nternational journal of computer integrated manufacturing , tập 23, số 8+9, 2010

EDITORIAL Semiotics-based Manufacturing System IntegrationThe primary intention of this special issue SI is to promote semiotics as a candidate framework for a new Manufacturing System I

EDITORIAL Semiotics-based Manufacturing System Integration

The primary intention of this special issue (SI) is to

promote semiotics as a candidate framework for a new

Manufacturing System Integration (MSI) paradigm

Considering semiotics as an emergent discipline in

engineering, there is a question: are there, and

(supposing there are) what are the implications of

semiotics for the MSI? In other words, could semiotics

be a new theoretical, or meta-theoretical, base for the

conceptual, theoretical or practical shift in MSI

discipline? Or, to be inclusive (or conservative): could

semiotics represent an extension of the solution space

in conceptual, theoretical, and/or practical terms, for

the MSI discipline?

The need for the new MSI paradigm is more and

more obvious for several reasons:

(1) There are problems for which the ‘traditional’

approaches, which are based on the semantic

field, have no capacity to resolve, referring

the problems that lead to the extremely high

percentage of ‘failed’ and ‘challenged’ projects,

such as the problems of acceptance of the

engineering solutions by the clients despite

technical correctness of the solutions, not

managing incomplete specifications because of

not understanding fully the clients’ needs, and

similar, i.e not managing the problems in

which the root is, in fact, not the technical

correctness and/or information transaction but

the technical solutions and information use,

interpretationand communication

(2) The growing complexity of the MSI issues,

manifested through the growing frequency of

emergence of new MSI tools of different kinds,

and, consequently the requirements for their

faster development and integration with the

existing tools – in other words, the growing

dynamics of the MSI issues

Put simply, the primary intention of this SI on

‘Semiotics-based Manufacturing System Integration’

could be understood as ‘opening’ the question on a

new paradigm of MSI

The secondary intention, or more ‘conservatively’

the secondary objective, of this SI is more ‘traditional’:

to present some research results on the use of semiotics

for MSI In this respect, this SI comprises 11 papersthat present some solutions, results, analyses andapproaches that use the semiotics instruments for theimprovement of various aspects of MSI These will bepresented in more detail later

The major question is, what exactly is semiotics?Semiotics could be seen as a meta-theoreticalframework for new research and development of theMSI discipline In its most simple definition, semiotics

is the science of ‘signs’ The signs could be linguistic ornon-linguistic.1The following two great scientists andthinkers are considered as the ‘fathers’ of semiotics:Charles Sanders Peirce (1839–1914), American logicianand founder of pragmatism, and Ferdinand deSaussure (1857–1913), Swiss linguist, the ‘father’ ofmodern linguistics and structuralism, who laid downindependently the basis of semiotics as a science on thetransition of the nineteenth to twentieth century.The domain of semiotics comprises three fields:syntax, semantics and pragmatics While syntax andsemantics are well known in the MSI science,pragmatics is almost totally unknown as a discipline

In a review of the publications in the field of MSI, itwas not possible to find any contribution thataddresses the pragmatics issues, except for verysingular references in a wider context of MSI andwithout any further elaboration, or presentation, ofresearch results and/or models However, in the areas

of enterprise integration/‘interoperability’ (EI) (notnecessarily addressing MSI) and information systems(IS) the situation is relatively different Consideringthe extent to which semiotics is used as an instrument

in and a meta-model of the science of engineering, thegreatest, implying a relevant, extent semioticsachieved was in the area of IS, while in the area of

EI the extent of achievement could be said very low,

if relevant at all, and in the area of MSI it ispractically inexistent The promotion of semiotics inthis SI has a significant success, originating todayalready a larger community and a greater number ofresearch works and publications

Semiotics could be seen as having a double relation

to the science2of MSI: it is both an instrument of MSIand a meta-model of the science of MSI Semiotics asthe instrument of MSI means that the MSI as adiscipline uses the models, mechanisms and proceduresVol 23, Nos 8–9, August–September 2010, 687–690

ISSN 0951-192X print/ISSN 1362-3052 online

Ó 2010 Taylor & Francis

DOI: 10.1080/0951192X.2010.513809

of semiotics as an independent science for the objective

of improvement of the MSI mechanisms, tools and

processes Semiotics as the model, or a

meta-theory, of the science of MSI means that semiotics of

the actual ‘traditional’ science, or discipline, of MSI is

being investigated From this perspective, semiotics is

used as an explanatory and unification framework, or

as an ‘organon’, of the sciences in general (Morris

1938), ‘since every science makes use of and expresses

its results in terms of signs’, and might be used (as it is

still not ) as an explanatory and unification

frame-work, or as an ‘organon’, of the science of MSI

In this SI, the relation of the MSI to semiotics is

primarily instrumental, meaning that semiotics is used as

a new instrument for improving MSI science and

discipline This is valid for nine papers while two papers

have used semiotics as a perspective for analysing two

advanced scenarios of emergent MSI domains

Of course, the ‘second’ relation between the MSI

and the semiotics, i.e the semiotics of the MSI (whether

‘traditional’ or ‘semiotic’), is expected to be one of the

subjects, or even an exclusive subject, of some future

SIs

Now, let me discuss about the papers that are

integrated in this SI

With regard to the semiotics applications in MSI,

the papers can be summarised as follows:

The papers in which the semiotics-based

instru-ments are referred to and/or employed are as follows:

The first paper represents an introductory paper byG.D Putnik and Z Putnik, on the semiotics-basedMSI concept The paper presents the findings of anexploratory research on the potential of semiotics forMSI The findings strongly suggest that semioticsmight be the base for a new paradigm of MSI In thefirst part, the paper introduces the basic notions ofsemiotics relevant for the MSI The second partpresents a framework for the semiotics-based MSIand a model of the semiotics-based MSI, called

‘Generative Integration’ In the third part, someexperimental set-ups, i.e prototype demonstrators ofthe Manufacturing Systems, are presented, as a plat-form for future research and development of thesemiotics-based MSI

The second paper by F.v Eijnatten and G.D.Putnik presents the technique of dialogue as anintegration mechanism The paper is expected to be

of interest for the readership as it clarifies thedifferences between the dialogue techniques andthe discussion, or argumentation, techniques whichare usually confused, leading to inconsistent use of theterms ‘dialogue’ and ‘discussion’ (or ‘argumentation’)and consequently to inconsistent applications of thesetechniques In terms of dialogue use in MSI, it is used

as a generative integration instrument for the creation

of a manufacturing system as a learning organisationand in those organisations looking for the organisa-tional renewal

The third paper by N Jing and S Lu presentscollaborative negotiation as another generative inte-gration instrument Concerning semiotics, the negotia-tion processes belong to pragmatics, representing a ‘co-construction process’ which is, in fact, a form ofintegration On the object-level, the authors present anew framework for improvement of the negotiationprocesses

