Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

715

Quality of Ser

41. Quality of Service (QoS) of Automation

Heinz-Hermann Erbe (Δ)

Quality of service (QoS) of automation involves

issues of cost, affordability, energy, mainte-

nance, and dependability. This chapter focuses

on cost, affordability, and energy. (The next chap-

ter addresses the other aspects.) Cost-effective

or cost-oriented automation is part of a strat-

egy called low-cost automation. It considers the

lifecycleofanautomationsystemwithrespect

to their owners: design, production, operating,

and maintenance, reﬁtting or recycling. Afford-

able automation is another part of the strategy.

It considers automation or automatic control in

small enterprises to enhance their competitiveness

in manufacturing and service. Despite relative ex-

pensive components the automation system can

be cheap with respect to operation and main-

tenance. As examples are discussed: numerical

controls of machine tools; shop ﬂoor control with

distributed information processing; programmable

logic controllers (PLCs) shifting to general-purpose

(PC); smart devices, i. e. information processing in-

tegrated in sensors and actuators; and distributed

manufacturing, and maintenance.

Energy saving can be supported by auto-

matic control of consumption in households, ofﬁce

buildings, plants, and transport. Energy intensity

is decreasing in most developing countries, caused

by changing habits of people and by new control

strategies. Centralized generation of electrical en-

ergy has advantages in terms of economies of scale,

but also wastes energy. Decentralized generation

41.1 Cost-Oriented Automation..................... 718

41.1.1 Cost of Ownership ........................ 718

41.1.2 Robotics...................................... 719

41.2 Affordable Automation ......................... 721

41.2.1 Smart Devices .............................. 721

41.2.2 Programmable Logic Controllers

as Components

for Affordable Automation............. 722

41.2.3 Production Technology.................. 723

41.3 Energy-Saving Automation.................... 725

41.3.1 Energy Generation ....................... 725

41.3.2 Residential Sector ........................ 726

41.3.3 Commercial Building Sector ........... 726

41.3.4 Transportation Sector.................... 727

41.3.5 Industrial Sector........................... 727

41.4 Emerging Trends .................................. 728

41.4.1 Distributed Collaborative

Engineering................................. 728

41.4.2 e-Maintenance and e-Service ....... 730

41.5 Conclusions .......................................... 731

References .................................................. 732

of electricity and heat in regional or local units are

of advantage. A combination of wind energy, solar

energy, hydropower, energy from biomass, and

fossil fuel in small units could provide electrical

energy and heat in regions isolated from grids.

These hybrid energy concepts are demanding

advanced, but low-cost, controls.

The term low-cost automation was born in 1986 at

a symposium in Valencia sponsored by the International

Federation of Automatic Control [41.1]. However, the

use of this term led to a misunderstanding of low-cost

automation as a technology with poor performance,

although the intention was to promote affordable au-

tomation devices and to reduce the life cycle cost

or cost of ownership of automation systems. The in-

tention was also to bridge the gap between control

theory and control engineering practice by applications

using low-cost techniques. Ortega [41.2] pointed out

that the transfer of knowledge between the academic

Part E 41

716 Part E Automation Management

community and the industrial user is far from satis-

factory, and it still is, particular regarding small- and

medium-sized industry. However, developments in con-

trol theory are based on principles that can appeal to

concepts with which the practical engineer is famil-

iar. Despite the successfully demonstration of advances

of modern control methods over classical ones, actual

implementations of automatic control in manufacturing

plants show a preference for classical proportional–

integral–derivative (PID) control. Two key factors are

considered [41.3]:

1. Return on investment (better, cheaper, and faster)

2. Ease of application.

To be utilized broadly, a new technology must

demonstrate tangible beneﬁts, be easier to implement

and maintain,and/or substantiallyimprove performance

and efﬁciency. Sometimes a new control method is

not pursued due to poor usability during operation

and troubleshooting in an industrial environment. The

PID, due to its simplicity, whether implemented ana-

log or digital, provides advantages in its application.

However, manufacturing systems are becoming more

complex. Control needs include single-input/single-

output SISO and multiple-input/multiple-output MIMO

controllers. Therefore control techniques beyond PID

control become necessary. State-space methods are well

developed and provide many advantages if well under-

stood [41.3].

