensions of
of materials is the most important region to the specimen as the specimen is stretched, thus
consider. maintaining its volume at a constant level.
The force and extension parameters need to be An important concept here is that of strain
normalized to enable comparisons between hardening which is best explained by understand-
different materials and specimen sizes. This can ing the behavior of the material at an atomic
be done by converting the force to a stress defined level. The dislocations in the atomic structure
as the force, F, divided by the area, A, over which that enable plastic deformation to occur
it is applied. with atoms sliding over one another become
Thus, stress s ¼ F/A. progressively more difficult as more and more
However, it is convenient to divide the force dislocations form and pile up against each other
by the original area of the specimen, viz., Ao. and against the grain boundaries of the metal.
Flow Stress, Flow Curve 533 F
Flow Stress, Flow Curve, 500
Fig. 4 Flow curve
450 B

Engineering Tensile stress (MN/m2)

400 C

350

300

250
A
200

150

100
F
50

0
0 0.1 0.2 0.3 0.4 0.5
Engineering Strain

This makes it increasingly difficult for plastic conditions, the material can in fact sustain much
deformation to occur; i.e., the metal becomes higher stresses without instability or failure
stronger. This is known as strain or work hard- occurring.
ening. Returning to the tensile test, after yield- Using the original cross-sectional area is not
ing the specimen, will continue to deform realistic when considering the stress in the plas-
plastically in a uniform manner, gradually tic deformation region, it being better defined as
increasing its strength. However, the cross- the force divided by the current cross-sectional
sectional area of the specimen is also reducing, area of the specimen. This is known as the true
thus increasing the stress on the specimen. These stress.
two effects eventually balance each other and Similarly, using the extension of the specimen
the load reaches a maximum at which point the divided by the original length is not realistic as
increase in strength caused by strain hardening a measure of the strain in the plastic region.
can no longer overcome the increased stress due Considering infinitesimal changes, an incre-
to the reduction in area. At this point the speci- ment of strain, using the symbol e, can be
men starts to form a local reduction in area, i.e., expressed as:
a neck, and the force drops off as the neck
extends and the specimen eventually fractures. de ¼ dl=l
The maximum load point is known as the point
of instability and is peculiar to tensile stress Integrating this over, the overall change of
systems. length then gives:
The point where the load reaches a maximum
point B in Fig. 2 defines the tensile strength of the e ¼ lnð11 =1o Þ
material.
Thus, tensile strength ¼ maximum force in where l1 and lo are the final and original lengths
tensile test/original cross-sectional area of respectively of the strained specimen.
specimen. e is referred to as the natural, true, or
This is sometimes described as the ultimate logarithmic strain.
tensile strength, but this term should not be used Such total or finite strains are known as natural
as it implies that it is the maximum stress that the or logarithmic strains the use of which was first
material can withstand. Under different loading suggested by Ludwig (1909).
F 534 FMS

Methods of Obtaining Flow Curves to use constitutive equations that incorporate

these effects into modelling of the processes.
In metal-forming operation, the strains are
often quite large with e > 1.0. The tensile test
is rather limited with strains never reaching this References
level prior to necking and fracture. Neverthe-
less, the tensile test is used quite extensively Altan T, Ngaile G, Shen G (eds) (2005) Cold and hot
forging: fundamentals and applications. ASM
to establish approximate stress–strain data for
International, Materials Park, OH
modelling/estimating metal-forming processes Cook M, Larke EC (1945) Resistance of copper and
(Czichos et al. 2006). Much larger strains can copper alloys to homogeneous deformation in
be obtained using simple compression tests compression. J Inst Met l71(12):371–390
Czichos H, Saito T, Smith LE (eds) (2006) Springer
but a different complication arises, that of
handbook of material measurement methods.
friction between the workpiece and the tools. Springer, Berlin, pp 303–304
Various methods have been proposed to over- Johnson W (1956) Experiments in plane strain extrusion.
come these difficulties (Johnson 1956; Loizou J Mech Phys Solids 4(4):269–282
Johnson W, Mellor PB (1962) Plasticity for mechanical
and Sims 1953; Cook and Larke 1945). In the engineers. Van Nostrand, Princeton
case of materials in sheet form, testing in biaxial Loizou N, Sims RB (1953) Yield stress of pure lead in
tension also enables much larger strains than in compression. J Mech Phys Solids 1(4):234–243
simple tension to be obtained (Mellor 1956). Ludwig P (1909) Elemente der technologischen Mechanik
(Elements of technological mechanics). Springer,
Plane strain compression (Watts and Ford
Berlin
1955) and ring tests (Altan et al. 2005) have Mellor PB (1956) Stretch forming under fluid pressure.
also been used. J Mech Phys Solids 5:41–56
Industrial metal-forming operations are Watts AB, Fard H (1955) On the basic yield stress curve
for a metal. Proc I Mech E 169:1141–1156.
carried out over a range of speeds and workpiece
temperatures, and in establishing relevant
stress–strain data for a particular condition, the
effect of strain rate and temperature needs to be FMS
accommodated in the testing regime. Increasing
the temperature of the material reduces the flow ▶ Flexible Manufacturing System
stress and, especially at high temperatures,
increasing the strain rate increases the flow stress.
There are many sources of flow data in
the literature and modelling software vendors Folding (for 180 Bends in Thin
incorporate such data into their packages. Materials Only)

▶ Bending (Sheets)
Empirical Flow Curves and Constitutive
Relations
Force-Based Feedrate Scheduling
It is common practice when making estimates of
load requirements for forming operation to use ▶ Process Optimization via Feedrate Scheduling
approximate or idealized relationships for the in Milling
flow curve such as rigid perfectly plastic, elastic
perfectly plastic, and rigid linear strain harden-
ing. These are explained more fully in Johnson
and Mellor (1962). Recognizing that in forming Forced Vibration
operations ranges of strain, temperature, and strain
rate are encountered, it is now common practice ▶ Vibration
Form Error 535 F
Theory and Application
Form Deviation
General: What Is a Form Error?
▶ Form Error Irregularities of surfaces or profiles can be
decomposed into form, waviness, and roughness
according to the ratio of the distance between
Form Error irregularities (deviations, waves, peaks, cracks,
etc.) to their depth. A form error is the overall
Albert A. Weckenmann parameter describing the maximum of the local
Chair Quality Management and Manufacturing deviations of the real form of a line or surface
Metrology, University Erlangen-Nuremberg, from the nominal – geometrically ideal – feature.
Erlangen, Germany The parameter has to be evaluated according to F
the specification (see below).
One can discriminate between “local form
Synonyms error” (here would be more precise: “local form
deviation”) and “form error” which is valid holisti-
Deviation of form; Form deviation; Shape devia- cally for the entire feature (Humienny et al. 2001).
tion; Shape error In the figure below the nominal workpiece and
an example of the workpiece with form devia-
tion/form error are demonstrated (Fig. 1).
Definition
Form Errors in the Technical Product
The form error deviation of the real Documentation
(manufactured) feature (i.e., line, axis, surface, The maximum admissible form deviation of
center plane) from the geometrically ideal nominal a feature is called form tolerance. A design engi-
feature measured and evaluated by analogy to the neer wishing to specify tolerances of form for
definition of the form tolerance in ISO 1101. It is geometrical features on workpieces can make
the minimum distance of the quasi-parallel bound- use of six different types of form tolerances:
aries of a zone, in which the entire surface of a real straightness, flatness, roundness, cylindricity,
feature, its line element, its axis, or its center plane profile of any line, and profile of any surface
remains. The boundaries of the zone are formed: (Table 1). The tolerances of form defined in
ISO 1101 (2004) are not mutually exclusive, so
For standard features: that several different form tolerance specifica-
• In 2D case by two geometrically ideal 2D tions are possible for a single toleranced feature
features (lines), which are in a common plane (e.g., the cylinder form tolerance includes
• In 3D case by two – geometrically ideal – straightness tolerance of the axis or generatrix
nominal features and the roundness tolerance of a cross section).
The nominal 3D features or the nominal 2D The form tolerance defined in any particular case
lines must show everywhere the same distance specifies a zone – of everywhere equal thickness
to each other (Euclidian distance evaluated for or width – within which the surface of the real
the entire feature mathematically by Chebyshev (manufactured) feature should remain. Note
criterion). well: Only the zone width is defined by the tol-
erance value but not the orientation or position of
For profiles (freeform lines or sculptured the considered feature.
surfaces): The shape of the toleranced geometrical features
• For “profile any line” or “profile any surface,” is usually defined by features of geometrically ideal
the zone is formed in another special way (see form, such as straight lines, circles, planes, and
“profile any line” and “profile any surface”). cylinders. For all deviations of form, it is important
F 536 Form Error

model of the desired ideal real workpiece with

design engineer workpiece form deviations
(exaggerated)

Form Error, Fig. 1 Example of a workpiece without and with form deviations

Form Error, Table 1 Form tolerances, excerpt from DIN Form Errors in the Manufacturing
EN ISO 1101 (2004) Every manufacturing process produces nonideal
Form tolerance: Tolerance characteristics and symbols features. Deviations of size, form, and position
are caused. The type of the form deviation is
Straightness dependent on the specific manufacturing process,
the applied tools, the type of the feature, and the
material. Figure 2 shows some examples of the
form deviations, which may occur in a turning
Flatness
process. Typical roundness deviations here are
the three-lobed parts and ellipses. Cylinder form
deviations can be, e.g., cone, concave, or convex
Roundness shaped, buckled, and three lobed.