An additionally interesting aspect for the readers,concerning the second paper, is the opportunity tocompare the negotiation processes based on argumen-tation with the dialogue technique (presented in thefirst paper) and to notice that the use of these twodifferent techniques is justified in the totally differentorganisational contexts

In the next paper, A Zelitchenko presents erative integration instruments on two levels On thelower level a ‘standard language with formal syntaxand informal unlimited vocabulary for the pragmaticdescription of the projects’, i.e of the co-operativeprojects is presented On the higher level another newMSI instrument is presented: a new type of the socialnetworks named ‘acting social network’, i.e the socialnetworks that facilitate ‘common action’, contrary tothe traditional social networks oriented to facilitation

gen-of social communication and exchange gen-of information

No ofpapers Paper(s)

Pragmatics –society 3 (þ2) 9–11 (þ4, 6)

The ‘acting social network’ aims to join potential

partners in a new-generation of manufacturing systems

(the virtual enterprises) The proposed mechanisms

imply, actually, an upgrade of the well-known

seman-tic web to the emerging pragmaseman-tic web – also virtually

a new term for the MSI discipline

The fifth paper by M Janssen and R Feenstra

provides a generative integration instrument

present-ing a service portfolio, as one of the pragmatic web

instruments The service portfolio used

communica-tion, synchronisation and generation of meaning

among the stakeholders for the purpose of decision

making on composition and reconfiguration of

man-ufacturing chains, enabling rapid composition of

manufacturing supply chain processes from reusable

components This paper is the second paper that

addresses the pragmatic web, through the specific

instrument service portfolio, as an MSI semiotic-based

instrument

The paper by L.A Ripamonti and C.A Peraboni

presents virtual worlds as one of the integration

mechanisms Concretely, the multi-user virtual

envir-onments (MUVE) are presented and discussed This

type of mechanism is especially oriented to the

integration of (manufacturing) virtual enterprises in

which the inherent dynamics of reconfigurations

impose higher levels and qualitatively different

bar-riers, such as personality, cultural, language,

organisa-tional and physical, for the effective and efficient

integration MUVEs augment the actual lives of their

users and support effectively the extension of people’s

actual social networks, fostering social interaction and

knowledge sharing, in fact supporting pragmatic and

social levels of the semiotic framework The MUVEs

also present a pragmatic web instrument and in this

way, this paper refers to the pragmatic web too

The seventh paper by B Provideˆncia and J

Ciurana presents how human communication as a

generative integration instrument could be integrated

with a traditional CAD–Rapid Manufacturing system

to create a powerful system for highly personalised

product development In the Guest Editor’s opinion,

the presented system might represent an initial model

of how the traditional manufacturing system too (not

only the new manufacturing system structures and

tools, such as, e.g social networks, dialogue,

negotia-tion, pragmatic web, etc.) might be enhanced by using

‘semiotic tools’

The next paper by J Andersson Schaeffer, J

Cadavida & T Backstro¨m presents a non-linguistic

generative integration instrument It is the ‘continuous

improvement area’ Actually, the paper explores

spatial design in continuous improvement areas and

also explains how spatial design may hinder or support

communication regarding improvements The semiotic

aspects of the spatial design for continuous ment areas in industry implies a different perspectiveand includes aspects of cognition, information, com-munication and treats how and what the elements inthe improvement areas communicate The improve-ment areas serve as a complement to the integration

improve-of manufacturing through computers Althoughnon-linguistic, the continuous improvement area isconsidered as a pragmatic-based instrument

In the paper by F Romero, a generative tion instrument on a social level is described Thisintroduces the term ‘institution’, where an ‘institution’

integra-is a socially devintegra-ised construct Examples of institutionsmay include, among others, organisations with educa-tional purposes (schools, universities), economic pur-poses (firms, trade unions, cooperatives), regulatorypurposes (certification organisations, supervising agen-cies) or political purposes (political parties, govern-ment agencies) The paper elaborates the hypothesis onthe institutions as the integration instruments throughsecondary analysis of the literature, and from primaryand secondary analyses of case studies, and shows thatsocial factors condition in important ways the shapesand even the possibility of the implementation andintegration of manufacturing systems

The tenth paper by P.J.G Garrido has usedsemiotics for analysing the organisational architecture

as a generative integration mechanism Concretely, theorganisation of Open Design that originates from theFree Software and Open Source Software organisa-tions is analysed The generative nature of theOpen Design organisational architecture, and thesemiotics-based integration instruments, relays inthe organisation’s openness From the other side, theorganisation’s openness implies a kind of an under-lying communication and action social network

In the Guest Editor’s opinion, the Open (Design,Manufacturing, etc.) organisational architecturesmight represent the emergent new generation, and anew paradigm, for manufacturing systems and enter-prises Obviously, these models are intrinsically depen-dent on the semiotics-based integration paradigm

In the eleventh and final paper in the issue by R.Jardim-Goncalves, A Grilo, T Hassan & A Steiger-Garc¸a˜o, an analysis is introduced which is based on theorganisational semiotics perspective of The EuropeanCommission’s social-technical study envisioning thesingle integrated information space, i.e the SingleElectronic Market The study was grounded at atechnological level on the ATHENA framework forsystems interoperability, complemented with non-Technological Interoperability Metrics (nT-IM) Thepaper analyses the vision of stakeholders and discussesthe correlation of the technical framework and non-technical interoperability measurement indicators with

semiotic levels and theories for MSI, identifying the

principal challenges for MSI when characterised by the

semiotic levels according to the stakeholders’

recog-nised issues

At the end, it is expected that this SI improves this

‘state-of-the-art’, at least to ‘open’ the research that

would result in definitive adoption or refutation of

semiotics as an instrument of MSI and/or a

meta-model of the science of MSI

There is also a hope that this SI on semiotics will

deserve the MSI research community’s attention

adequate to the great potential of semiotics for the

MSI, and for engineering in general

Acknowledgments

First, our acknowledgments and greatest thanks go to Prof

Stephen Newman, Editor-in-Chief of the International

Journal of Computer Integrated Manufacturing (IJCIM),

and Dr Aydin Nassehi, Managing Editor, for their highest

support and professionalism and, more importantly, their

highest collaboration, understanding and patience during the

development of this SI We would like to underline

Prof Newman’s vision and openness to such an innovative

project as this issue intends to be (the readers will judge)

Next, our acknowledgments go to the authors, for their

contributions and collaboration and especially to the

authors who also served as reviewers, and other reviewers,for their great effort during the review process and for thesuggestions they provided to the authors, without which thiscomplex and challenging project would not have been madepossible

Notes

1 For some semioticians, semiotics is a study of linguistic signs only The study of linguistic signs belongs

non-to linguistics as a separate discipline

2 Paraphrasing Morris (1938) for the purpose of MSI

ReferenceMorris, C 1938 Foundations of the theory of signs In: O.Neurath, R Carnap, and C Morris, eds Internationalencyclopedia of unified science, Vol 1, no 2 Chicago:University of Chicago Press, 1–2 (reprint in: Morris, C