Low-cost automation is now established as a strat-

egy to achieve the same performance as sophisticated

automation but with lower costs. The designers of au-

tomation systems have a cost frame within which they

have to ﬁnd solutions. This is a challenge to theory and

technology of automaticcontrol, asthe main parts of au-

tomation. Low-cost automation isnot anoxymoron, like

military intelligence or jumbo shrimps. It opposes the

rising cost of sophisticated automation and propagates

the use of innovative and intelligent solutions at afford-

able cost. The concept can be regarded as a collection of

methodologies aiming at the exploitation of tolerance

to imprecision or uncertainties to achieve tractability,

robustness, and ﬁnally low-cost solutions. Mathemati-

cally elegant designs of automation systems are often

not feasible because of their neglect of real-world prob-

lems, and in addition are often very expensive for their

owners.

Cost aspects are mostly considered when designing

automation systems. However, in the end, industry is

looking for intelligent solutions and engineering strate-

gies for saving cost but that nevertheless have secure,

high performance. Field robots in several domains such

as manufacturing plants, buildings, ofﬁces, agricul-

ture, and mining are candidates for reducing operation

cost. Enterprise integration and support for networked

enterprises are considered as cost-saving strategies.

Human–machine collaboration is a new technological

challenge, and promises more than cooperation. Last

but not least, condition monitoring of machines to re-

duce maintenancecost andavoid downtime of machines

and equipment, if possible, is also a new challenge, and

promotes e-Maintenance [41.4] and e-Service [41.5].

The reliability of low-cost automation is indepen-

dent of the grade of automation, i.e., it covers all

possible circumstances in its ﬁeld of application. Often

it is more suitable to reduce the grade of automa-

tion and involve human experience and capabilities to

bridge the gap between theoretical ﬁndings and practi-

cal requirements [41.6]. On the other hand, theoretical

ﬁndings in control theory and practice foster intelli-

gent solutions with respect to saving costs. Anyway,

reliability is a must for all automation systems, al-

though this requirement has no one-to-one relation to

cost.

As an example one may consider computer-

integrated manufacturing (CIM). The original concept

of CIM connected automatically the design of parts

to machines at the workshop via shop ﬂoor planning

and scheduling software, and therefore used a lot of

costly components and instruments. After a while this

kind of automation turned out not to be cost effec-

tive, because the centralized control system had to ﬁght

against uncertainties and unexpected events. Decen-

tralization of control and the involvement of human

experience and knowledge along the added-value chain

of the production process required less sophisticated

hardware and software and reduced manufacturing cost,

which allowed CIM to break through even in small and

medium-sized enterprises [41.7].

Low-cost automation also concerns the implemen-

tation of automation systems. This should be as easy

as possible and also facilitate maintenance. Mainte-

nance is very often the crucial point and an important

cost factor to be considered. Standardization of compo-

nents of automation systems could also be very helpful

to reduce cost, because it fosters usability, distribu-

tion, and innovation in new applications, for example

ﬁeldbus technology in manufacturing and building

automation.

The components of an automation/control system,

such as sensors, actuators and the controller itself, can

incorporate advances in information technology. Lo-

Part E 41

Quality of Service (QoS)ofAutomation 717

cal information processing allows for integration at the

level of components, reducing total cost and provid-

ing new features in control lines. They may be wireless

connected because of low data stream rates. Smart

sensors and actuators are new developments for indus-

trial automation [41.8]. The cost of wireless links will

fall while the cost of wired connections will remain

about constant, making wireless increasingly a logi-

cal choice for communication links. However, wireless

links will not be applicable in all situations, notably

those cases where high reliability and low latencies are

required. The implication for factory automation sys-

tems is that processing and storage will become cheap:

every sensor, actuator, and network node can be eco-

nomically provided with unlimited processing power.

If processing and storage systems become inexpen-

sive relative to wiring costs, then the trend will be to

locate processing power near where it is needed in or-

der to reduce wiring costs. The trend will be to apply

more processing and storage systems when and where

they will reduce the cost of interconnections. The cost

of radio and networking technology has fallen to the

point where a wireless connection is already less ex-

pensive than many wired connections. New technology

promises to further reduce the cost of wireless connec-

tions [41.9].

Distributed collaborative engineering, i.e., the con-

trol of common work over remote sites, is now an

emerging topic in cost-oriented automation. Integrated

product and process development as a cost-saving strat-

egy has been partly introduced in industry. However, as

Nnaji et al. [41.10] mention, lack of information from

suppliers and working partners, incompleteness and in-

consistency of product information/knowledge within

the collaborating group, and incapability of processing

information/data from other parties due to the problem

of interoperability hamper effective use. Hence, collab-

orative design tools are needed to improve collaboration

among distributed design groups, enhance knowledge

sharing, and assist in better decision making. (See also

Chaps. 26, 88.)

Mixed-reality concepts could be useful for col-

laborative distributed work because they address two

major issues: seamlessness and enhancing reality.