Measurement of Form Errors

The measurement of form errors is to be looked at
Cylindricity
in two aspects:
• Sampling points on the surface and attributing
coordinate values to them
Profile any line • Evaluation algorithm of the points in order to
calculate the form error
We cannot measure the entire surface with an
Profile any surface infinite number of points. The surface is
a continuum and can only be described completely
by an analogue description or by an infinite number
of points. So, in practice we take a more or less
to note that the default evaluation must be random sample of the surface’s entity. In DIN EN
performed by the minimum-zone method as ISO 14660–1 (1999), there is in addition
defined in ISO 1101 (2004) (example for a circle a distinction made between geometrical features
is shown in Fig. 5, second image from left). describing the surface of a workpiece (cylinder
Recent developments and strategic plans for surface, sphere surface etc.) and those which are
future work for dimensioning and tolerancing not real existing surfaces or lines (like symmetry
standards are described by Srinivasan (2012) planes and axis). To make this clear, the definitions
and Nielson (2012). are shown in Fig. 3 for the example of a cylinder.
Form Error 537 F

Form Error, Fig. 2 Cause of form errors in manufacturing, as described in (Noppen and Sigalla 1990)

drawing workpiece representation of workpiece

extraction association

Nominal integral real (GPS) feature Extracted integral Associated integral

feature feature feature

Form Error, Fig. 3 Terms

of geometrical features for
the example of a cylinder, Nominal derived Extracted derived Associated derived
as described in DIN EN feature feature feature
ISO 14660–1 (1999)

The number and distribution of the sampling of probing points for typical form measurements is
points in combination with the type and value of in the range of several thousand points).
the real form deviation will influence the value of The evaluation procedure must follow the
the evaluated form error (Fig. 4). So the number of criteria given in ISO standards (Humienny et al.
points should be high enough according to the 2001). Usually the Chebyshev criterion, also
expected accuracy of the result. (Often the number called minimum line/zone (Figs. 4a, 5: second
F 538 Form Error

Form Error,
Fig. 4 Influence of
evaluation method (a) and
probing strategy (b and c),
as described by
Weckenmann et al. (1995)

Form Error, Fig. 5 Associate element calculation for a circle, as described in DIN EN ISO 12181–1 (2011) and by
Weckenmann (2012)

image from left), is applied. The calculation fol- Cross-References

lows nonlinear optimization algorithms. The
optimization criterion is just the distance of ▶ Error
the inner/outer or lower/upper feature. So neither ▶ Metrology
the orientation nor the location is a fixed param- ▶ Roundness
eter. For the optimization process, they are given ▶ Straightness
free unless they are not restricted by further ▶ Tolerancing
specifications.
In some cases the form error is not related to
Chebyshev approximating (minimum zone) fea- References
ture but due to functional reasons to maximum
inscribed feature or minimum circumscribed fea- DIN EN ISO 1101 (2004) Geometrical product specifica-
tions(GPS) – Geometrical tolerancing – Tolerances
ture (for circle, cylinder, sphere) or even to Gauss-
of form, orientation, location and run-out. Interna-
ian (least square) feature or to tangential feature tional Organization for Standardization (ISO), Beuth,
(plane or 2D straight line), as shown in Fig. 5. Berlin
Formability (Damage) 539 F
DIN EN ISO 12181–1 (2011) Geometrical product spec- Definition
ifications (GPS) – Roundness – Part 1: vocabulary and
parameters of roundness. International Organization
for Standardization, Beuth, Berlin The formability is the capability of a material to
DIN EN ISO 14660–1 (1999) Geometrical product spec- undergo plastic deformation to a given shape with-
ifications (GPS) – Geometrical features - Part 1: Gen- out defects. Formability limits are a hard con-
eral terms and definitions. International Organization straint when sheet metal parts are manufactured,
for Standardization, Beuth, Berlin
Humienny Z, Weckenmann A, Bialas S, Osanna PH, but also in bulk metal forming, formability limits
Tamre M, Blunt L, Jakubiec W (2001) Geometrical can be reached leading to faulty parts.
product specification – course for technical universi- In general the term “formability” usually
ties. Printing House, Warsaw refers to sheet metals.
Nielson HS (2012) Recent developments in ISO-GPS
Standards and strategic plans for future work. In:
Jiang X (ed) Proceedings of the 12th CIRP conference F
on computer aided tolerancing – programme and Theory and Application
abstracts, pp 33–43
Noppen G, Sigalla J (1990) Form- und Lagetoleranzen –
Grundlagen [Form and position tolerancing – Sheet Metal Formability
fundamentals]. Technische Mitteilungen [Engineering At the end of the nineteenth century, due to the
notes] Nr. 8904, Augsburg, MAN Technologie AG development of the sheet forming technology,
(in German) sheet metal formability became a research
Srinivasan V (2012) Reflections on the role of science in
the evolution of dimensioning and tolerancing stan- topic. Starting from that time, different tests
dards. In: Jiang X (ed) Proceedings of the 12th CIRP have been proposed. All of them are based on
conference on computer aided tolerancing – the measure of the maximum deformation that
programme and abstracts, pp 18–32 a sheet metal sample can reach.
Weckenmann A (2012) Koordinatenmesstechnik: Flexi-
ble Strategien f€
ur funktions- und fertigungsgerechtes Sheet metal-forming processes can be charac-
Pr€ufen [Coordinate metrology: flexible strategies for terized by two basic types of deformation
function and manufacting related testing], 2nd edn. patterns: drawing and bending. Concerning draw-
Hanser, M€ unchen, p 273 (in German) ing one has to distinguish stretching (all the in-
Weckenmann A, Eitzert H, Garmer M, Weber H (1995)
Functionality-oriented evaluation and sampling plane strain components are greater than zero)
strategy in coordinate metrology. Precision Eng and deep drawing (one of the in-plane strain
17(4):244–252 components is less than zero).
Figure 1 shows that formability is a complex
characteristic of the materials since different
aspects affect its amount. Among them, material
Formability (Damage) properties, process parameters, and strain
bounding criteria are probably the most relevant.
Luigino Filice1 and Dorel Banabic2 As an example, the influence of various
1
Department of Mechanical Energy and parameters on the formability in deep drawing
Management Engineering, The University of is presented in Fig. 2.
Calabria, Rende, Italy The formability, expressed by the drawing
2
CERTETA-Research Center in Sheet Metal ratio ß ¼ D/d (D being the blank diameter and
Forming, Technical University of Cluj, Romania, d the cup diameter), depends on the strain limit-
Cluj Napoca, Cluj, Romania ing criterion as well as on a process parameter
(blank holding pressure p).
The forming region represents the “window”
Synonyms in which material formability is allowed.
A typical defect of formability in bulk metal
Forming limit; Workability forming is shown in Fig. 3.
F 540 Formability (Damage)

Formability (Damage),
Fig. 1 Parameters Mechanical εu, E, R p 0.2, n, r, m
proprieties
influencing sheet metal
formability (Banabic 2000)
Material Metallurgical Texture, size, shape,
Proprieties proprieties density of the voids

Chemical Chemical
proprieties composition

Stress
state

Strain
rate

Process
parameters Temperature

Part
configuration

Formability Lubrication

Tearing

Strain
localization

Strain bounding Wrinkling

criteria

Roughness

Springback

Among the methodologies to evaluate form- circles. The research in this field was pioneered
ability, forming limit diagrams plays a relevant by Keeler (1961).
role. Later, Goodwin (1968) plotted the curve for
the tension/compression domain (e1 > 0; e2 < 0)
Forming Limit Diagram by using different mechanical tests. In this case,
The maximum values of the principal in-plane transverse compression allows for obtaining high
strains e1 and e2 can be determined by measuring values of tensile strains like in rolling or drawing.
the strains at failure (necking, fracture, wrinkling, The diagrams of Keeler (right side) and
etc.) on sheet components covered with grids of Goodwin (left side) are currently called the
Formability (Damage) 541 F
Formability (Damage),
Fig. 2 Influence of various
W
parameters on formability OR
in deep drawing (Marciniak KP
IE
1978) CE
G ST
LIN RE

DRAWING RATIO
K NG
NI
RI TH
W EX
CE
ED
ING

FORMING REGION

GA
LLI
F

NG
BLANKHOLDER PRESSURE

a b

Shear
Deodzone
Flow
Suction

Formability (Damage),
Fig. 3 Formability defect
in radial extrusion
(Arentoft et al. 1998)

forming limit diagram (FLD); see Fig. 4. FLD’s are determined, for necking and for frac-
Connecting all of the points corresponding to ture; see Fig. 5.
limit strains leads to a forming limit curve
(FLC). The FLC splits the “fail” (i.e., above the Experimental Determination of the FLD
FLC) and “safe” (i.e., below the FLC) regions. FLDs must cover as much as possible the strain
FLC is plotted on a forming limit diagram domain which occurs in industrial sheet metal-
(FLD). The intersection of the limit curve with forming processes. The diagrams are established
the vertical axis, which represents the plane strain by experiments that provide pairs of values of the
deformation (e2 ¼ 0), is an important point of the limit strains and obtained for various loading
FLD and is noted FLD0. The position of this point patterns (equi-biaxial, biaxial, uniaxial, etc.). In
depends mainly on the strain hardening coeffi- order to determine an FLD, one must generate
cient and also on thickness. load paths ranging from equi-biaxial tension
Today, depending on the kind of limit (e1 ¼ e2) to pure shear (e1 ¼
e2). In practice
strains that is measured, different types of the state of simple tension (e1 ¼
e2 for isotropic
F 542 Formability (Damage)

Formability (Damage), 120

Fig. 4 Forming limit
CRITICAL STRAIN
diagrams defined by Keeler
GOODWIN
and Goodwin (Goodwin 100 FAIL

MAJOR ENGINEERING STRAIN , %

1968)

FAIL
80
CRITICAL STRAIN
KEELER

60

40 SAFE SAFE

20 TENSION– TENSION–
COMPRESSION TENSION

0
−40 −20 0 20 40
MINOR ENGINEERING STRAIN , %

materials) is never exceeded in the blank holder

Major region.
Strain
+ ∈1 It is necessary to deform the specimen along
a linear strain path, i.e., the trajectory followed by
a point in the e1, e2 plane until reaching the
Fracture
forming limit must be a straight line. Although
several tests have been developed in the past,
Nakazima’s one is commonly utilized in practice.
Forming Limit