1971 Writings on the general theory of signs, Hague, TheNetherlands: Mouton, 17)

Goran D PutnikDepartment of Production andSystems Engineering, Campus of Azurem

University of MinhoGuimara˜es, PortugalEmail: putnikgd@dps.uminho.pt

A semiotic framework for manufacturing systems integration –

Part I: Generative integration model

Goran D Putnika,b* and Zlata Putnikba

Department of Production and Systems Engineering, University of Minho, Portugal;bInterdisciplinary Centre for Production

Technologies and Energy, University of Minho, Portugal(Received 2 May 2010; final version received 20 July 2010)This paper presents the findings of exploratory research on the potential of semiotics for manufacturing systemsintegration (MSI) The findings strongly suggest that semiotics might be the basis for a new paradigm for MSI Inthe first part of the paper a number of needs for the new semiotic-based integration paradigm are presented Thesecond part of the paper introduces the basic notions of semiotics and provides a discussion on the use of semiotics

in MSI The third part presents a framework for the based MSI, together with a model of the based MSI, entitled ‘generative integration’ (GI) In the final part, some experimental set-ups, i.e prototypedemonstrators of the manufacturing systems, elements and systems, are presented as a platform for future researchand development of the semiotics-based MSI

semiotics-Keywords: manufacturing system; CIM; integration; interoperability; generative integration; semiotics; pragmatics;semantics

1 Introduction

The relevance of the semiotic approach in a social

context in engineering has emerged in response to the

failure of the traditional ‘technocentric’ approach to

today’s information systems (IS) and organisations’

requirements as well as to the ‘software development

crisis.’ (Note: semiotics has been introduced in

different engineering fields at various times and to

different levels of extent Considering the extent to

which semiotics is used as an instrument and a

meta-model of science of engineering, the greatest, implying

the most relevant, extent was achieved in the area of

information systems (IS), while in the area of EI the

extent of achievement could be said to be relatively

very low, if relevant at all, and in the area of MSI it is

practically nonexistent However, because of the need

for brevity in this paper an extensive presentation of

the state-of-the art in the use of semiotics in

engineer-ing will be presented in a subsequent publication.) The

software development crisis is manifested by a very

high percentage of failed and/or ‘challenged’ projects,

the percentage that goes up to 70% of all projects (the

percentages are in fact different in different sectors.)

(The Standish Group Int 2005) These percentages are

also referred to as ‘shameful numbers’ (Note: The

Standish Group categorises projects into three

resolu-tion types: (a) Successful: the project is completed on

time and on budget, with all the features and functions

originally specified (b) Challenged: the project iscompleted and operational, but over-budget, over thetime estimate, and with fewer features and functionsthan initially specified (c) Failed: the project iscancelled before completion or never implemented.)While it can be safely claimed that the use ofadvanced computer technologies is contributing tofurther improvements of (integration) system’s effi-ciency (e.g reduction of the throughput time), thecontribution of the ‘technology-oriented’ approach tothe projects failure rate (in the above terms) is lessobvious, if not detrimental Actually, according to theStandish Group Int.’s report the top five factors forproject success are not technological (in fact, theformal methodologies, and standard tools and infra-structures – that could be considered as technologicalmeans – are ranked 9th and 10th in the list of 10) Inother words, ‘Tools by themselves do not promotesuccess; the proper use of the tools does.’

According to R Stamper [‘A Dissenting Position’

in FRISCO Report (Falkenberg et al 1998)], thetraditional approach fails ‘not for technical reasons –most delivered software performs efficiently to speci-fication – but for organisational reasons – they do notrelate correctly to the world of business reality Thesad fact is that in general, technical people do notunderstand business problems and business-orientedpeople do not understand the need for detailed, formal

*Corresponding author Email: putnikgd@dps.uminho.pt

Vol 23, Nos 8–9, August–September 2010, 691–709

ISSN 0951-192X print/ISSN 1362-3052 online

Ó 2010 Taylor & Francis

DOI: 10.1080/0951192X.2010.510292

precision, which contributes to the problem.’ In the

FRISCO Report, it is identified that ‘there are at least

three major sources of problems: (a) the large variety

of interest groups, (b) conflicting philosophical

posi-tions, and (c) the lack of understanding

communica-tion.’ Also, it was realised that ‘the social, cultural and

organisational aspects play more decisive roles than

technology itself’ (Liu 2000) According to Moor and

Weigand (2002), information systems should be

approached ‘much more as communication systems

than computation systems’ To deal with the

commu-nication systems, the ‘move away from the traditional

information flow paradigm, in which positivistic

modelling of symbol manipulating functions aimed is

needed (Stamper 2000) The information systems

built on the information field paradigm do not produce

sterile data, but aim to generate and communicate

information that can lead to knowledge that helps

people to perceive, understand, value, and act in the

world’ (Moor and Weigand 2002) A good illustration

of what it means is given in Tables 1 and 2

When considering manufacturing systems

integra-tion (MSI), it is obvious that the MSI problem is just a

particular case of the information systems Therefore,

it could be claimed that there is a major integration/

interoperability problem relating to MSI, e.g Newman

et al (2008) Actually, the transaction-based

integra-tion is, in many cases, effective only within limited,

well-organised domains and on the lower levels ofcommunication Even for the traditional enterprisedomains, e.g intra-enterprise domains such as en-terprise resource planning (ERP), the traditionalapproaches failed to provide effective and efficientsolutions

Example 1 For example, theoretically, the system integration architecture (based on a standards,i.e based on a ‘neutral-format’ data file) is consideredmore flexible and more efficient than federatedintegration architecture However, the practice refutesthis idea In practice, an integration-federated archi-tecture that uses 250 prebuilt adaptors ‘is capable ofimmediately connecting to virtually any informationsystem, rapidly integrating more data sources on moreplatforms and across more network protocols than anyother integration solution’ (InterSystems 2004) Devel-opment of those 250 adaptors is probably less time andenergy-consuming than the development of standardsfor the same application domain (for problems onstandards and ontologies development, see, e.g Libes

open-et al (2004) and Nell (1998))

Example 2 Another example is from the area ofCAD, where supposedly the product geometrical dataneutral format standards are well defined However, ithappens that original equipment manufacturing(OEM) companies require, or more correctly, forcetheir cooperation on partners to have exactly the sameCAD software as the OEM companies This requiresthe OEM supply companies to solely have totallyidentical processors and post-processors for the samestandards used by the OEM company The truemeaning of this is that the data standards arephenomenologically irrelevant because an OEM couldchoose some totally proprietary CAD software andforce the partners to use it For each OEM’s partner itmeans that it is required to have as many differentCAD softwares for each OEM it cooperates Thisreally portrays the reality of how reliable commercialtools are considered for integration based on standarddata formats Ironically, the researchers and fundingbodies believe the integration problem to be solved,with researchers now investigating areas such asoptimisation of the CAD processes management.From the semiotics perspective, the above examplesare explained considering the semiotic fields, theirordering and abstraction degrees The above examplesmean that although the standards might be welldefined logically, formally and socially – in terms ofthe social agreement on definitions, the practice issomehow different and the real solutions pass on theside of the socially agreed standard definitions Whilethe standards could be considered the solutions on thesyntactic and semantic levels, the integration solutionsfrom the above examples could be considered as