In mixed-reality distributed environments information

ﬂow can cross the border between reality and virtual-

ity in an arbitrary bidirectional way. Reality may be

the continuation of virtuality, or vice versa [41.11],

which provides a seamless connection between the

two worlds. This bridging or mixing of reality and

virtuality opens up some new perspectives not only

for work environments but also for learning or train-

ing environments [41.12, 13]. (See also Chaps. 15,

86.)

The changing global context is having an impact on

local and regional economies, particularly on small and

medium-sized enterprises. Global integration and inter-

national competitive pressures are intensifying at a time

when some of the traditional competitive advantages –

such as relatively low labor costs – enjoyed by certain

countries are vanishing.

One can see growing emphasis on strategies for

encouraging supply-chain (vertical) and horizontal net-

working. These could be means for facilitating agile

manufacturing. Agile manufacturing is built around the

synthesis of a number of independent enterprises form-

ing a network to join their core skills, competencies,

and capacities to be able to operate proﬁtably in a com-

petitive environmentcharacterized by unpredictable and

continually changing customer demands. Agile manu-

facturing is underpinned by collaborative design and

manufacturing in networks of legally separate and spa-

tially distributed companies. Such networks are useful

in optimizing current processes and mastering sporadic

change in demand, material, and technology [41.14].

Distributed collaborative automation in manufacturing

networks is therefore an emerging area for automatic

control.

Energy-saving strategies and also individual solu-

tions for reducing energy consumption are challenging

politicians, the public, and researchers. The cost aspect

is very important. Components of controllers such as

sensors and actuators may be expensive, but one has to

calculate savings in energy costs over a certain period

or life cycle of a building or plant.

Energy provision and consumption based on ﬁnite

resources can cause conﬂicts between customers, and

between customers and suppliers. Energy savings are

necessary due to these ﬁnite resources. Effective use of

available energy can be supported by automatic control

of consumption in households, ofﬁce buildings, plants,

and transportation. Energy intensity is decreasing in

most developing countries. This is caused by changing

habits of people but also by new control strategies. The

use of energy in a country, in the residential sector, the

commercial building sector, the transportation sector,

and the industrial sector, inﬂuences the competitiveness

of the economy, the environment, and the comfort of the

inhabitants.

Building automation is generally a highly cost-

oriented business. The goal is to provide acceptable

comfort conditions at the lowest possible cost in terms

Part E 41

718 Part E Automation Management

of implementation, operation, and maintenance. Energy

saving is one of the most important goals in building

automation [41.15]. (See also Chap.62.)

In Sect.41.1 the strategy of cost-oriented automa-

tion will be explained with the examples of cost of

ownership and robotics. Section 41.2 continues, cov-

ering affordable automation with subsections on smart

devices, i.e., information processing embedded in sen-

sors, and actuators, programmable logic controllers

as important components of an affordable automa-

tion, and examples of low-cost production technology.

Section 41.3 covers energy saving with automatic con-

trol using the aforementioned strategies. (See also

Chap.40.)

41.1 Cost-Oriented Automation

Cost-oriented automation as part of the strategy of low-

cost automation considers the cost of ownership with

respect to the life cycle of the system:

•

Designing

•

Implementing

•

Operating

•

Reconﬁguring

•

Maintenance

•

Recycling.

Components and instruments could be expensive if life

cycle costs are to decrease. An example is enterprise

integration or networked enterprises as production sys-

tems that are vertically (supply chain) or horizontally

and vertically (network) organized.

Cost-effective product and process realization has

to consider several aspects regarding automatic con-

trol [41.13]:

•

Virtual manufacturing supporting integrated prod-

uct and process development

•

Tele- or web-based maintenance (cost reduction

with e-Maintenance systems in manufacturing)

•

Small and medium-sized (SME)-oriented agile

manufacturing.

Agile manufacturing here is understood as the synthesis

of a number of independent enterprises forming a net-

work to join their core skills.

As mentioned above, life cycle management of

automation systems is important regarding cost of

ownership. The complete production process has to

be considered with respect to its performance, where

maintenance is the most important driver of cost. Nof

et al. [41.16] consider the performance of the complete

automation system, which interests the owner in terms

of cost, rather than only the performance of the control

system; i. e., a compromise between cost of mainte-

nance and cost of downtime of the automation system

has to be found.