Nakazima Test The test (Nakazima and Kikuma

1967; Nakazima et al. 1971) consists of drawing
rectangular specimens having different widths
using a hemispherical punch and a circular die.
2
1 4 By varying the width of the specimen and the
3 lubricant, one may obtain both the positive and
the negative domain of the FLD. The shape of the
specimen recommended by ISO 12004 standard is
presented in the Fig. 6. The recommended length
− ∈2 Minor Strain, + ∈2 is longer than 25 % of punch diameter. The punch
1 : PURE SHEAR diameter and the die diameter are recommended
2 : SIMPLE TENSION the values 100 and 105 mm, respectively.
Advantages of the test are the simplicity of the
3 : PLANE STRAIN TENSION
tools, the simple shape of the specimens, and
4 : BIAXIAL TENSION
the possibility of covering the entire domain of
Formability (Damage), Fig. 5 Forming limit diagrams the FLDs. Disadvantages are the possibility of
for necking and for fracture wrinkling and errors of measurement caused by
Formability (Damage) 543 F
the curvature of the punch. This method is used Factors Influencing the FLC
actually as standard method by the ISO 12004 FLDs are influenced by several factors; thus their
standard “Metallic materials. Determination of generalization depends on the choosing of some
the forming limit curves” (ISO 12004 2006). parameters. For this reason a great attention has
to be paid when FLDs are used as design tools.
a length However, in the numerical codes, many of the
affecting factors are taken into account. For the
sake of completeness, the most affecting ones are
remainder here cited: sheet thickness, grid size, strain path,
of mechanical properties, punch curvature, temper-
blank width
ature, strain rate, and normal pressure.
The complex strain paths undergone by indus- F
R 20–30 mm trial parts during multi-pass forming can be
simulated in laboratory tests by applying to the
b same specimen successively two different load
paths, for example, a tensile test followed by deep
drawing or biaxial expansion followed by a local
tensile test. The analysis of the specimens pro-
vides results similar to those for real industrial
components:
• A path change has a strong influence on FLDs
both for necking and fracture.
Formability (Damage), Fig. 6 (a) Shape of the speci-
• As long as the initial deformation implies
mens used in the Nakazima test (Nakazima and Kikuma
1967); (b) a set of specimens for a complete FLD (Cour- e2 < 0 (compression test, simple tensile test)
tesy by GOM company) followed by biaxial expansion, the limit curve

e1
0,7

0,6

II
0,5
I
0,4

0,3

III
0,2

0,1
Formability (Damage),
Fig. 7 Influence of the
strain paths on the FLC
(Nakazima et al. 1971) −0,3 −0,2 −0,1 0 0,1 0,2 0,3 0,4 e2
F 544 Formability (Damage)

Tension-compression Tension-tension
ε1 (deep drawing) (stretch forming)
320
2 4 6 8 10 Severity

280
B⬙ B

240
C⬘
Failure area
A
B⬘ A⬙ 200

Major strain, %
Safe area
A⬘
160
C
Critical
120 strain
level
−ε2 O +ε2 80
Severity
Formability (Damage), Fig. 8 Examples of the shifting 10
of critical points 40 8
6
Safe area 4
0
−120 −80 −40 0 40 80 120
II is higher than the FLD I (corresponding to
Minor strain, %
simple, linear load paths, see Fig. 7).
• On the contrary, if e2 > 0 during the first stage Formability (Damage), Fig. 9 Definition of the severity
and compression or simple tensile load fol- indices (Chatfield and Keeler 1971)
lows later, the limit curve III is below the
FLD I for simple paths.
The most important influence is due to the maximum strains in the part must be known as
modification of the strain path. More details well as the forming limit diagram of the material.
about the factors influencing the FLC are By comparing the points corresponding to the
presented in the reference (Banabic 2000, 2010; maximum strains in the parts with the FLD, one
Banabic et al. 2010). can estimate whether fracture or necking could
appear during forming.
Use of Forming Limit Diagrams in Industrial If no defect is to be expected, one has infor-
Practice mation about how far from the limit the material
The most realistic and efficient way to estimate is deformed.
the technological possibilities of producing a part If the points defined by the maximum strain
and to avoid unsuccessful design is to apply the are beyond the limit curve, some modifications
FLD method. This enables to determine: have to be made of:
• The range of safety for deep drawing • Working conditions (blank holding,
• The critical zones where necking or fracture lubrication)
are most likely to occur • Design of the part itself (fillet radii, angles,
• The strain level etc.)
• The favorable working conditions (blank • Material (quality, thickness)
holding, lubrication, etc.) The effect of these modifications is
In practice this method is applied as follows. illustrated by Fig. 8. In the first two cases, the
After the shape, dimensions, and material quality characteristic point is shifted either from A to
of a given part have been prescribed by the A0 or A00 or from B to B0 or B00 . In the third case,
designer, the forming technology and the tools the forming limit curve itself is shifted to the
have to be designed. For this purpose the dashed line.
Formability (Damage) 545 F
Formability (Damage),
Fig. 10 Structure of an Major
Strain strain
expert system for the rate Fail

True stress
analysis of sheet metal Neck
formability (Lee et al. Safe
1988)

True strain Minor strain

Constitutive equations Forming limit diagrams
(material behavior) (failure criteria)

Geometry,
friction,
die speed, F
and so
forth

Analytical model
(strain prediction)

Analysis File CAD

program conversion program

(Boundary conditions)

File
conversion

(Geometry)

Material Material Geometry Blueprint

data file loading file

Terminal

The FLD method also gives an estimation of is small and the safety margin is so large that
the severity of deformation through the so-called material is wasted (point C in Fig. 8).
severity index (Chatfield and Keeler 1971). This Therefore it is possible either to modify the
parameter is defined as shown in Fig. 9. If the forming process by increasing the strains (move
point of the maximum strain is below and far from C to C0 ) or to use a material having a lower
from the forming limit curve, the severity index forming limit curve (yet still beyond the point of
F 546 Formability (Damage)

maximum strains). A good compromise between has been achieved in numerical solution methods,
reasonable safety margins and prevention of especially the finite element method.
material waste appears to be in the zone where The structure of an expert system for the anal-
the severity index is 7–8. The safety margin ysis of sheet metal formability is illustrated by
depends on the quality of the material as well as Fig. 10 (Lee et al. 1988). Besides software pack-
on the process parameters. The choice of the ages of finite element analysis and CAD, com-
material not only influences the FLD but also puter-aided process design also requires
the strain path and the deformation front. databases with FLDs for various working condi-
The determination of the maximum strains to tions. Experimental testing is a very valuable
compare with the FLD can be difficult for complex input for these databases, but it is also rather
parts, since the deformations cannot be calculated expensive.
by analytical methods. In such cases two other
methods are available for the estimation of the Damage
maximum strains: experimental studies on small Material damage occurs when the plastic energy
scale models or numerical simulation. In the past absorbed by the material leads to crack.
experimental methods have been most widely Usually we refer to ductile fracture,
used. However, in the last two decades, numerical underlining the role of plastic deformation
simulation has become more and more the method which precedes failure of the part. The fracture
of choice. On the other hand, important progress occurs on the surfaces, for instance, in tensile test
or in the inner (central bursting in extrusion).
Shear stress could play a relevant role as it occurs
in tension or compression tests in which the fail-
ure follows the surfaces characterized by the
larger shear stress (Fig. 11).
While in sheet metal forming FLDs are com-
monly used to evaluate formability, in bulk metal
forming some criteria have been developed.
However, they can be classified in two cate-
gories (Teixeira et al. 2009):
1. Criteria based on the continuous damage
mechanics
Formability (Damage), Fig. 11 Cup and cone shape in 2. Criteria based on the analysis of strain-stress
broken parts in a tensile test (along the shear bands) tensor

A=0
F B=3
A C=12
A D=15
D E=24
E F=30
C

Formability (Damage), Fig. 12 Application of a shear plastic energy-based criterion to a backward extrusion
forging process
Forming Presses (Hydraulic, Mechanical, Servo) 547 F
To the first category belong criteria based on Marciniak Z (1978) Sheet metal forming limits. In:
material porosity in which some microvoid Koistinen DP, Wang NM (eds) Mechanics of sheet
metal forming. Plenum Press, New York/London,
development functions are developed. This pp 215–235
approach is probably more coherent with reality Nakazima K, Kikuma T (1967) Forming limits under
but its complexity usually vanished the effective- biaxial stretching of sheet metals. Testu-to Hagane
ness when applied to industrial cases. 53:455–458 (in Japanese)
Nakazima K, Kikuma T, Hasuka K (1971) Study on the
On the other hand, criteria based on stress, formability of steel sheets. Yawata Tech Rep
strain, or plastic energy model the material as 284:678–680
homogeneous and perfect but calculate step by Teixeira P, Santos AD, César de Sá JMA, Andrade Pires
step a damage function able to individuate the FM, Barata da Rocha A (2009) Sheet metal formability
evaluation using continuous damage mechanics. Int
domain in which material breaks (Fig. 12). J Mater Form 2:463–466
F

Cross-References
Forming Limit
▶ Anisotropy
▶ Bending (Sheets) ▶ Formability (Damage)

References Forming Presses (Hydraulic,

Arentoft M, Bjerregaard H, Andersen CB, Wanheim T
Mechanical, Servo)
(1998) The influence of the constitutive behaviour of
materials on the formability in radial extrusion of Konrad Wegener
tubular component. J Mater Process Technol Institute of Machine Tools and Manufacturing
75:122–126
(IWF), Swiss Federal Institute of Technology
Banabic D (2000) Forming limits of sheet metal. In:
Banabic D, Bunge HJ, Pöhlandt K, Tekkaya AE (eds) (ETH), Z€urich, Switzerland
Formability of metallic materials: plastic anisotropy,
formability testing, forming limits. Springer,
Heidelberg/Berlin
Banabic D (2010) Sheet metal forming processes: consti-
Synonyms
tutive modelling and numerical simulation. Springer,
Heidelberg/Berlin Press machines
Banabic D, Barlat F, Cazacu O, Kuwabara T (2010)
Advances in anisotropy and formability. Int J Mater
Form 3:165–189
Chatfield DA, Keeler SP (1971) Technology for using Definition
sheet steel designing for formability. Metal Prog
99:60–63 Forming presses are forming machines with lin-
Goodwin GM (1968) Application of strain analysis to
ear relative movements of the two main parts of
sheet metal forming problems in the press shop. Soc
Automot Eng 680093:380–387 the tools. Presses have a drive system, which
ISO 12004 (2006) Metallic materials. Sheet and strip. generates a linear movement of at least and in
Determination of the forming limit curves. Part 2: most cases one ram carrying the first main part of
Determination of forming-limit curves in the
laboratory. ISO, Geneva
the tool, which is bound by the linear guiding
Keeler SP (1961) Plastic instability and fracture in sheet system. To counterbalance the press forces,
stretched over rigid punches. PhD Thesis. Massachu- presses have a stout frame, supporting the drive
setts Institute of Technology, Boston system for the ram(s), supporting the linear
Lee D, Majlessi SA, Vogel JH (1988) Process modelling
and simulation for sheet forming. In: Semiatin SL (ed)
guides of the ram, and supporting some kind of
Metals handbook, vol 14, 9th edn, Forming and table for the fixture of the second main part of the
forging. ASM, Metals Park, pp 911–927 tool. The frame also supports all kinds of
F 548 Forming Presses (Hydraulic, Mechanical, Servo)