Table 1 From information to communication systems

(Moor and Weigand 2002)

Informationsystems

Communicationsystems

Supports: Transaction

processes

CommunicationprocessesDesign objects: Clear specifications ‘Fuzzy’ process

definitionsDevelopment

Information flow IS information field

Responsibility: Anonymous Individual

responsibilitiesDesign process: Representation Interpretation

value, act

solutions on the pure pragmatic level From the

semiotics point of view the differences between the

semantics and pragmatics-based solutions, in relation

to reality, are due to different degrees of abstractions

of semantics and pragmatics, which confirms the need

of consideration of pragmatics, i.e semiotics in

integration problems resolution

Further, when integration is considered in a

complex, dynamic, nonlinear, ‘chaotic’ environment,

the problems might be even more difficult Actually, in

a complex, dynamic, nonlinear, ‘chaotic’, ‘uncertain’

environment, MSI implies 1) dynamic establishment

and management of the interactions, and 2) novel,

emergent, instantaneous as well as ‘ad-hoc’

(‘synchro-nic’) integration solutions, among manufacturing

systems tools, agents and stakeholders, intra- and

especially inter-organisations (intra- and

inter-enter-prises) It is not realistic to expect that each tool, agent

and stakeholder within the organisation and each

perspective partner in an inter-organisation

relation-ship, is in possession of ‘perfect’ standard-based

solutions, the state-of-the-art technology, and even

the ‘ideal’ knowledge It means that in such imperfect

situations, which are part of the real world, the

partners must have the ability to create or to

synthesise efficiently an effective integration solution

Moreover, the partners must have the ability to

implement integration process management, as MSI

is a complex process and not only a data

transac-tional process For that reason, the authors have

coined the term ‘generative integration’ The first

formulation of the generative integration appeared in

Putnik et al (2005) This paper presents an improved

formulation

The rest of the paper presents, as the second part,

an introduction to basic notions of semiotics The third

part discusses the generative integration as a model of

semiotics-based MSI and the fourth part is where some

experimental set-ups, i.e prototype demonstrators ofthe manufacturing systems, elements and systems, arepresented, as a platform for the future research anddevelopment of semiotics-based MSI

Concerning the scientific method, the paper sents an exploratory research based on the secondarydata analysis

pre-2 Semiotics – an introduction to basic notions

In its most condensed form, the definition of semiotics,

or semiology, is the study of signs In modern use,semiotics, or semiology, was conceived as a generaltheory of signs at the turn of the 20th century in theworks of two great scientists and thinkers: CharlesSanders Peirce (1839–1914), American logician andfounder of pragmatism (Figure 1), and Ferdinand deSaussure (1857–1913), Swiss linguist, the ‘father’ ofmodern linguistics and structuralism (Figure 2).Fascinatingly, both formulated semiotics, or semiology,respectively, independently and almost at the sametime

2.1 ‘Sign’

The central concept of semiotics is the ‘sign’ Peircegave a number of definitions of the ‘sign’, as well ascomments on various aspects of the ‘sign’ from which

we select the following:

‘I define a Sign as anything which is so determined bysomething else, called its Object, and so determines aneffect upon a person, which effect I call its Interpretant,that the latter is thereby mediately determined by theformer.’ (A Letter to Lady Welby, SS 80–81, 1908)

‘No sign can function as such except so far as it isinterpreted in another sign’ (CP 8.225n10 1904.07[Draft probably of a letter to Paul Carus])

Peirce’s definition of the ‘sign’ (S) is a ‘structure’ ofthree constituents, or a ‘triadic relation’: object (O),

Figure 1 Charles Sanders Peirce (1839–1914) Figure 2 Ferdinand de Saussure (1857–1913)

representamen (R), interpretant (I), or, in other words,

the sign is a triple: S¼ 4 (O, R, I) These are usually

graphically represented as in Figure 3

2.2 Sign process

The ‘sign process’ means the process of creation of

signs through effects produced by some other sign(s)

This process is called ‘semiosis’ Peirce described

semiosis as:

‘But by ‘semiosis’ I mean, on the contrary, an action, or

influence, which is, or involves, a cooperation of three

subjects, such as a sign, its object, and its interpretant’,

CP 5.484, 1907 (Robin 318, Pragmatism)

As the sign may produce another triadic relation,

i.e the sign may be interpreted in another sign, we

have a sequence of interpretations Consequently, ‘the

process of referring effected by the sign is infinite.’,

Jean-Jacques Nattiez (1990: 7) (cited in

http://psycho-logy.wikia.com/wiki/Signifier), e.g Figure 4

2.3 Field of observation and/or ‘information field’

Change of the sign along time Saussure has called

diachrony, or diachronic dimension/perspective of sign,

while its state, and relations with other signs, without

any consideration of time, i.e without any

considera-tion of diachronic perspective, is called synchrony, or

synchronic dimension/perspective of sign Concerning

these two dimensions of the sign we have a

phenom-enon of the major importance for our thesis on

semiotic-based MSI (and of IS in general) This

phenomenon is the ‘field’ of observation The ‘field’ of

observation emerges from the fact that the user is

unaware of diachronic dimension and considers only

the synchronic dimension

‘ Speech contains the seeds of every change, each

one being pioneered in the first instance by a certain

number of individuals before entering into general

usage This form, constantly repeated and accepted

by the community, became part of the language

They enter our field of observation only when they havebecome accepted by the community.’ (Saussure 1916,p.97) [bold formats by GP, ZP]

Implication of the ‘field of observation’ is thatactually it is hardly possible to exist an ‘absolute’,common and universal, interpretation of reality, but,rather, there are multiple interpretations by multiplecommunities and in different times

This phenomenon is similarly referred in tion systems (IS) by Stamper (1999) (referredindependently), as the information field The informa-tion field is defined by overlapping groups of peoplethat share norms (Stamper 2009), or informationfield(s) is(are) field(s) of norms shared by organisa-tional agents and govern their behaviour (Filipe2004), or information field(s) is(are) field(s) of

informa-‘subcultures with shared norms’ (Stamper 1999).(Note: Norms exist in a community and will governhow members behave, think, make judgements andperceive the world The shared norms are whatdefine a culture or subculture A subculture may be ateam who know how to work effectively together,and their norms include a solution to their organisa-tional problems Norms are, actually, a special kind

of signs A norm is more like a field of force thatmakes the members of the community tend tobehave or think in a certain way.’ (Stamper et al.2000) Also, a norm is ‘the unifying concept acrossall the layers of the organisational structure’ (Filipe2004)) Information field forms one of the basicconcepts within the semiotic framework for IS (seeSection 1) Obviously, the ‘group of people that