41.1.1 Cost of Ownership

A large amount of the cost of a manufacturing plant

over its lifetime is spent on implementation, ramp up,

maintenance, and reconﬁguration. Cost-of-ownership

analysis makes life cycle costs transparent regard-

ing purchase of equipment, implementation, operation

costs, energy consumption, maintenance, and recon-

ﬁguration. It can be used to support acquisition and

planning decisions for a wide range of assets that have

signiﬁcant maintenance or operating costs throughout

their usable life. Cost of ownership is used to support

decisions involving computing systems, vehicles, lab-

oratory and test equipment, manufacturing equipment,

etc.

It brings out the hidden or nonobvious ownership

costs thatmight otherwisebe overlooked in making pur-

chase decisions or planning budgets. The analysis is not

a complete cost–beneﬁt analysis. It pays no attention to

beneﬁts otherthan cost savings when different scenarios

are compared. When this approach is used in decision

support, it is assumed that the beneﬁts from all alterna-

tives are more or less equal, and that choices differ only

on the cost side.

In highly industrialized countries the type and

amount of automation mostly depends on the cost of

labor. However, automation in general is not always

effective [41.17]. A recent study carried out by the

Fraunhofer Institute of Innovation and System Tech-

nology regarding the grade of automation within the

German industry [41.6] found that companies are re-

fraining from implementing highly automated systems.

The costs of maintenance and reconﬁguration are con-

sidered too high; it would be more cost effective to

involve well-qualiﬁed operators. The same considera-

tions were discussed by Blasi and Puig [41.18]:

Automation is not useful by itself: there are a lot

of additional requirements to accomplish its func-

Part E 41.1

Quality of Service (QoS)ofAutomation 41.1 Cost-Oriented Automation 719

tion as needed, not only putting automation into the

factories, putting automation just where it is needed

and economically justiﬁed.

Blasi and Puig [41.18] consider the challengesof manu-

facturing processes of consumer goods. They stress that

automation helps in providing homogeneous operation,

but for many operations humans can sometimes carry

out the task better than automation systems. However,

humans are failure prone, emotionally driven, and do

not constantly operate at the same level, which also has

to be considered.

Morel et al. [41.19], as mentioned in the Introduc-

tion, stress the consideration of the whole performance

of a plant, rather than the control performance only,

which is what interests the owner regarding cost; i. e.,

a compromise between cost of maintenance and cost of

downtime of the automation systems in a plant regard-

ing commitments to customers or market must be con-

sidered. Scheduled maintenance can be shifted, but it is

of course better to implement condition-based mainte-

nance, in which degradation of parts can be observed

and the decision to run a machine or piece of equipment

for a certain extra time to fulﬁll a relevant customer

order can be made reasonably based on collected data.

Blasi and Puig [41.18] consider the conditions for

successful automation in industrial applications. Based

on their experiences, a manufacturing engineer dealing

with plant automation should bear in mind that proper

automation is much more than a matter of machines and

equipment. The organization of work in the whole plant

involving the human experience at all stages and always

considering possible improvements with automation or

not is the challenge for engineers and management

to save costs. Automation helps in providing homo-

geneous manufacturing processes. In many operations,

humans can sometimes carry out the task better without

automation. Humans may perform better than any auto-

matic machine, for the time being, but it is impossible

for them to assure constant quality. Where reasonable

cost is critical, inspecting the image quality of a screen,

for example, is a task reserved, for the moment, to hu-

mans, because computer vision is too expensive.

To summarize this section Fig.41.1 shows the main

cost contributions which are of interest to the owner of

the machine and manufacturing equipment.

41.1.2 Robotics

Robots were created in order to automate, optimize,

and ameliorate work processes that had up to then been

Design Stage

Lifecycle of

machine

and

equipment

Engineering design cost

Drawing cost

Computer processing cost

Design modification cost

Management cost

Production preparation cost

Implementation cost

Material cost

Facility cost

Production cost

Maintenance cost

Retrieval cost

Reconfiguration cost

Disassembly cost

Recycling cost

Ramp-up

stage

Disposal and

recycling stage

Manufacturing

stage

Fig. 41.1 Cost contributions of the life cycle of machine and equip-

ment

carried out by humans alone. For reasons of safety, hu-

mans may not enter the working space of the robot.

Recently it has been shown, however, that robots can-

not come close to matching the abilities or intelligence

of humans. Therefore, new systems which enable col-

laboration between humans and robots are becoming

increasingly important. The advantages of using inno-

vative robots can especially be seen in medical or other

service-oriented areas as well as in the emerging area of

humanoid robots.