auxiliary systems as transfer, ejectors, die cush- into the workpiece during the travel of the tool
ions, material infeed, lubrication and coolant, and under acting process forces, which distinguishes
systems for the exchange of tools and tool parts. them, e.g., from die casting machines.
Presses are subclassified in path-driven, force-
driven, and energy-driven machines, depending
on the drive system. Theory and Applications
In path-driven machines the drive system
defines the path of the ram, while the forces are Forming presses deform material. Figure 1 shows
defined by the feedback from the forming process. a forming machine with material region between
Typical realization of such machines is the an upper and a lower die, which are the two
mechanically driven presses. Force-driven main parts of the tool. The deformation, whatever
machines define by their drive system the force material it is, is governed by three equations at
at every position of the ram, while the path is least.
defined by the feedback from the forming process.
Typical realizations of such machines are hydrau- div s ¼ 0 (1)
lic presses, where the pressure and the cross sec-
tion of the piston define the press force. Both s ¼ f ðe; e_ Þ (2)
machines suffer from energetic limitations of
their drive system, which influences accelerations e¼ s gradðuÞ (3)
and thus speed. If only the energy for the process is
defined by the machine, the machine is classified Equation 1 is the continuum equilibrium con-
as energy-driven press machine. Typical realiza- dition, Eq. 2 are the constitutive equations, which
tions are hammers and flywheel spindle presses. specify the material behavior, and Eq. 3 are the
A servo press is a press machine with a closed- kinematics, the displacement-deformation condi-
loop control of the position and/or force. Those tions. The forming machine defines the boundary
machines can be operated in a path-driven, force- conditions for this set of partial differential equa-
driven, or energy-driven mode and thus mark tions. It can be seen without going into details
a fourth class of press machines. But those that those equations are essentially grade two in
machines can only be operated in one of those the displacements, which means that boundary
modes, which is decided by the control system. conditions are in the displacements or in the first
A path-driven mode in servo presses requires derivative of displacements. First derivatives of
a position feedback loop, a force-driven mode, displacements are in all constitutive equations
and a current feedback loop, while the energy- connected to stress at least due to elasticity,
driven mode does not require any feedback loop. which is common for all materials. From these
All kinds of presses are further subdivided in two types of press machines can be derived,
metal-forming, bulk-forming or sheet metal- namely, path-driven machines, where boundary
forming, and polymer-forming presses by the conditions for the displacements are prescribed,
kind of raw material to be processed and in and force-driven machines, where stress bound-
cold, hot, and semi-hot forming machines, ary conditions are prescribed, which are inte-
depending on the temperature at which the mate- grated to the press force. From this it follows
rial is fed in. Those criteria define the design also that despite the possibilities of servo presses
specification of the whole press. to operate under different modes of the drive,
Forming presses are press machines that are path, force, or energy, nothing really new is
designated and utilized to deform some kind of added, because the drive can only introduce
material, mostly metals, which means that the either boundary condition at a time.
main operations within the machine are processes The press force is transmitted on the slide in
from the second main group of processes of DIN pressure points. Thus single- to four-point presses
8580. Forming presses therefore introduce energy can be distinguished.
Forming Presses (Hydraulic, Mechanical, Servo) 549 F
Forming Presses
(Hydraulic, Mechanical,
Servo), Fig. 1 Path-
driven and force-driven
presses: determined by the
boundary conditions for the
governing differential
equations

The accuracy of press machines is limited to the translation of the slide. With clutch and
mainly due to play in the drive system and com- brake the drive system is coupled to the
pliance of drives and frame, giving rise to dis- flywheel or stopped by coupling to the frame.
placements of the slide under load in vertical and Mechanical presses mainly differ in the
horizontal direction and by tilting. O-type frames link drive. Simple link drives are excentric
suffer from horizontal displacements and shaft – connection rod, which gives mainly
tilting of the slide only with excentric loads a cosine of the crank angle as slide path.
from the process, while C-type frames have Knuckle joints and modified knuckle joint drives
both also with centric loads. Even today measur- as shown in Fig. 2 or rocker arm drives as six-
ing of dynamic behavior of press machines under component link drives or eight-component link
dynamic loads is not standardized, but static drives as shown in Fig. 3, which is necessary to
compliances are standard quality parameters of distort the slide path – crank angle correlation
press machines. especially for deep drawing operations. Mechan-
Centric loads are thus the requirement to be ical presses are able to deliver the force necessary
fulfilled for good and accurate parts, which need for forming only within the nominal force
to be optimized in the planning of the forming angle or nominal force stroke as shown in
steps, when multiple forming steps are Fig. 4. On the other hand, as the force is defined
included within one tool on one slide. The prob- by the forming process, the forces may become
lem to be solved is that the force-time functions infinite theoretically in the lower dead center,
for each forming step are different and at best at which requires either an overload protection or
all times within the stroke, the force center special attention by the user.
should be aligned with the center of the slide. Mechanical presses are very energy efficient,
Hydraulic presses sometimes have closed-loop as they are able to run continuously, which is with
controls for keeping the slide despite excentric the exception of friction and the energy for the
loads parallel. forming process only shifting around of kinetic
and potential energy. Furthermore, they act as
power concentrators: while the motor charges
Mechanical Presses the flywheel continuously over a whole crank
angle cycle, only within a small time span, for
The drive system consists of electric motor, example, 30 crank angle, the power for forming
which drives a fly wheel, a gear for speed is withdrawn from the flywheel. A mechanical
reduction, and a link drive to convert rotations press needs approx. 1/12 of the forming power
F 550 Forming Presses (Hydraulic, Mechanical, Servo)

MK, ωK
b c
MK, ωK
α MK, ωK

α
h
H=2r
h

Forming Presses (Hydraulic, Mechanical, Servo), Fig. 2 Link drive mechanisms from Lange (1984) and Schuler
(1996): (a) excenter drive system, (b) knuckle joint drive, and (c) modified knuckle joint drive

Forming Presses (Hydraulic, Mechanical, Servo), Fig. 3 Link drive systems for mechanical presses from
Schuler (1996): (a) excenter drive, (b) rocker arm drive, and (c) 8-component link drive

as motor power, while the motor can operate all energy withdrawal from the flywheel, which
the time in a fairly energy-efficient mode. reduces the speed of flywheel and motor.
Essential layout parameter is the work capac-
ity of the machine, which normally is specified as
nominal stroke times nominal force. But this is Hydraulic Presses
available only in a specified range of strokes per
minute, as it is limited by the power of the motor Hydraulic presses differ from mechanical presses
to high speeds and to low speeds by the permitted by substituting the drive system by cylinder and
Forming Presses (Hydraulic, Mechanical, Servo) 551 F

Forming Presses (Hydraulic, Mechanical, Servo), upstroke pressurizing the ring cylinder underneath the
Fig. 4 Cycle of a hydraulic press cylinder from (Schuler piston, squeezing oil from the upper cylinder room back
1996): (a) fast downstroke by gravity of the slide, (b) into tank
forming stroke, pressurized by the pump, and (c) fast

piston, where the piston is fixed to the slide and the stroke, the hydraulic press can deliver the full
the cylinder to the frame as shown in Fig. 4. A fast press force. No overload can occur. This flexibil-
downstroke is made by opening the prefill valves, ity is counterbalanced by the disadvantage that
which connect the cylinder to the oil tank situated the whole power for forming must be delivered
above the cylinder on the frame. The oil then is by the pump, which means that maximum power
sucked out of the tank by the dead weight of the of a hydraulic press is by a factor of approx. 10
slide, while the oil in the ring cylinder space higher than for mechanical presses. A possible
underneath the piston is pushed into the tank or solution to overcome this problem is the applica-
in the upper cylinder chamber. For forming the tion of nitrogen accumulators, which have the
pump pressurizes the cylinder room giving rise to same task as the flywheel for mechanical presses,
press force F: namely, to distribute the power over the whole
cycle. The pump then fills the accumulators when
F¼pA (4) no pressurized oil is needed for the drive and is
unloaded during the forming cycle.
where p is the pressure, usually not above 300 For hydraulic presses the stroke of the slide
bar, and A is the piston surface. The fast upstroke consisting of acceleration, fast prestroke, deceler-
is realized by a very stout piston rod, leaving only ation, forming stroke, change of direction in the
a very thin ring cylinder between the rod and the lower return point, fast upstroke, deceleration, and
cylinder wall, which only needs very little upper rest needs to be programmed by the user
amount of oil to lift the slide. with graphical interface. Hydraulic presses are
This system enables a very fast and efficient normally by a factor of at least 1.5 slower than
downstroke and prefilling. At every position of a comparable mechanical press, for instance, ten
F 552 Forming Presses (Hydraulic, Mechanical, Servo)

main drive

additional
brake 31 planetary gear
additional 1 32 with crank
drive
2

4 5
output

standard machine

machine with additional drive

forward motion braking
0.5
0.4
0.3
ram velocity

Un
0.2
0.1
[m/s]

0 OT ram velocity [m/s] UT

−0.1
−0.2
−0.3
−0.4
−0.5
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
return motion acceleration
stroke [m]

Forming Presses (Hydraulic, Mechanical, Servo), top-right flywheel with planetary gear to couple the
Fig. 5 Flexibilization of the main drive of mechanical servo motor, and speed stroke diagram showing the supe-
presses by a planetary gear as coupler for the variable rior speed of the flexible drive system
speed drive. Top-left scheme of the drive system,

versus 18 strokes per minute for large area auto- additional servo drives coupled in the link drive
motive presses. While the ram speed of mechani- system or into the rotary gear by planetary gear as
cal presses changes during the forming stroke, shown in Fig. 5 are discussed. Those machines
hydraulic presses are able to keep the speed fairly have only the capability of limited flexibilization
good constant. Hydraulic press drive systems are depending on the ratio of servo drive power to
more compliant than mechanical press drives for power of the main drive. Full flexibilization of the
comparable press forces, which gives rise to larger main drive system is achieved, when the flywheel
slide displacements under loading conditions. is discarded and the electric motor becomes
a servo motor. This then necessitates motors
delivering the power required for the forming
Servo Presses stroke. Depending on the power and torque
requirements, one or several servo motors are
The limitations of mechanical presses concerning required. Especially machine designs with one
the path of the slide have driven machine devel- drive for each pressure point of the slide enable
opment not only towards multicomponent link to keep the slide parallel under excentric loads, if
drives as seen in Figs. 2 and 3 but also towards direct measuring systems are applied. Three clas-
programmable modifications of the slide paths. In ses of servo presses exist today:
Neugebauer et al. (2007), some technical solu- 1. Servo press with link drive system: The ser-
tions for presses with flywheel drive and vomotors replace the flywheel. One or more
Forming Presses (Hydraulic, Mechanical, Servo) 553 F