Figure 3 Representation of the Peirce’s sign elementary

structure

Figure 4 Peirce’s sign: representation of the sequence ofsign interpretations in other signs

share norms’ are Saussure’s communities that share

signs The concept could be graphically represented

as in Figure 5:

‘Shared norms constitute what is called the ‘social

reality’ – something not given at once for all, but

constantly in the process of being redefined and

renegotiated’

2.4 Semiotic fields

Three particular fields of semiotics study have been

identified by Charles Morris (1946): syntactics,

seman-tics and pragmatics These fields are called semiotic

fields, (not to be confused with ‘field of observation’ or

‘information field’), which are the fields of dyadic

relations among the three correlates of the triadic

relation of semiosis Charles Morris (1946) defined

them as follows:

‘pragmatics deals with the origin, uses and effects of

signs within the behavior in which they occur;

semantics deals with the signification of signs in all

modes of signifying;

syntactics deals with combination of signs without

regard for their specific significations or their relation to

the behaviour in which they occur.’ (Morris 1946:302)

The universally accepted order among the three

semiotic fields, introduced by Carnap (1942), is based

on their degree of abstractness in relation to complete

signs and semiosis:

‘If in an investigation explicit reference is made to the

speaker, or, to put it in more general terms, to the user

of language, then we assign it to the field of

pragmatics If we abstract from the user of the

language and analyse only the expressions and their

designate, we are in the field of semantics And if,

finally, we abstract from the designata also and analyse

only the relations between the expressions, we are in

(logical) syntax.’ (Carnap 1942: 9) (cited in Recanati

(2004))

This criterion could be considered of the maximumimportance as it ‘reveals’ proximity to the reality ofsyntactics, semanticsand pragmatics

3 Generative integration (GI)Generative integration (GI) is characterised by theability to create, synthesise or generate the integrationsolution – effectively in the first place and thenefficiently

The fundamental and qualitative new feature ofgenerative integration is employment of semiotics as

an instrument, especially employing pragmaticsinstruments

Speaking about pragmatics, pragmatics is notanother model or another representation of informa-tion, as sometimes misinterpreted and/or sometimeswrongly reduced to semantical representation.Rather, pragmatics is an information meta-modelwhich treats information as a nondeterministicprocess in which any attempt of ‘writing’, i.e to

‘fix’ or to formalise information in fact prevents thecreative process of interpretation, i.e prevents co-design and/or co-creativity Thus, a pragmaticapproach is live communication – speech, i.e ‘parole’(Saussure) – in a synchronic, or paradigmatic,dimension of the communication language, and notlanguage, i.e ‘langue’ (Saussure), as a formal, orformalised, or normed, structure, in its diachronic, orsyntagmatic, dimension

Therefore, for the semiotics-based MSI, and itsgenerative integration model, pragmatics is a distin-guishing feature

The fundamental integration mechanism for thesemiotics-based MSI, and its generative integrationmodel, is a live human communication – speech(‘parole’), mainly by verbal signs, but also by non-verbal signs, with associated underlying cognitiveprocesses

Figure 5 (a) Information fields, overlapping groups of people that share norms (Stamper 1999); (b) an example of communitiessharing their own information fields (Stamper 2009)

This is quite different than the ‘traditional’

perspective on organisations for which achievement is

mainly up to semantics and which

‘ assumes that communication consists of an exchange

of information in and out of an organization

Organiza-tion structure is conceived to be uni-dimensional and

fixed, and for this reason it is thought to be amenable to

managerial design’ (Saludadez and Taylor 2006)

The new, semiotic-based paradigm provides us with

a different perspective on, and capability for,

develop-ment of organisations, MS and MSI For example,

Saludadez and Taylor (2006) wrote:

‘In the communication perspective , by contrast,

organisation is conceptualised as grounded in a social

process of interpretation ( ) [The original text refers

various sources/references which are omitted on

assumption that they are not critical in the context of

the citation purpose in this paper Of course the reader

could consult them through the original text The

symbol ‘( )’ means that on that place is referred a single

source, while the symbol ‘( )’ means that on that place

is referred multiple source.] Organization is created

and recreated ( ) in and through the everyday

sense-making activities of its members ( ) From this

perspective, organizational structure is conceived to

be multiple ( ) rather monolithic ( ); fluid or even

fragmented ( ), rather than fixed; socially constructed

( ) rather than static ( ); context-sensitive ( ) and

historical ( ) rather than acontextual or ahistorical;

emergent ( ) rather than consciously designed

Accord-ing to this view, communication and organisation are

coconstructing ( ) In other words, it is through the

process of communication that organisational forms

emerge And communication is in turn framed and

informed by organisation

At the core of this view of communication is the

principle of a circular dynamic: from conversation to

text (text being interpreted broadly to include verbal as

well as written expression of ideas), and from text to

conversation (conversation being understood as

interac-tion grounded in, and concerned with, practice) It is the

circularity of the communicative dynamic – what

Giddens (1984) called its recursive character - that

explains organising Situations, patterns of interaction,

contexts, and motives are all sited in the everyday

ongoing flow of experience and yet they only take on

meaning when they have been interpreted, tively, to become not merely an account of whathappened but as what actually did happen (Varey,chap 10, this volume), now transformed into the context

retrospec-of future interaction.’ (Saludadez and Taylor 2006)

Consequently, humans are at the centre of tion The integration fields are individuals andcollectives, their interior and exterior, and the integra-tion fields location is in humans

integra-Individuals and collectives form different nities intra- and inter-organisations, creating multipleand dynamic fields of observation and/or ‘informationfields’ as the integration fields

commu-Indeed, a semiotics-based Integrated ManufacturingSystem is a set of semiotic-based MSI models, i.e a set

of communication models, i.e a set of ‘fields ofobservation’ or ‘information fields’, in continuouschange, Figure 6

Paraphrasing Guiraud’s (1975) scheme, Figure 7presents an explanation of the sign changes and creationand changes of ‘fields of observations’, or ‘informationfields’, for MSI, and influence of the ‘individual/community’ and ‘society’ on the MSI paradigm, throughthe enrichment or ‘impoverishment’ of the potential forinnovation and novelty of the MSI and of the potentialfor MSI effectiveness or efficiency

Promoting space for the individual and nity-based generation and interpretations of signs, thesemiotics-based MSI could be seen as – paraphrasingGuiraud (1975) – the generator of the creative power inmanufacturing systems and MSI systems, as the

commu-‘maker’ and ‘inventor’ of new integration instruments,providing higher levels of coherence with the environ-ment, and social reality Figuratively, semiotics is thepoie´te(‘maker’) of MSI

In other words, ‘making’, and the ‘inventing’ of MSI

is grounded in pragmatics, which are based oninterpretations Katambwe and Taylor (2006) wrote onthe ambiguity, as a manifestation of interpretations, that

‘ambiguity is an inevitable consequence of beingsimultaneously different and unified As Eisenberg(1984) and others have argued, ambiguity is ‘an