Although robot technology is mostly regarded as

costly, one can see today applications not only in

customized mass production (automobile industry) but

also in small and medium-sized enterprises (SMEs),

manufacturing small lots of complex parts. Increasing

product variety and customer pressure for short deliv-

ery times put robots in the focus. Robots can be used

for loading and unloading of machine tools or other

equipment such as casting machines, injection mold-

ing machines, etc. Here mostly the robots are stationary,

and their control is coordinated with the control of the

machine or equipment they serve.

Evenﬁeld robotsare applicablewith affordable cost.

Ollero et al. [41.20] describe as an example (among

other applications) an autonomous truck, which could

also be an unmanned vehicle in a manufacturing site.

Low cost should be referred to the components to be

used as well as to system design and maintenance.

Sometimes a trade-off between general-purpose com-

ponents and components tailored to the application has

Part E 41.1

720 Part E Automation Management

Intelligent power assisting device

Realistic approach for complex

assembly and handling processes

in industry and service branche

Passive

handling

manipulator

Safety

Low costs

Simple operation

Industrial robot

Precision

Path control

Sensor-based control

Fig. 41.2 Cobots, a new class of handling devices, which combine the characteristics of robots and hand-guided manip-

ulators [41.24]

to be considered. Given the high cost of components,

the design of the ﬁeld robot should consider modularity,

and simple assembly, reliability maintenance, and fault-

detection properties are important for reducing cost in

terms of the life cycle. Sasiadek and Wang [41.21]

report on low-cost positioning of autonomous robots

fusing, data sensed by global positioning system (GPS)

and inertial navigation system (INS). Automation in

deep mining with autonomous guided vehicles (AGV)

saves cost in terms of healthcare of workers, provision

of energy (using fuel cells),and compressedair required

to maintain working conditions. For navigation, GPS is

not available for this application, therefore radio bea-

cons together with INS can be used. The vehicles and

their control are discussed by Dragt et al. [41.22]and

Sasiadek and Lu [41.23].

Wang et al. [41.25] developed a low-cost robot plat-

form for control education.

Applications of robots in automated assembly and

disassembly fail mostly because the environment can-

not be sufﬁciently structured. This was the reason for

the development of collaborative robots (cobots) or in-

telligent (power) assisting devices (I(P)AD) (Fig.41.2).

It was also motivated by ergonomic problems in as-

sembly of parts, where parts weight endangered the

human body. Cobots offer a cost-effective solution for

material handling. Complete automation of assembly

processes is complicated, if not impossible. Reconﬁg-

uration of robots performing assembly tasks could be

very costly. Humans, on the other hand, have capabili-

ties that are difﬁcult to automate, such as parts-picking

from unstructured environments, identifying defective

parts, ﬁtting parts together despite minor shape varia-

tions, etc. Cobots are passive systems which are set in

motion and guided by humans. The cobot concept sup-

poses that shared control, rather than ampliﬁcation of

human power, is the key enabler [41.26]. The main task

of the cobot is to convert a virtual environment, deﬁned

in software, into physical effect on the motion of a real

payload, and thus also on the motion of the worker.

Overhead gantry-style rail systems used in many shops

can beconsidered cobots but without the virtual surface.

Virtual surfaces separate the region where the worker

can freely move the payload from the region that cannot

be penetrated. These surfaces or walls have an effect on

the payload like a ruler guiding a pencil. Technically

cobots are based on continuously variable transmis-

sions (CVT). Peshkin et al. [41.26] developed spherical

CVTs, and Surdilovic et al. [41.24] developed CVTs

based on a differential transmission.

Part E 41.1

Quality of Service (QoS)ofAutomation 41.2 Affordable Automation 721

41.2 Affordable Automation

This strategy of low-cost automation focuses on mak-

ing systems affordable for their owners with respect to

the problem to be solved. An example is a manufactur-

ing systemin asmall ormedium-sized enterprise,where

automation increases productivity and therefore com-

petitiveness. Although small enterprises agree that at

least routine work can be done better when automated,

there is still a fear that automation will be sophisti-

cated, failure prone, need experts for maintenance and

reconﬁguration, and therefore would be costly.

Soloman [41.27] points to shortening product life

cycles that need more intelligent, faster, and more

adaptable assembly and manufacturing processes with

reduced setup, reconﬁguration, and maintenance time.

Machine vision, despite costly components, can reduce

manufacturing cost when properly applied [41.28]. In

order to survive in a competitive market it is essential

that manufactures have the capability to deploy rapidly

affordable automation to adapt to a changing manu-

facturing environment with increased productivity but

reduced production costs.

41.2.1 Smart Devices

Smart devices (sensors, actuators) with local informa-

tion processing in connection with data fusion are in

steady development, achieving cost reduction of com-

ponents in several application ﬁelds such as automotive,

robotics, mechatronics, and manufacturing [41.29,30].