Forming Presses (Hydraulic, Mechanical, Servo), closed-loop control for each drive, and c servo press with
Fig. 6 Three types of servo presses: (a) link drive servo two linear motors to compensate the horizontal magnetic
press with four motors for two pressure points, (b) servo forces
press with two ball screws for two pressure points and

servo motors operate in parallel and drive the do not reverse, or with reversing motion, which the
gear system as shown in Fig. 6a. latter enables to adjust also the lower dead center.
2. Servo press with ball screw drive system: This Benefits for the forming process directly are,
press system is shown in Fig. 6b and resembles for instance, the possibilities of superposition of
more the hydraulic drive system, if the hydrau- vibrations and counter shearing, movements with
lic cylinders are replaced by ball screws with rest under load for processes which require time
a motor for each pressure point, alternatively and accurate lower return point.
by one motor and a gear system to move the
thread spindles synchronously.
3. Servo press with linear motor drive system:
References
This press system is shown in Fig. 6c and Lange K (ed) (1984) Umformtechnik–Handbuch f€ ur
could only be applied to presses with small Industrie und Wissenschaft, 2nd edn. Springer, Berlin.
forces up to 100 kN, which can be used for [Metal Forming–Handbook for Research and
micro-forming operations. As they omit all Industry]. (in German)
Neugebauer R, Denkena B, Wegener K (2007)
kinds of mechanical gears, the predefined
Mechatronic systems for machine tools. CIRP Ann
path can be followed very exactly to within Manuf Technol 56(2):657–686
microns (see Wegener and Schepp 2000). Osakada K, Mori K, Altan T, Groche P (2011) Mechanical
servo press technology for metal forming. Ann CIRP
60(2):651–667
Those servo presses and their capabilities Schuler GH (1996) Handbuch der Umformtechnik.
and opportunities are described within Springer, Berlin. [Metalforming Handbook].
Osakada et al. (2011). The flexibility of servo (in German)
presses with link drive systems is today mainly Wegener K, Schepp F (2000) Innovative Antriebskonzepte
f€
ur Maschinen der Zukunft [Innovative Drive Concepts
exploited for increasing the productivity. Here the for Future Machines]. In: Groche P, Schmoeckel
variability of the speed helps to better adjust the D (eds) Anforderungen, Innovationen, Perspektiven
cycle to forming and transfer of parts from one [Demands, Innovations, Perspectives]: 7.
step to the other, so that with the same forming Umformtechnisches Kolloquium Darmstadt [7th Metal
Forming Conference, Darmstadt], TUD, Inst. f€ ur
speed and the same transfer speed faster idle
Produktiostechnik und Umformmaschinen [Institute
movements are possible. Link drive systems can for Production Engineering and Forming Machines].
be driven in continuous motion, where the gears Meisenbach, Bamberg, pp 117–132 (in German)
F 554 Forming Tools (Die, Punch, Blank Holder)

In forming technology, the workpiece geome-

Forming Tools (Die, Punch, Blank try is entirely or partially determined by the tool
Holder) geometry. Thereby, in contrast to machining
techniques, much larger tools and therefore
Mathias Liewald higher forces are required for forming. Due to
Institute for Metal Forming Technology, complex part shapes, the workpieces are often
University of Stuttgart, Stuttgart, Germany not produced in one step. Depending on part
geometry, production is carried out in several
processing steps using one or more production
Synonyms procedures, such as forming and cutting, heat,
or surface treatment, whereby one step can
Metal forming die; Metal forming tool; Tool include several processes.
In the construction phase of the tools, the
required manufacturing processes as well as the
Definition order and number of production steps are deter-
mined in the method plan or sequence of opera-
Device for giving shape to solid materials by tions plan. The available machines, the planned
mechanical means. number of pieces, and any other operational
boundary conditions are thereby taken into
consideration.
Theory and Application The goal of tool development is to minimize
the number of required steps while at the same
Forming Tool: General Considerations time ensuring a high level of operational reli-
All forming methods can be regarded as a system ability. The parts are already greatly simplified
reaching from a plasticity approach and metallur- in the draft stage in close cooperation between
gical principles right up to questions of produc- the parts design and production departments, in
tion (Fig. 1). The forming procedure, which is the order to be able to combine several forming or
central aspect, must be viewed in connection with cutting operations into one production station.
the forming tool and press (Fig. 1). However, the more operations that are

8 5
1 ... Forming Zone
4
1 2 ... Properties of Material Prior Forming
3 ... Properties Workpiece
2
3 4 ... Boundary Between Workpiece and Tool
5 ... Forming Tool
6 ... Surface Development
7 ... Metal Forming Unit
8 ... Environmental Impact
6

Forming Tools (Die, Punch, Blank Holder), Fig. 1 Systematic consideration of forming procedures
Forming Tools (Die, Punch, Blank Holder) 555 F
Forming Tools (Die,
Punch, Blank Holder),
Fig. 2 production tool for
deep drawing of side panels
for the automotive industry
(Source: Audi AG)

integrated into one tool, the more complex is the to progressive cutting tools. They combine sev-
configuration of that tool. This results in higher eral cutting operations in one tool.
costs, reduced working velocity, and greater Bending tools are used to machine-angled
susceptibility to failure. parts. They are also called snakers when V- or
When designing and constructing the tool, the U-shaped angles are formed. Counterpunches are
specification for the forming machine to be used often used during U bending to ensure that the arc
later must always be taken into consideration. In of the bent part does not bulge. In die bending, the
addition to the geometrical installation condi- V or U shape is formed most precisely when there
tions, the mode of operation of the press, pressing is significant stamping pressure on conclusion of
forces, the sleeve diagram, the kinematics of the forming.
ram, permissible eccentricities, die space, ejector The constructive design of drawing tools for
and lifter functions, and much more play a role deep drawing is determined by the type of process
here (Altan et al. 2010). (deep drawing, redrawing, reverse drawing) and
the available press (single or double acting).
Forming Tool: Sheet Metal Forming In the case of large single-acting auto body
Sheet metal forming and cutting technology tools drawing tools like i.e. shown in Fig. 2, the draw-
can be divided into drawing and composite tools ing punch is arranged on the press bed and the
as well as cutting, bending, and stamping tools. blank holder lies on the sleeves of the die cushion,
Additionally, there are other types of tool for which is integrated in the base of the drawing
special procedures within metal forming. press. The drawing die is hereby always fixed at
Cutting tools are production means for cut- the ram, so that the drawn part when fully drawn
ting out, cutting in, cutting off, and profiling, as (but still in the tool) looks something like a hat
well as for punching, notching, and deburring with the drawing flange beneath (Fig. 3).
workpieces or making cuts from sheets, strips, Composite dies are production tools that
reels, and profiles of metallic materials. The implement several processes such as cutting,
procedure is divided into shearing, precision forming, and joining. These can be classified as
blanking, and counter cutting. Cutting opera- progressive dies (Fig. 4) and complete composite
tions can also be in a sequence. One refers here dies.
F 556 Forming Tools (Die, Punch, Blank Holder)

Progressive dies combine several cutting and production of various workpiece geometries.
forming operations in one tool, whereby the sin- With special tools either semifinished products
gle working stages are arranged one after the with determined cross-sectional dimensions can
other. In complete composite dies, various pro- be produced by drawing or extruding or compo-
duction methods are carried out within one stroke nents with determined geometric properties (e.g.,
of the press (Altan et al. 2010). by means of die forging or warm and cold
forging).
Forging Tool Furthermore, the tools for forging are differ-
Forging tools can be divided into universal and entiated according to the forming method that can
special tools. Universal tools are used for open be implemented by them (e.g., die forging or ring
die forging, which can be shared into the rolling). With this differentiation the forging
tools are divided into the forming operations
that can be carried out by them (e.g., during die
forging, divided into: upset tools, blocking tools,
finishing tools, trimming tools, and calibration
tools). Forging tools can be designed for one
forming procedure (usually to create relatively
simple component geometries) or several forging
operations (to create complex component geom-
etries, e.g., using multistage extrusion tools).
The simplified schematic diagrams in Fig. 5
show an open die forging tool and an extrusion
tool for a forging press with a simple ram move-
ment. A forging tool is normally comprised of an
upper and lower die (Fig. 5a), and the typical
design of an extrusion tool presupposes a punch
and a die (Fig. 5b). Sample images in Fig. 5c, d
Forming Tools (Die, Punch, Blank Holder), show an open die forging tool and a tool for
Fig. 3 Deep drawing tool for single action press cup-backward extrusion (Schuler 1998).

Forming Tools (Die,

Punch, Blank Holder),
Fig. 4 Progressive die
(Source: Fischer IMF
GmbH & Co. KG)
Functional Correlation 557 F
a b

upper die Punch

workpiece
workpiece

F
flash die

lower die

c d

Forming Tools (Die, Punch, Blank Holder), open die forging tool (Source: LEIBER Group GmbH &
Fig. 5 Elementary representation of (a) an open die forg- Co. KG) and (d) a tool for cup-backward extrusion
ing tool, (b) an extrusion tool, and sample images of (c) an

References
Functional Correlation
Altan T, Ngalle G, Shen G (2010) Cold and hot forging:
fundamentals and applications. ASM International, Christopher A. Brown
Materials Park Worcester Polytechnic institute, Worcester,
Schuler GmbH (1998) Metal forming handbook. Springer,
Berlin/Heidelberg MA, USA

Free Abrasive Machining Synonyms

▶ Lapping Roughness relations

F 558 Functional Correlation

Definition literature. This would appear to be especially true

of functional correlations of the second kind.
Functional correlation in surface metrology Surface topography is also called texture,
refers to the relations between the surface and particularly in mechanical and manufacturing
phenomena that influence or are influenced by engineering (ASME B46.1 2009), and is
the topography. Establishing functional correla- commonly referred to as roughness.
tions is one of the principal objectives of research Functional correlations of the first kind
in surface metrology. describe how surface topographies are modified
by interactions with their environment. Knowl-
edge of functional correlations of the first kind is
Theory and Application useful when designing manufacturing processes
(Brown 2000). Functional correlations of the first
There are two kinds of functional correlations. kind are also used to understand the use-wear that
These are called correlations of the first kind created surfaces on artifacts for archeologists
when they are between phenomena that created (Stemp et al. 2009) and what foods created dental
or modified the surface and the topography. They microwear for physical anthropologists
are called correlations of the second kind when (Scott et al. 2005) and to identify weapons in
they are between the topography and behavior, or forensics (Gambino et al. 2011). They are also
performance, of the surface (Fig. 1). useful in monitoring and controlling degradation
Currently there are probably many more phe- in cultural heritage preservation.
nomena that are topographically related than Functional correlations of the second kind
there are functional correlations reported in the describe how interactions with surfaces are