Figure 6 A semiotics-based integrated manufacturing system as a set of semiotics-based MSI models, i.e a set ofcommunication models, i.e a set of ‘fields of observation’ or ‘information fields’, in continuous change

inherent and sometimes necessary element of human

interaction and social life’ ( )’ (Katambwe and Taylor

2006)

and demonstrated

‘the value of ambiguity as an integrative mode of

organising paradox or not, the achievement of

both (differentiation and integration – GP and ZP),

simultaneously, is an outcome, when it happens, of

communication, and not a desideratum both

differentiation and integration have to be enacted

in the conversation of members of the organisation

on a continuing basis Maintaining differentiation

with integration, we argue, is not a once-and-forever

achievement; both have to be renegotiated in the

everyday jostling of running a business (or

depart-ment of governdepart-ment, or nonprofit association) It

was the challenge of discovering how the mix

integration and differentiation is constructed,

prag-matically, in the course of managerial conversation,

that stimulated us to focus on exchange such as the

one we cited at the outset.’ (Katambwe and Taylor

2006)

Table 3 presents some fundamental dimensions of

semiotics-based MSI and its Generative Integration

analytical, instrumental and solution space Obviously,

semiotic-based MSI space is multi-dimensional Most

of these dimensions are already discussed in the

previous text However, some other dimensions, such

as the dimension of integration fields from the

(organisational) complexity perspective, are also

cri-tical Table 4 presents some integration instruments for

the semiotic-based MSI

The generative integration implies an integration

life cycle, Figure 8 (Putnik et al 2005) It consists of

three global phases: synthesis (or design), operation,

and termination

In the phase ‘integration synthesis’ (or design, orgeneration), representing a communication-based inte-gration process, integration generation mechanisms, ortools, are used, some of which are presented in Table 4 Inthis phase the integration processes locations are humans.Therefore, it implies further that the underlying philoso-phy, or the paradigm, is manufacturing system seen as ahuman centred organisation, in which humans are thesubjects of organisation, autonomous and conscious, trueagents of creation, innovation and actions

Figure 7 Influence of the ‘individual/community’ and ‘society’ on the MSI paradigm characterisation

Table 3 Integration dimensions of semiotics-based tion

integra-Integration dimensions – semiotics basedSemiotics relation to the MS

InstrumentalMeta-theoreticalSemiotics fieldPragmaticsSemanticsSyntacticsIntegration fieldsIndividual internal – IntentionalIndividual external – BehaviouralCollective internal – CulturalCollective external – SocialChange

Synchronic – ParadigmaticDiachronic – SyntagmaticCognitive

Interpretations Knowledge LearningType of signsLinguisticNonlinguistic

However, it should be clear that the role of humans

in semiotic-based integration might be quite different

than in the ‘traditional’ semantics-based integration,

strongly informed by the cybernetic paradigm, where

humans also participate but are instrumentalised,

making their role the role of objects and instruments

of the system, complementing the system, or

substitut-ing misssubstitut-ing system’s machine and automaton

compo-nents (which miss, i.e are not realised, or not

implemented, either because of costs or because of a

lack of knowledge of how to implement them)

Therefore, the authors call the ‘traditional’

manufac-turing system paradigm a ‘cybernetic’ or ‘automaton’

paradigm, phenomenologically intended to achieve

manufacturing system and its integration as absolute

automaton (e.g flexible, adaptive ‘unmanned factory’,

‘intelligent factory’, etc as a mechanistic and

cyber-netic dream of the ‘final solution’ for production

(manufacturing))

In the generative integration’s ‘integration

synth-esis’ phase the issues such as how the particular MS

architecture affects the integration process, how do the

social requirements, culture, (international) law,

ecol-ogy, energy, etc., affect the particular integration

solutions, how the particular individuals affect in

particular, and similarly are taken into account in

accordance with the concrete, particular and

instantaneous needs and an effective integrationsolution is created

It could be said also that in this phase of thegenerative integration life-cycle the organisational per-spective of the manufacturing system and MSI is infocus

In the phase the ‘integration operation’, afterfinishing the ‘integration synthesis’ phase, the issuessuch as data formats, shared databases, standards,ontologies, deterministic algorithms, wrappers, con-tracts (as fixed documents), etc., are of concern Thisphase actually represents a ‘traditional’ transaction-based integration process, which is an integrated partand an underlying process of the generative integrationand which assures transmission of the signs as well asprovides capability of automation of particular aspectsand subprocesses, especially on the physical andsemantical/syntactical levels

While in the phase ‘integration synthesis’ theintegration fields are individuals and collectives, andthe integration fields location is in humans, in thephase ‘operation’ the integration fields and theirlocation is in machines, i.e in computers and in IT.Thus, for the integration life-cycle phase ‘opera-tion’ it could be said that the technological perspective

of the MS and MS integration is in focus

Finally, the ‘integration termination’ phase of theintegration process life cycle corresponds to either asimple disconnection of the two elements (in bothgenerative integration and in transaction-based inte-gration), or it may require further sub-processes toterminate, especially in the generative integration, asthe communication-based integration

It is important to notice that the semiotic-basedMSI, i.e its generative integration model, does not

‘eliminate’ or ‘abrogate’ the ‘traditional’ based integration, but, rather, represents an extensionand a wider and inclusive paradigm For example, thegenerative integration life-cycle phase ‘integrationoperation’ is in fact the traditional ‘transaction-basedintegration’ process, which demonstrates that the

transaction-‘transaction-based integration’ is embedded in erative integration In other words, the semiotics-basedMSI – generative integration represents a new para-digm of the computer integrated manufacturing (CIM)and a step ‘ahead’ from the ‘traditional’ field of CIMparadigm enriching the ‘traditional’ paradigm withnew instruments, AND moving the location of theintegration field from the machines to humans,supporting, and promoting, new MS paradigms-based

gen-in concepts such as learngen-ing organisations, complexity,and other emergent

It is expected that this shift will provide, at least, aqualitative shift towards higher potential for fullintegration of manufacturing system

Table 4 Integration instruments for semiotic-based

open organisational architectures, open-architecture

organisations and enterprises

Figure 8 Semiotics-based MS integration process lifecycle

basic model (Putnik et al 2005)

A framework for semiotics-based MSI is presented

in Table 5 The header of the table represents the

determinants of the framework, and also serves as the

criteria or a reference, for 1) guiding the development

of the semiotics-based MSI instruments, and 2) for

evaluating a concrete MSI solution’s compliance with

the semiotics-based MSI paradigm The second row of

the table – ‘Pragmatic’ – presents the ‘values’ of the

semiotics-based MSI framework determinants

Addi-tionally, the next two rows of the table – ‘semantic’ and

‘syntactic’ (coloured grey) – represent the values of the

same integration paradigm determinants but for the

‘traditional’ semantic, or transaction-based integration

for the purpose of comparison and distinction The last

two rows of the table – ‘empirical’ and ‘physical’