One of the ﬁrst discussions on smart devices re-

garding cost aspects was given by Boettcher and

Traenkler [41.31].

The developments allow for computer-based au-

tomation system to evolve from centralized architec-

tures to distributed ones. The ﬁrst level of distribution

in order to reduce the wiring cost consisted of ex-

changing inputs and outputs through a ﬁeldbus as

a communication support. The second level integrates

data processing in a modular setting as close as possi-

ble to sensors and actuators. These smart sensors and

actuators can communicate, self diagnose or make deci-

sions [41.32]. One step to realize smart sensors is to add

electronics intelligence for postprocessing of outputs of

conventional sensors prior to use by the control sys-

tem. The same can be applied to actuators. Advantages

are tighter tolerance, improved performance, automatic

actuator calibration, etc.

Smart devices used in continuous systems bene-

ﬁt from the addition of microelectronics and software

that runs inside the device to perform control and di-

agnostic functions. Very small components such as

inputs/outputs blocks and overload relays are too small

to integrate data processing for technical/economic rea-

son. However, it is possible to develop embedded

intelligence and control for the smallest factory ﬂoor

devices. It is not always possible to implement data

storage or processing on each sensor or actuator. The al-

ternative solution is to implement data-processing units

connected to some sensors or actuators, connected to-

gether by communication links inorder toobtain remote

inputs/outputs. Current trends are for the development

of smart equipment associated to the ﬁeldbus, which

leads to a distributed architecture. Automation sys-

tems have evolved from a centralized to a distributed

architecture, yielding an automation system with an in-

telligent distributed architecture. Robots are following

this evolution, and increasingly becoming decomposed

into divided subsystems, each of which realizes an

elementary function. Distributed automation systems

yield several advantages, such as greater ﬂexibility,

simplicity of operation, and better commissioning and

maintenance. Today’s smart ﬁeld devices consist of two

essential parts: sensor or actuator modules, and elec-

RFID

Range

UMTS

GPRS

WI-Max

Zig Bee

Wi-Fi

NFC

Very

loosely

coupled

Loosely

coupled

Normally

coupled

Closely

coupled

Tightly

coupled

Blue Tooth

UWB

Fig. 41.3 Range and coupling modes of wireless technolo-

gies [41.9](UMTS – universal mobile telecommunications system,

GPRS – general packet radio service, WI-Max – worldwide inter-

operability for microwave access, Wi-Fi – wireless interoperability

ﬁdelity, RFID – radio frequency identiﬁcation, NFC – near ﬁeld

communication, UWB – ultra wire band)

Part E 41.2

722 Part E Automation Management

tronic modules. Microcomputers, initially responsible

for PID control, communication, etc., now have diag-

nosis functions included. Advanced diagnosis addresses

fault detection, fault isolation, and root analysis. The

early detection of anomalies, either process or device

related, is a key to improving plant availability and re-

ducing production costs.

When smart devices come together with wireless

communication a great cost saving can be achieved.

This infrastructure will be ﬂexible for reconﬁguration.

The reconﬁguration of existing software for the new

conﬁguration is sometimes very costly and has to be

considered. However, wireless communication eases

the problem of physically inﬂexible communication in-

frastructures.

In mobile devices wireless connections are manda-

tory. Distributed sensors in a wide area do not need

wireless communication, unless wiring is cost pro-

hibitive. Without cables, cost-intensive wiring plans are

not necessary. The freedom to place wireless sensors

and actuators anywhere in a plant or a building be-

comes limited if the devices need a main power source,

in which case power cables become necessary. It de-

pends on the sensors if they can use internal batteries or

can harvest energy from the environment. Several tech-

nologies for wireless communication are available atthe

market, with different standards and ranges (Fig.41.3).

Cardeira et al. [41.9] discuss the pros and cons of the

available technologies. They conclude that, in spite of

some initial skepticism, wireless communication is im-

posing itself as a complement to wired communication.

Location awareness is a new feature of wireless de-

vices. This feature may have a strong impact on service,

where the physical location of a device is important for

tracking, safety, security, and maintenance.

41.2.2 Programmable Logic Controllers

as Components

for Affordable Automation

Programmable logic controllers (PLCs) canbe regarded

as the classic components of affordable automation. An

affordable application was already reported by Jörgl

and Höld [41.33]. PLCs are meanwhile available with

full capabilities for less than US$100. A PLC can

be deﬁned as a microprocessor-based control device,

with the original purpose of supplementing relay logic.