Performance scale Process scale

Topography Topography formation
performance Topography Process

Measurement Scale

MEASURED
TEXTURE

Analysis Scale

performance TEXTURE process

CHARACTERIZATION CHARACTERIZATION CHARACTERIZATION

second kind first kind

FUNCTIONAL CORRELATIONS

Functional Correlation, Fig. 1 The importance of scales in determining functional correlations of the first and
second kinds
Functional Correlation 559 F
influenced by the topography. Knowledge of important in establishing functional correlations:
functional correlations of the second kind is first, the geometrical nature of the topography
useful for the design of products. and, second, the scale of the characterization
Both kinds of functional correlations are (Brown and Siegmann 2001; Berglund et al.
useful for the advancement of surface science 2010a, b). The appropriate characterization of
and for understanding geological and biological a measured surface also depends on making the
surfaces. The latter can assist in the development measurements at the appropriate scales.
of biomimicry for applications in manufacturing The interactions that collectively produce the
and manufactured products. topographically related phenomena on rough sur-
A rational approach to the design of surface faces, which would be described by functional
topographies for products and the processes to correlations, occur at certain scales of interaction.
create them could be based on functional corre- These scales of interaction need to be included in F
lations of the second and first kinds, respectively. the scales of measurement in order to discover
To the extent to which functional correlations are functional correlations (Fig. 1).
used in product and process design would appear Form in some cases, particularly when it
to be hidden in trade secrets. deviates from what it is supposed to be, can
There is little in the literature on rigorous alter behavior or performance that was thought
experimental verification, especially on functional to be related to the roughness. This can exacer-
correlations of the second kind. Establishing bate the experimental determination of functional
functional correlations of the second kind usually correlations of the second kind.
requires more laborious experimental efforts than One of the simplest functional correlations of
establishing functional correlations of the the first kind with a theoretical and experimental
first kind. basis can provide an example of the transition
Initial understandings of quantitative func- between smooth and chaotic topographies
tional correlations can be developed through (Brown et al. 1996). This is the relation
modeling. These would then need to be verified between feed per revolution (f) in conventional
experimentally. Experimental quantitative func- machining by turning, the tool nose radius (r),
tional correlations should be reported with some and the peak-to-valley roughness (Rt or Rz).
analysis relating topographic characterization The peak-to-valley roughness is the maximum
parameters and phenomena of interest. minus minimum surface height. The relation
Information on functional correlations of the is given by
first and second kinds, of a qualitative sort, can be
found in textbooks and handbooks (e.g., Black Rz  f 2 =8r (1)
and Kohser 2012; Whitehouse 1994). For
addressing functional correlations of the first The development of this model is based on
kind, there are tables available that typically smooth geometries (Whitehouse 1994), even
suggest what range of roughness might be though it is below the scale considered as form in
achievable with what kinds of manufacturing this manufacturing operation. The scale of what
processes. Whitehouse (1994) gives a thorough would normally be considered to be the form is
discussion of experimental and theoretical bases given by the grosser movements of the tool and
for functional correlations of the first and second workpiece during machining. The development of
kind; however, there is little experimental verifi- the equation is based on the round nose of the tool,
cation and no regression testing of functional r, scraping out a helical arc-shaped groove in the
correlations reported. surface about a cylindrical workpiece (Fig. 2).
Establishing quantitative functional correla- When viewed in cross section, these arcs have
tions depends on being able to describe the topog- a maximum cord length equal to the feed, f.
raphy appropriately. There are two aspects of An image from a measurement of a turned
characterizing the topography that appear to be surface is shown in Fig. 3. The surface was turned
F 560 Functional Correlation

Functional Correlation, R z ª f 2 / 8r
Fig. 2 Modeling turned
surface roughness for f: feed
correlations of the first kind r: tool nose radius
at the scale of the feed and Regular
tool nose radius tool roughness, larger
scales
feed

r
Rz

Irregular roughness
finer scales

feed
tion
l mo
too

μm

700
18

Functional Correlation, 9

350
Fig. 3 Image of a surface

μm
turned with a feed of 0
150 mm and a tool nose 700
400 mm, rendered from
a measurement with 350
scanning laser confocal μm

0
microscope 0

with a tool with a nose radius of 400mmat a feed Figure 4 shows a functional correlation of the first
of 150 mm. The measurement was made with kind. The feed per revolution is regressed
a laser scanning confocal microscope with with the peak-to-valley roughness in turning.
a 50 objective. The large-scale tool form is The deviation between the theoretical and
clearly visible, and the regular peak-to-valley actual roughness is shown. It appears that the
roughness described by Eq. 1 is clear. At the minimum roughness that could be achieved
fine scales, there are details of a chaotic, or would be an of about 1.5 mm.
irregular, roughness that are not captured by the One inadequacy of Eq. 1 Rt as a characteriza-
peak-to-valley roughness described by Eq. 1. tion of the topography of the turned surface is
Experimentally the roughness has been shown their failure to capture the characteristics of the
to be highly correlated with the feed with topography at the finer scales. It is often at these
R2greater than 0.95 (Brown et al. 1996). At fine scales that the phenomena interact with the
feeds Eq. 1 applies as long as the chaotic topography and influence the performance. Func-
roughness inside the groove left by the tool nose tional correlations of the second kind that are
is less than that described by the equation. strong enough to be used in product design
At large feeds the equation applies as long as might not be achievable using only Rt as
the feed is not larger than the cord describing the parameter to characterize the roughness.
the limits of the round portion of the tool nose. Topographically dependent phenomena, like
Functional Correlation 561 F
Functional Correlation, 25
Fig. 4 A functional
correlation of the first kind
showing theoretical and measured
20
measured peak-to-valley Rz = f 2 / 8r + 1. 4 m m

Roughness Rt in μm
roughness, Rz
15

theoretical
10
R z = f 2 / 8r

5
F
0
0 50 100 150 200 250 300
Feed in μm/rev

reflectivity and adhesion, tend to depend on the Cross-References

finer scale, i.e., irregular or chaotic, roughness
inside the regular grooves left by the tool nose. ▶ Roughness
Examples of functional correlations of the ▶ Surface Parameter
second kind which have been discovered
experimentally include relating rake face
roughness to tool-chip friction (Koshy and
References
Tovey 2011), substrate roughness to adhesive
strength of thermal spray coatings (Brown and ASME B46.1-2009 (2009) Surface texture (Roughness,
Siegmann 2001), and the roughness to sheet Waviness and Lay), The American society of
metal friction (Berglund et al. 2010a, b). Theo- mechanical engineers, New York
Berglund J, Agunwamba C, Powers B, Brown CA,
retical support for these functional correlations
Rosén B-G (2010) On discovering relevant scales in
should relate the mechanisms of interaction with surface roughness measurement. An evaluation of a
the rough surface that influence the performance band-pass method. Scanning 32:244–249
to the topographic characterization parameters. Berglund J, Brown CA, Rosén B-G, Bay N (2010b)
Milled die steel surface roughness correlation with
In the case of rake face roughness, the skew
steel sheet friction. CIRP Ann – Manuf Technol
could be an indication of the tool-chip contact 59(1):577–580
area which is known to influence machining Black JT, Kohser RA (2012) DeGarmo’s materials
forces (Brown 1983). The correlation with ther- and processes in manufacturing, 11th edn. Wiley,
Hoboken, p 1037
mal spray adhesion is based on a characterization
Brown CA (1983) Practical method for estimating
of the area that is available for bonding. machining forces from tool-chip contact area. CIRP
The correlation appears to be specific at a scale Ann – Manuf Technol 32(1):91–95
which could be related to the bonding interaction. Brown CA (2000) Issues in modeling machined surface
textures. Machining Sci Technol 4(3):539–546
It appears that the sheet metal die friction could
Brown CA, Siegmann S (2001) Fundamental scales of
be related to the inclinations on the die surface at adhesion and area-scale fractal analysis. Int J Mach
a certain scale (Berglund et al. 2010b). However Tools Manuf 41:1927–1933
many kinds of topographic characterization Brown CA, Johnsen WA, Butland RM (1996) Scale-
sensitive fractal analysis of turned surfaces. Ann
parameters were found to correlate strongly
CIRP 45(1):515–518
with friction, provided the correct scale range is Gambino C, McLaughlin P, Kuo L, Kammerman F,
identified (Berglund et al. 2010a). Shenkin P, Diaczuk P, Petraco N, Hamby J,
F 562 Functional Specifications