(coloured dark grey) – are not of concern for the

semiotics but they are present in any paradigm as they

represent underlying physical implementations and

processes

4 Computers in semiotics-informed (or based)

integration of manufacturing system

From the semiotics point of view, there are three

dimensions or roles of the computer usage in

semiotic-based MSI These are: 1) the computer

itself, as a nonverbal sign, 2) the computer as

generators of signs, and 3) the computer as transmitter

of signs

4.1 The computer itself as a nonverbal sign

This role includes actually noncomputing functions of

the computer, e.g computer as a part of the working

space, e.g of the meeting place, to enhance the

pragmatic effects on the interlocutors In this case,

computers are symbols/signs by themselves For

example, in Figure 9 a meeting room (in a research

laboratory) is shown with two large 320 computer

screens with the exclusive role to substitute the cork

boards for exactly the same purpose Functionally, and

of course technologically, there is a difference between

two ‘technologies’, but the exposed content on the

screens is the same as on the cork board It means

that these screens are not used for traditional

multimedia presentations, for which purpose there

are other screens in the room (not visible in the

picture) However, there is an explicit intention of

transmitting the message of the laboratory as an

advanced working environment The success of the

message was confirmed through a number of

com-ments by the visitors

The role and analysis of computers as the

nonverbal symbols and/or signs by themselves will be

referred as semiotics of computers

4.2 The computer as generators of signsThis role refers to the computer as a ‘semioticmachine’ However, for this aspect the discussion inNake and Grabowski (2001) should be considered

As computers by themselves are incapable ofgenerating signs, i.e incapable of performing prag-matic function of communication, their interactionbeing reduced only to information transaction, interms of semiotics and in the context of MSI, this casewill be called nonsemiotic or pseudo-semiotic MSI(actually, more formally it is a transaction-basedMSI) Figure 10 presents elementary architectures forthis case These are architectures that ‘connect’ twocomputers, Figure 10(a), and computer and human,Figure 10(d)

In all architectures only semantic and syntax aretransmitted, without pragmatics In the case of human

on one side, although the human is susceptible topragmatics, there is no pragmatics (because there is nointerlocutor) However, humans may perform process

of interpretation as a cognitive process (Althoughinterpretation is a part of pragmatics, in this case not allconditions are fulfilled to have true pragmatics effects.)The architecture presented in Figure 10(a) could besimply modularly assembled in other nonelementaryarchitectures, of which the special interest for MSI hasthe architectures presented in Figure 10(b) and (c).When the machines (computers) in the middle have theroles of translators (compilers), it is recognised asthe well-known ‘federated’ integration architecture(Figure 10(b)) (e.g P2P integration), and so-called

‘open’ integration architecture, based on a format’ data file (Figure 10(c)) (For the sake ofcompleteness, the architecture in Figure 10(a) is theelementary architecture of the ‘direct’, or ‘proprietary’integration)

‘neutral-Actually, the integration architectures in Figure10(a), (b) and (c) are the architectures of the

‘traditional’ information transaction-based MSI digm In this way, the transaction-based MSI para-digm could be interpreted as a special case of thesemiotic-based MSI

para-4.3 The computer as a transmitter of signsThis role refers to the computer as a supporter/mechanism/physical_channel/medium/ for commu-nication, i.e for transmission, of signs and signprocesses – semiosis, between humans In this casethere is a pragmatic function and this case representsuse of the computer (machine) in true semiotic-basedintegration Concerning the ‘computer integratedmanufacturing’, semiotic-based integration mightseem as ‘noncomputer’ integration, because of the

focus on human communication This is true when a

computer is not used, as in the elementary

archi-tectural patterns for ‘noncomputer’, semiotic-based

MSI (Figure 11(a)) However, when the computer is

used for communication, for transmitting signs

between the interlocutors, whether in elementary

architectural patterns for semiotic-based MSI

pre-sented in Figures 11(b) and (c), then it is justified to

call such communication architectures

‘semiotic-based ‘computer’ integration’ with federated

archi-tecture and ‘direct’ communication archiarchi-tecture, and

‘semiotic-based ‘computer’ integration’ with ‘open’

architecture, based on ‘neutral-format’ data file and

‘virtual’ communication architecture, respectively

The federated and direct communication

archi-tecture (DCA), Figure 11(b), usually uses tools such

as videoconferencing (VC), computer-supported

co-operative work (CSCW) tools (VC tools actually

might be classified as a CSCW tool), virtual reality

(VR) and especially immersive virtual reality (IVR),

other virtual environment (VEnv) tools, such as

‘Metaverse’ tools, and internet through Pragmatic

Web, and similar The ‘open’ architecture, based on

‘neutral-format’ data file and virtual communication

architecture (VCA) (Figure 11(c)), usually uses tools

such as virtual reality (VR), immersive virtual reality

(IVR), other virtual environment (VEnv) tools, such

as Metaverse tools, and internet through Pragmatic

Web, which might be combined with CSCW tools,

because the role of the ‘open’/VCA architectures is to

‘hide’, i.e to virtualise the communication partners

(More of DCA and VCA architectures, as well as

some results on their technical performance (but not

on the semiotic results – which is for future work) are

presented in Putnik et al (2008))

Further justification for calling the semiotic-based

MSI architectures ‘semiotic-based ‘computer’

integra-tion’, or ‘semiotic-based ‘computer’ integrated’

sys-tems is in the fact that the signatures of these

architectures and the architectures of the ‘information

transaction’-based SMI (the ‘traditional’ CIM) are in

fact equal in the ‘middle part of the computerintegration ‘chain’ The difference is at the ends ofthe ‘chain’ While in the ‘traditional’ computerintegration at the ends of the chain the messagesender and receiver are machines, i.e computers(algorithms), meaning that the location of the(semantic only) interpreter of the messages is in themachines–computers (algorithms), in the semiotic-based integration at the ends of the chain the messagesender and receiver are humans, meaning that thelocation of the (now pragmatic) interpreter of themessages is in the humans – compare e.g Figures10(a), (b) and (c) and Figures 11(b) and (c)

Resuming:

(1) It means that the paradigmatic distinctionbetween the ‘traditional’ transaction-based in-tegration and the semiotic-based integration is

in the location of the interpreter, whether theinterpreter is in machine (semantics only) or inhuman (pragmatic interpretation capability)respectively

(2) When the computer (machine) is used as thetransmitter of signs, but NOT as a sign generator,then the semiotic-based MSI will be called

‘semiotic-based ‘computer’ integration of MS’ or

‘semiotic-based Computer MSI’ or ‘semiotic-basedCIM’, or similar (This syntagm could beconsidered as a regular one since the commu-nication between humans in modern, and future,technical systems is, or will be, practicallyimpossible without use of computers)

5 An experimental platform for research anddevelopment of the semiotics-based integration5.1 ‘Functional requirements’ for IT tools supportingthe semiotics-based integration

Importance of the semiotics-based MSI grows withgrowth of complexity of manufacturing systems

Figure 9 Computer itself as a nonverbal sign: enhancing the pragmatic effects of communication within the meeting place