Early PLCs were only able to perform logical opera-

tions. PLCs can now perform more complex sequential

control algorithms with the increase in microprocessor

performance. They can admit analog inputs and out-

puts. The main difference from other computers is the

special input/output arrangements, which connect the

PLC to smart devices as sensors and actuators. PLCs

read, for example, limit switches, dual-level devices,

temperature indicators, and the positions of complex

positioning systems. On the actuator side, PLCs can

drive any kind of electric motor, pneumatic or hydraulic

cylinders or diaphragms, magnetic relays or solenoids.

The input/output arrangements may be built into a sim-

ple PLC,orthePLC may have external I/O modules

attached to a proprietary computer network that plugs

into the PLC.

PLCs were invented as less expensive replacements

for older automated systems that used hundreds or

thousands of relays. Programmable controllers were

initially adopted by the automotive manufacturing in-

dustry, where software revision replaced rewiring of

hard-wired control panels. The functionality of the PLC

has evolved over the years to include typical relay

control, sophisticated motion control, process control,

distributed control systems, and complex networking.

There are other ways for automating machines, such

as a custom microcontroller-based design, but there are

differences between the two approaches: PLCs contain

everything needed to handle high-power loads, while

a microcontroller would need an electronics engineer

to design power supplies, power modules, etc. Also

a microcontroller-based design would not have the ﬂex-

ibility of in-ﬁeld programmability of a PLC,whichis

why PLCs are used in production lines. Typically they

are highly customized systems, so the cost of a PLC is

low compared with the cost of contacting a designer for

a speciﬁc one-time-only design.

The earliest PLCs expressed all decision-making

logic in simple ladder diagrams (LDs) inspired by elec-

trical connection diagrams. Electricians were quite able

to trace out circuit problems with schematic diagrams

using ladder logic. This was chosen mainly to address

the apprehension of technicians. Today, the line be-

tween a personal computer and a PLC is thinning.

PLCs have connections to personal computers (PCs)

and Windows-based software-programming packages

allow for easy programming and simulation. With the

IEC 61131-3 standard, it is now possible to program

using structured programming languages and logic ele-

mentary operations. A graphical programming notation

called sequential function charts is available on cer-

tain programmable controllers. IEC 61131-3 currently

deﬁnes ﬁve programming languages for programmable

control systems: function block diagram (FBD), ladder

diagram (LD), structured text (ST; similar to the Pas-

Part E 41.2

Quality of Service (QoS)ofAutomation 41.2 Affordable Automation 723

cal programming language), instruction list (IL; similar

to assembly language), and sequential function chart

(SFC). These techniques emphasize logical organiza-

tion of operations. Susta [41.34] presents a method to

convert a PLC program in either of the languages men-

tioned above into programming statements, which can

be used either for low-cost emulation of the program

or as an auxiliary tool when a debugged PLC program

is moved to another cheaper hardware, for instance, to

a PC.

Continuous processes cannot be accomplished fast

enough by PLC on–offcontrol. Thecontrol systemmost

often used in continuous processes is PID control. PID

control can be accomplished by mechanical, pneumatic,

hydraulic, or electronic control systems, as well as by

PLCs. PLCs, including low-cost ones, have PID control

functions included, which are able to accomplish pro-

cess control effectively. To program control functions

IL, ST or derived function block diagrams (DFBD) are

used.

41.2.3 Production Technology

Within the last years so-called shop-ﬂoor-oriented

technologies have been developed [41.17], and have

achieved success at least but not only in small and

medium-sized enterprises (SMEs). These technologies

are focused on agile manufacturing, which involves

using intelligent automation combined with human

skill and experience on the shop ﬂoor. With shop-

ﬂoor-oriented production support, human skill and

Surface modeler

for moldmaking

CAD/CAM

system

CNC milling

Link between CAD

and CNC-control

CNC program of

CAD/CAM via

post-processor

Direct using of

CNC programs

Tool catalogue and

data for tool setting

available at all CNC

CNC turning CNC milling Tool setting

equipment

LAN

Shopfloor oriented

programming systems

Planning board

(available

at all CNC-controls)

Link to

intra- or

Internet

Link of

digitized

geometry

Precision scan

(digitizing parts)

Fig. 41.4 Modules of a shop ﬂoor network

automation create synergetic effects to master the man-

ufacturing process.

Automation provides the necessary support to ex-

ecute tasks and rationalize decisions. This represents

a strand of low-cost automation. Running the manu-

facturing process effectively is not only a question of

technology, though this is essential. Together with ad-

equate work organization in which human skills can

be developed, it establishes the framework for cost-

effective, competitive manufacturing in SMEs.