Petraco NDK (2011) Forensic surface metrology: tool Theory and Application
mark evidence. Scanning 33(1–7):272
Koshy P, Tovey J (2011) Performance of electrical dis-
charge textured cutting tools. CIRP Ann Manuf Introduction
Technol 60:153–156 The concept of fuzzy logic emerged in 1965
Scott RS, Ungar PS, Bergstrom TS, Brown CA, Grine FE, within the development of the theory of fuzzy
Teaford MF, Walker A (2005) Dental microwear tex- sets by Lotfi A. Zadeh, professor of computer
ture analysis within-species diet variability in fossil
hominins. Nature 436(4):693–695 science at the University of California in
Stemp WJ, Childs BE, Vionnet S, Brown CA (2009) Berkeley (Zadeh 1965).
Quantification and discrimination of lithic use-wear: Later, in 1972, Michio Sugeno of the Tokyo
surface profile measurements and length-scale fractal Institute of Technology introduced the concepts
analysis. Archaeometry 51(3):366–382
of fuzzy measure and fuzzy integral. One of the
first control systems built using fuzzy set theory
was proposed in 1975 by Ebrahim Mamdani of
the University of London to control a steam
Functional Specifications engine and boiler combination by synthesizing
a set of linguistic control rules obtained from
▶ Requirement Specification skilled human operators. It is known as Mamdani
fuzzy inference method, and it is the most
commonly seen fuzzy methodology.
The basic concept underlying FL is that of
a linguistic variable, that is, a variable whose
Fuzzy Logic values are words rather than numbers. In effect,
much of FL may be viewed as a methodology for
Alessandra Caggiano computing with words rather than numbers.
Fraunhofer Joint Laboratory of Excellence on Although words are inherently less precise than
Advanced Production Technology, Department numbers, their use is closer to human intuition.
of Chemical, Materials and Production Furthermore, computing with words exploits
Engineering, University of Naples Federico II, the tolerance for imprecision and thereby lowers
Naples, Italy the cost of solution. Fuzzy logic effectively deals
with the trade off between significance and
precision. FL provides the opportunity for model-
Definition ing conditions that are inherently imprecisely
defined (Russell and Norvig 2003).
The term fuzzy logic has two different meanings.
More specifically, in a narrow sense, fuzzy logic, Fuzzy Sets and Membership Functions
FLn, is a logical system which may be viewed as Fuzzy logic starts with the concept of fuzzy set.
an extension and generalization of classical A fuzzy set is a set without a crisp, clearly defined
multivalued logics. But in a wider sense, fuzzy boundary that can contain elements with only
logic, FLw, is almost synonymous with the theory a partial degree of membership. A fuzzy set
of fuzzy sets (Zadeh 1975). defines a mapping between elements in the
input space (usually referred to as the universe
Extended Definition of discourse) and values in the interval [0, 1] (Teti
Basically, fuzzy logic (FL) is a multivalued logic, and Kumara 1997). The basic concept of the
which allows intermediate values to be defined theory is that of membership function, which
between conventional evaluations like true/false, specifies to what degree an element belongs to
yes/no, and high/low. Fuzzy logic is an extension a fuzzy set.
of the traditional logic to intermediate and Let X be a space of points (objects), with
approximate values. a generic element of X indicated by x. Thus,
Fuzzy Logic 563 F
X ¼ {x}. A fuzzy set A in X is characterized by subjective ideas and that can be interpreted
a membership function fA(x) which associates slightly differently by various individuals typi-
with each point in X a real number in the interval cally involve fuzzy propositions. Most natural lan-
[0, 1], with the value of fA(x) at x representing guage is fuzzy, in that it involves vague and
the “degree of membership” of x in A. Thus, the imprecise terms. Statements describing
closer the value of fA(x) to 1, the higher the a person’s height or weight or assessments of
grade of membership of x in A. When A is a set people’s preferences about colors or menus can
in the ordinary sense of the term, its membership be used as examples of fuzzy propositions (Ross
function can assume only the two values 0 and 1, 2004). The truth value assigned to a proposition
with fA(x) ¼ 1 or 0 according as x does or does not P can be any value in the interval [0, 1]. The
belong to A. Thus, in this case fA(x) reduces to assignment of the truth value to a proposition is
the familiar characteristic function of a set A actually a mapping from the interval [0, 1] to the F
(Zadeh 1965; fuzzy sets). universe U of truth values, T, as indicated
A membership function (MF) can be seen as
a curve characterizing how each point of the input T : u 2 Uð0; 1Þ
space is mapped to a membership value
(degree of membership or truth degree) between Fuzzy propositions are assigned to fuzzy sets.
0 and 1. The membership function can be any Suppose proposition P is assigned to the fuzzy set
arbitrary curve, the shape of which can be defined A, then the truth value of a proposition, T(P), is
as a function suitable from the point of view of given by T ðPÞ ¼ mA ðxÞ where 0  mA  1:
simplicity, convenience, speed, and efficiency. The equation above indicates that the degree
Typical membership function shapes are triangu- of truth for the proposition P: x ∈ A is equal to the
lar, rectangular, trapezoidal, Gaussian, sigmoi- membership grade of x in the fuzzy set A.
dal, etc. Of these, the simplest is the triangular The fundamental difference between classical
membership function composed of three points propositions and fuzzy propositions is in the
forming a triangle. The trapezoidal membership range of their truth values. While each classical
function is a truncated triangle curve with proposition is required to be either true or false,
a flat top. These straight-line membership the truth or falsity of fuzzy propositions is
functions have the advantage of simplicity. a matter of degree (Klir and Yuan 1995).
Other membership functions are built on the Assuming that truth and falsity are expressed by
Gaussian distribution curve and have the advan- values 1 and 0, respectively, the degree of truth of
tage of being smooth and nonzero at all points. each fuzzy proposition is expressed by a number
In summary: in the unit interval [0, l].
• A fuzzy set admits the possibility of partial
membership in it. Fuzzy Quantifiers
• The degree an object belongs to a fuzzy set In general, fuzzy quantifiers are fuzzy numbers
is denoted by a membership value between that take part in fuzzy propositions. They are of
0 and 1. two kinds. Fuzzy quantifiers of the first kind are
• A membership function associated with defined on R and characterize linguistic terms such
a given fuzzy set maps an input value to its as about 10, much more than 100, and at least
appropriate membership value. about 5. Fuzzy quantifiers of the second kind are
defined on [0, 1] and characterize linguistic terms
Fuzzy Propositions such as almost all, about half, and most.
Fuzzy logic is a method for reasoning with logical
expressions describing membership in fuzzy sets. Linguistic Hedges
A fuzzy logic proposition, P, is a statement involv- Linguistic hedges are special linguistic terms by
ing concepts without clearly defined boundaries. which other linguistic terms are modified.
Linguistic statements that tend to express A linguistic hedge or modifier is an operation
F 564 Fuzzy Logic

that modifies the meaning of a term or a fuzzy set. FL is based on a collection of IF-THEN rule
For example, if hot is a fuzzy set, then very hot, statements used to formulate conditional fuzzy
more or less hot, and extremely hot are examples propositions.
of hedges that are applied to that fuzzy set (Klir A single fuzzy IF-THEN rule assumes the
and Yuan 1995). Hedges can be viewed as oper- form:
ators that act upon a fuzzy set membership func-
tion to modify it. There are hedges that intensify if x is A then y is B
the characteristics of a fuzzy set (very,
extremely), that dilute the membership curve where A and B are linguistic values defined by
(somewhat, rather, quite), that form the comple- fuzzy sets on the universes of discourse X and Y,
ment (not), and that approximate a scalar to respectively. The if-part of the rule “x is A” is
a fuzzy set (about, close to, approximately). The called the antecedent or premise, while the then-
mechanics underlying the hedge operation is gen- part of the rule “y is B” is called the consequent or
erally heuristic in nature. For example, it is used conclusion.
frequently to implement the hedge slightly. An example of such a rule might be (Teti and
Zadeh’s original definition of the hedge very Kumara 1997):
intensifies the fuzzy region by squaring the mem- If service is good, then tip is average.
bership function at each point in the set. On the Fuzzy rule sets usually have several anteced-
other hand, the hedge somewhat dilutes the fuzzy ents that are combined using fuzzy operators.
region by taking the square root of the member- The combination is called a “premise,” and it
ship function at each point along the set. The generates a single truth value that determines
complement of very is a hedge group represented the rule’s outcome. In general, one rule by itself
by somewhat, rather, and quite. These hedges is not sufficient, but two or more rules that can
basically dilute the force of a fuzzy set member- play off one another are needed. The output of
ship function. A generalization of the dilator each rule is a fuzzy set, but in general, the output
hedge simply replaces the exponent of the inten- of an entire collection of rules should be a single
sification function with a real positive number number.
less than unity, expressed as a fraction (1/n).
Fuzzy Inference
Fuzzy Operators The actual process of mapping from a given input
In fuzzy logic, operators such as AND, OR, and to an output using fuzzy logic is called fuzzy
NOT are implemented by fuzzy intersection or inference and consists of the five steps illustrated
conjunction (AND), fuzzy union or disjunction in Fig. 1 (Teti and Kumara 1997).
(OR), and fuzzy complement (NOT). There are Step 1: Input variables fuzzification
various ways to define these operators, but com- This step consists in determining the degree
monly, AND, OR, and NOT logic operators are to which the inputs belong to each of the
implemented by the min, max, and complement appropriate fuzzy sets through membership
operators. The fuzzy truth, T, of a complex sen- functions.
tence is evaluated in this way: Step 2: Application of fuzzy operators
If a given rule has several parts, once the
T ðA L BÞ ¼ minðTðAÞ; TðBÞÞ inputs have been fuzzified, the fuzzy operators
TðA V BÞ ¼ maxðTðAÞ; TðBÞÞ are applied to resolve the antecedents to
T ðAc Þ ¼ 1
TðAÞ a single number between 0 and 1.
Step 3: Application of implication method
The implication method is defined as the
If-Then Rules shaping of the output membership functions.
Fuzzy sets and fuzzy operators are the subjects The input for the implication process is
and verbs of fuzzy logic. The processing core of a single number given by the antecedent, and
Fuzzy Logic 565 F

Fuzzy Logic, Fig. 1 Fuzzy inference steps

the output is a fuzzy set. Two commonly used Decision-Support Systems and Artificial
methods of implication are the minimum and Intelligence
the product. Fuzzy models are employed to support decision-
Step 4: Output aggregation making processes, sometimes combined with F
Aggregation is the process through which artificial intelligence paradigms. In the literature
the outputs of each rule are unified. The input related to manufacturing, fuzzy logic has been
to the aggregation is the output fuzzy sets exploited to support functions such as supplier
returned by the implication process for each management, through a fuzzy multi-agent
rule. The output is the combined output fuzzy decision-making strategy (Zhang et al. 2003), or
set. The most common aggregation methods to select the best automated manufacturing
are MAX (maximum), PROBOR (probabilis- system, through a fuzzy analytic hierarchy pro-
tic or), and SUM (simply the sum of each cess (AHP) method able to manage uncertainty
rule’s output set). and to consider productivity and flexibility issues
Step 5: Output defuzzification (Monitto et al. 2002). A fuzzy AHP method has
The resulting set must be defuzzified or been also applied to support decision making in
resolved to a single number (crisp value). evaluating machine tool alternatives (Ayag and
Some defuzzification methods are center of Özdemir 2006). In another case, a fuzzy program-
area (CoA), bisector, middle of maximum ming model has been developed to select process
(MOM), largest of maximum, and smallest of cutting speeds in a production plan as
maximum (Teti and Kumara 1997). Perhaps a compromise between particular and global tar-
the most popular defuzzification method is the gets (Noto La Diega et al. 1995). Fuzzy multi-
center of area (CoA), which returns the center criteria optimization methods have been
of area under the curve. efficiently applied to the network design of
In Fig. 2, the fuzzy inference flow from lin- integrated e-supply chains, as they support
guistic variable fuzzification to defuzzification of decision making even in case of subjective and
the aggregate output is shown: it proceeds up qualitative knowledge (Dotoli et al. 2007).
from the inputs in the lower left, then across Fuzzy systems usually do not incorporate auto-
each row, or rule, and then down the rule outputs matic learning abilities and adaptive features. It
to end in the lower right. seems that a very high performance can potentially
The described inference method, due to be obtained by combining neural networks (NN)
Mamdani (1974), is the most popular: other infer- and fuzzy logic approaches and integrating their
ence methods for fuzzy systems based on linguis- benefits (Monostori and Egresits 1994). Thus, the
tic rules are Sugeno models and Tsukamoto neuro-fuzzy (NF) approach can comply with real-
models (Mamdani 1974; Sugeno 1985). time management, uncertainty handling, and
learning ability. As an example, a neuro-fuzzy
Applications system is used to approach optimum tool inven-
In recent years, the number and variety of tory sizing of CBN grinding wheels for nickel-
applications of fuzzy logic have increased based alloy turbine blade fabrication (Teti and
significantly. Applications in manufacturing D’Addona 2003). A fuzzy-based methodology
cover a wide range including. combining artificial neural networks and a fuzzy
F 566 Fuzzy Logic

Fuzzy Logic,
Fig. 2 Fuzzy inference
flow

1. if and then

2. if and then

Input 1 Input 2

Output

inference system has been employed to reduce Fuzzy Control

computational complexity and to deal with multi- Fuzzy control is a methodology for constructing
ple performance criteria in the real-time schedul- control systems in which the mapping between
ing of dual resource constrained (DRC) systems real-valued input and output parameters is
where the shop capacity is constrained by machine represented by fuzzy rules. Fuzzy control has
and worker capacity (Araz and Salum 2010). An been very successful in commercial products
intelligent approach combining fuzzy logic and such as automatic transmissions, video cameras,
analytic hierarchy process (AHP), referred to as and electric shavers. In industrial automation,
fuzzy AHP, has been used for machine tool selec- fuzzy logic controllers can be used to minimize
tion to manage the large number of quantitative work in progress, production surplus, and
and qualitative attributes of the existing alterna- production backlog. In the literature, methodolo-
tives (Ayag and Özdemir 2006). A fuzzy approach gies developed for the design of such controllers
has been proposed to select the suitable material within manufacturing systems involve the use of
handling equipment in the presence of vagueness evolution strategies rather than genetic algorithms
(Onut et al. 2009). (Porter et al. 1997). Fuzzy logic controllers are
applied to manage complex production systems,
Robotics to develop optimal control of unreliable machines,
Fuzzy models have been employed in robotics to and to develop a manufacturing control system
establish the inverse dynamic model for a robot able to respond to system level disturbances
manipulator in its joint space (Qiao and Zhu (Homayouni et al. 2009; Yuniarto and Labib
2000) or to avoid complex analytical formulation 2005; Monfared and Yang 2004).
of isotropic target impedance and uncertainty of
parameters related to the robot and environment Pattern Classification with Fuzzy Sets
model through a new fuzzy impedance control Fuzzy pattern recognition methods have been
law (Petrovic and Milacic 1998). Furthermore, proposed in the literature for several applications,
fuzzy inference has been introduced into variable and they are the basis for classification expert
structure adaptive control for the nonlinear robot systems. As an example, they were used in
manipulator systems giving robustness against manufacturing to classify parts, based on the
system uncertainties and external disturbances idea that any characteristics of a part
(Zhao and Zhu 1995). corresponding to a certain class are fuzzy and an
Fuzzy Logic 567 F
absolute distinction does not exist (Zhu and The principal constituents of soft computing
Zhang 1987). They were also used in manu- the followings: fuzzy logic, neural computing,
facturing processes to monitor diamond tool evolutionary computation, machine learning,
wear status, considering the ambiguity in classi- and probabilistic reasoning.
fication as well as the weakness of the cutting
force variation (Ko et al. 1992). In recent years,
pattern recognition algorithms for data mining Cross-References
have greatly increased, from web data mining to
intelligent search engine and agents, as fuzzy ▶ Neural Network
logic can be used in making useful, human-
understandable, deduction from semi-structured
information (Nikravesh et al. 2002; Pal and References F
Mitra 2007).
Araz ÖU, Salum L (2010) A multi-criteria adaptive
control scheme based on neural networks and fuzzy
Fuzzy Modeling and Simulation inference for DRC manufacturing systems. Int J Prod
Fuzzy modeling has been effectively applied in Res 48(1):251–270
machining: as an example, Takagi-Sugeno- Ayag Z, Özdemir RG (2006) A fuzzy AHP approach to
evaluating machine tool alternatives. J Intell Manuf
Kang (TSK) fuzzy modeling has been employed
17(2):179–190
to accomplish the integration of multisensor Dotoli M, Fanti MP, Mangini AM (2007) Fuzzy multi-
information and tool wear information in turn- objective optimization for network design of
ing processes (Ren et al. 2011). Moreover, integrated e-supply chains. Int J Comput Integ Manuf
20(6):588–601
fuzzy logic is particularly suitable to model
Eversheim W, Hack T (1996) Managing multiple product
flexibility strategies; in the literature, applica- variants in assembly control with a fuzzy Petri-Net
tions of fuzzy modeling to the problems of approach. CIRP Ann Manuf Technol 45(1):45–48
manufacturing flexibility are illustrated (Zukin Homayouni SM, Hong TS, Ismail N (2009) Development
of genetic fuzzy logic controllers for complex produc-
and Young 2001). Modeling and simulation
tion systems. Comput Ind Eng 57:1247–1257
techniques such as fuzzy discrete event simula- Klir GJ, Yuan B (1995) Fuzzy sets and fuzzy logic: theory
tion have been applied to deal with production and applications, 1st edn. Prentice Hall, Upper Saddle
situations characterized by uncertainties, River, p 592
Ko TJ, Cho DW, Lee JM (1992) Fuzzy pattern recognition
which are often referred to as fuzzy production
for tool wear monitoring in diamond turning. CIRP
environments (Perrone et al. 2001). Fuzzy Petri Ann Manuf Technol 41(1):125–128
nets have been employed to represent knowl- Mamdani EH (1974) Application of fuzzy algorithms
edge about desired states or changes where for control of simple dynamic plant. IEEE
121(12):1585–1588
the degree of fulfillment is expressed by mem-
Monfared MAS, Yang JB (2004) Design of an intelligent
bership functions or transitions between states manufacturing scheduling and control system using
are modeled by fuzzy sets (Eversheim and fuzzy logic: sensitivity analysis and parameter
Hack 1996). optimization. J Intell Fuzzy Syst 15(2):89–104
Monitto M, Pappalardo P, Tolio T (2002) A new fuzzy AHP
method for the evaluation of automated manufacturing
Soft Computing systems. CIRP Ann Manuf Technol 51(1):395–398
Fuzzy logic represents one of the main elements Monostori L, Egresits C (1994) Modelling and monitoring
of soft computing. The latter differs from con- of milling through neuro-fuzzy techniques. In: Second
IFAC/IFIP/IFORS workshop on intelligent
ventional hard computing as it is tolerant to
manufacturing systems, 13–15 June 1994, Vienna
imprecision, uncertainty, partial truth, and pp 381–386
approximation. The principle behind soft Nikravesh M, Loia V, Azvine B (2002) Fuzzy Logic and
computing is to exploit the tolerance for the Internet (FLINT): Internet, World Wide Web, and
search engines. Int J Soft Comput 6(Spec Iss):287–299
imprecision, uncertainty, partial truth, and
Noto la Diega S, Passannanti A, Perrone G (1995)
approximation to achieve tractability, robustness, Machining economics in FMS by a fuzzy approach.
and low cost. CIRP Ann Manuf Technol 44(1):417–420
F 568 Fuzzy Logic

Onut S, Kara SS, Mert S (2009) Selecting the suitable Sugeno M (1985) Industrial applications of fuzzy control.
material handling equipment in the presence of vague- Elsevier Science, New York
ness. Int J Adv Manuf Technol 44(7–8):818–828 Teti R, D’Addona D (2003) Grinding wheel management
Pal SK, Mitra P (2007) Pattern recognition algorithms for through neuro-fuzzy forecasting of dressing cycle
data mining: scalability, knowledge discovery, and time. CIRP Ann Manuf Technol 52(1):407–410
soft granular computing. Chapman & Hall, Teti R, Kumara SRT (1997) Intelligent computing
Washington methods for manufacturing systems. CIRP Ann
Perrone G, Zinno A, Noto La Diega S (2001) Fuzzy 46(2):629–652
discrete event simulation: a new tool for rapid analysis Yuniarto MN, Labib AW (2005) Optimal control of an
of production systems under vague information. unreliable machine using fuzzy-logic control: from
J Intell Manuf 12(3):309–326 design to implementation. Int J Prod Res
Petrovic PB, Milacic VR (1998) A concept of an intelli- 43(21):4509–4537
gence fuzzy control for assembly robot. CIRP Ann Zadeh LA (1965) Fuzzy sets. Inf Control 8(3):338–353
47(1):9 Zadeh LA (1975) Fuzzy logic and approximate reasoning.
Porter B, Zadeh NN, Chisholm AWJ (1997) Evolutionary Synthese 30(3–4):407–428
design of fuzzy-logic controllers for manufacturing Zhang HC, Li J, Merchant ME (2003) Using fuzzy
systems. CIRP Ann Manuf Technol 46(1):425–428 multi-agent decision-making in environmentally con-
Qiao B, Zhu JY (2000) Fuzzy modeling of inverse scious supplier management. CIRP Ann Manuf
dynamics for robotic manipulators based on genetic Technol 52(1):385–388
algorithm. CIRP Ann Manuf Technol 49(1):321–324 Zhao DB, Zhu JY (1995) A fuzzy variable structure
Ren Q, Balazinski M, Baron L, Jemielniak K (2011) TSK adaptive control design for robot manipulators. CIRP
fuzzy modeling for tool wear condition in turning Ann Manuf Technol 44(1):363–368
processes: an experimental study. Eng Appl Artif Zhu JY, Zhang YZ (1987) An expert system of the part
Intel 24(2):260–265 classification. CIRP Ann Manuf Technol
Ross TJ (2004) Fuzzy logic with engineering applications, 36(1):351–353
2nd edn. Wiley, Chichester Zukin M, Young RE (2001) Applying fuzzy logic and
Russell SJ, Norvig P (2003) Artificial intelligence: constraint networks to a problem of manufacturing
a modern approach, 2nd edn. Prentice Hall, Upper flexibility. Int J Prod Res 39(14):3253–3273
Saddle River