Recent achievements for manufacturing are:

•

Shop ﬂoor planning and control based on operators

experience

•

Low-cost numerical controls for machine tools and

manufacturing systems (job-shop controls).

Shop ﬂoor control is the link between the administrative

and planning section of an enterprise and the manufac-

turing process at the shop ﬂoor. It is the information

backbone of the entire production process. What at least

small and medium-sized enterprises need is shop ﬂoor

control support to use the skills and experience of the

workforce effectively. However, it should be stressed

that this is support, not determination of what to do

based on centralized automatic decision making.

In small-batch or single production (molds, tools,

spare parts), devices for dynamic planning are desired.

Checking all solutions to the problems arising on the

shop ﬂoor while taking into account all relevant restric-

tions with short-term scheduling outside the shop ﬂoor

either by manual or automatic means is not very effec-

Part E 41.2

724 Part E Automation Management

tive. It is senseless to schedule manufacturing processes

exactly for weeks ahead. However, devices that are ca-

pable of calculating time corridors are of advantage.

The shop ﬂoor can do the ﬁne planning with respect to

actual circumstances much better then central planning.

Human experience regarding solutions, changing

parameters, and interdependencies is the very basis of

shop ﬂoor decisions and needs to be supported rather

than replaced.

Figure 41.4 shows a network system linkingall rele-

vant modules to be used by skilled workers. Apart from

necessary devices such as tool-setting, an electronic

planning board is integrated. The screen of the board is

available at all computer numerical control (CNC) con-

trols to be used at least to obtain information on tasks

to be done at certain workplaces to a certain sched-

ule. As all skilled workers in a group are responsible

for the manufacturing process they have, beyond access

to information, the task of ﬁne planning of orders they

received with frame data from the management. They

use electronic planning boards at the CNC controls or

alternatively at PCs besides the machines.

Planning and scheduling support with soft con-

straints was developed by [41.35], called Job-Dispo

(Fig.41.5). It consists mainly of an intelligent planning

board with drag-and-drop functionality, a graphic ed-

itor, and a structured query language (SQL) database

running on PCs under Windows. In SMEs, manufac-

turing groups are empowered to regulate their tasks

themselves based on frame data from the manage-

ment. The operators receive only rough data for orders

Fig. 41.5 Electronic, PC-based planning board for shop ﬂoor

use [41.35]

from central management, concerning delivery time,

material, required quality, supply parts, etc. Using this

support the workers decide on the sequences of tasks

to be done at the different parts of the manufactur-

ing system. The electronic planning board simulates

the effects of their decisions. Normally more than

one task needs the same resource at the same time.

The operators are enabled to change the resource

limits of workplaces and machines (working time,

adding or changing shift work, etc.) until the simu-

lation results in acceptable work practice and fulﬁlls

customers’ demands. If this cannot be achieved, cen-

tral management has to be involved in the decision.

Job-Dispo is partly automatic but allows for soft con-

straints to make use of the experience of human

operators.

Machine tools have a key position in manufactur-

ing. Their functionality, attractiveness, and acceptance

are determined by the efﬁciency of their control sys-

tems. Typical control tasks can be divided into the

numerical controller, programmable logic control, con-

trols for drives, and auxiliaryequipment. Manufacturers

of computer numerical control (CNC) systems are in-

creasing effort on developing new concepts permitting

ﬂexible control functions with a broad scope of free-

dom to adapt the functions to the speciﬁc requirements

of the planned application in order to increase the use

of standardized components. This simpliﬁes the inte-

gration of CNC of different manufactures into the same

workshop environment. Even today, CNC systems are

mainly offered as closed manufacturer-speciﬁc solu-

tions andsome machinetool builders developtheir own,

sometimes sophisticated, controls. This keeps the shop

ﬂoor inﬂexible, but ﬂexibility is only needed in small

and medium-sized enterprises. Recently there has been

a growing tendency to use industrial or personal com-

puters as low-cost controllers in CNC technology, with

a numerical control (NC) core on a card module, but

connected to standard operating systems such as Win-

dows. Because PC hardware is easily available and

under constant development, CNC based on it beneﬁts

from technological advances in this ﬁeld.

One of the low-cost applications of this kind of

CNC is calledjob-shop control,which arenowavailable

on the market (e.g., Siemens, Heidenhain). Machine

tools with job-shop control can be operated manually

and if needed or appropriate also with additional soft-

ware support. This is intelligent support, enabling the

switch in process from manual to numerical support.

These controls are suitable for job shops with small lots

and are easy to handle manually (conventional turning

Part E 41.2