Growing complexity implies/means growing

hetero-geneity, relations’ dynamics, nonlinearity and ‘chaos’

For example in a ‘supply-chain’ type network with

several partners, e.g Figure 13(a), one can expect

effective unification and integration based on

seman-tics, i.e transaction-based integration However, as

much as manufacturing systems move towards more

complex networks, e.g Figure 13(b) and (c), up to

hyper-complex structures, such as ubiquitous

manu-facturing systems (UMS), Figure 13(d), it is less and

less realistic to expect effective semantics, i.e

transac-tion, based MSI Thus, the only instrument that gives

us the potential for effective, and at the same time

efficient, MSI, is semiotics, i.e pragmatics (Note: The

ubiquitous manufacturing systems (UMS) as a

hyper-complex network, that maps the ubiquitous computing

systems (UCS) on its architecture, is one of the UMS

paradigms The second UMS paradigm uses UCS as

the underlying operating system (Putnik 2010) For the

second UMS paradigm, see also Suh et al (2008) and

Suho et al (2009).)

Building the effective, and at the same timeefficient, MSI depends largely on a new generation oftools, which do not depend too much on new IT butrather on conceptual solutions oriented to providefunctionalities for enabling live communication andpragmatics, i.e pragmatic effects

Therefore, while the ‘traditional’ tools are mainlyoriented to provide, defined as, the ‘semantic’ level-based communication ‘channels’ (for short ‘semantic’communication channels), the new generation ofMSI tools should provide, i.e should add, thefunctionalities for the ‘pragmatics’ level-based com-munication ‘channels’ (for short ‘pragmatic’ commu-nication channels) Obviously, these new ‘pragmatic’communication channels should enable overcomingthe barriers of the space and time, i.e employing adiversity of tools such as, already mentioned,videoconferencing (VC), computer supported coop-erative work (CSCW) tools, virtual reality (VR) andespecially immersive virtual reality (IVR), othervirtual environment (VEnv) tools, such as

Figure 10 Elementary architectural patterns for nonsemiotic, pseudo-semiotic MSI, or transaction-based MSI: (a) proprietaryarchitecture; (b) federated architecture; (c) ‘open’ architecture, based on ‘neutral-format’ data file; (d) hybrid architecturebetween machine and human

‘Metaverse’ tools, and Internet through Pragmatic

Web, for overcoming the ‘space barrier’, and e.g

ubiquity of resources through UMS, for overcoming

the ‘time barrier’

Also, both the ‘semantic’ communication channelsand ‘pragmatic’ communication channels should enablethe generative integration process life-cycle, i.e the phases

of integration synthesis, operation and termination

Figure 11 Elementary architectural patterns for semiotic-based MSI: (a) semiotic-based ‘noncomputer’ integration; (b)semiotic-based ‘computer’ integration with federated architecture and ‘direct’ communication architecture; (c) semiotic-based

‘computer’ integration with ‘open’ architecture, based on ‘neutral-format’ data file and ‘virtual’ communication architecture

5.2 The semiotic-based MSI oriented UMS cells – the

elementary and laboratory unit

The elementary experimental set-up for research of

semiotic-based MSI is the semiotic-based integration

oriented UMS cell(for short UMS cell) (‘UMS’ stands

for ‘ubiquitous manufacturing system’, see the next

section) It is a laboratory unit, carefully conceived to

provide rich ‘pragmatic’ communication channels for

the remote operation of machine tools (MT) inherent

to the UMS, integrating the client with the remote MS

cell local operator (Figure 12(a)) but, at the same time,

to be an elementary set-up for other MSI domains,

such as CAD, CAPP, ERP, when abstracting machine

tool, integrating remote partners in these application

domains

The ‘pragmatic’ communication channels are

implemented basically through installation of

multiple video-cameras and VC systems, Figure

12(b) The purpose of video cameras is not to be

sensors for eventual algorithmic ‘image processing’

(‘pattern recognition’) but as the instruments, in the

context of pragmatics, for enhancing interpretations

of the environment in order to provide a richer

content for the communication between the client

and the machine operator Obviously, the VC system

is an obligatory component, which could/should be

embedded in a richer ‘client-server (local cell

operator, or partner)’ interface (see below)

Concerning the enhancement of the perception and

positioning of the cameras, it is possible to conceive a

robotic system to manipulate cameras instead of havingmultiple cameras, which could be a better solution.From the semiotics-based MSI perspective the problem

of robot control is irrelevant However, the way ofmanipulating the camera in the context of the percep-tion and cognition enhancement is a relevant issue

5.3 Ubiquitous manufacturing system (UMS)experimental platform as environment for research anddevelopment of the semiotics-based MSI

The above laboratory unit UMS cell is a basic blockfor building larger manufacturing systems for research

of the semiotics-based MSI

The experimental set-up on a system’s ‘higher level’

is represented by a network in which nodes arelaboratory units – UMS cells In its physical imple-mentation, in the first phase the experimental platformconsists of 4–5 physical nodes (Figure 13(a)) On

‘lower objective levels’, the experimentations over thisnetwork should positively identify and evaluate thefactors and performance for simulation of the semio-tic-based MSI in the conditions of large hyper-complexmanufacturing networks such as ubiquitous manufac-turing system (UMS) (Figure 13(d)) that might consist

of millions of nodes, with inherent manifestations ofdynamics, nonlinearity, uncertainty and ‘chaos’ Theresearch of special interest is the research of integrationinstruments for semiotic-based integration referred to

in Table 4, for which the UMS concept is a good andnatural environment

Figure 12 Semiotic-based integration oriented UMS cells: (a) an informal specification, (b) physical implementation ofmultiple video-cameras for enhancing interpretations of the working area for communication between the client and theoperator

On ‘higher objective levels’ the main research

objective is validation of the scientific thesis on

effectiveness and efficiency of the semiotic-based MSI

in comparison with the ‘traditional’ semantics, i.e

‘transaction’ based, integration paradigm

Concerning the physical network as the

experi-mental set-up, its expansion will be relatively limited,

e.g Figure 13(b) and (c), and might include not only

laboratory units but the real-life industrial units as well

as virtual units

Ubiquitous manufacturing system (UMS)

experi-mental platform, besides the network of UMS Cells and

clients, integrates as well additional functionalities such

as brokering services and services inherent to the

‘meta-organisational’ environment (‘Market of Resources’

is one of the meta-organisational environments)

Figure 14 presents an informal specification of the

semiotics-based MSI-oriented UMS

The important feature of the experimental

plat-form is that the interfaces between the clients,

brokers and UMS cells (machining resources), in

order to enable the experiments inherent to the

semiotic-based MSI, should provide both ‘pragmatic’

communication channels and ‘semantic’ communication

channels

Both the ‘semantic’ communication channels and

the ‘pragmatic’ communication channels enable the

generative integration process life-cycle, i.e the phases

of integration synthesis, operation and termination It

is important to note that use of the ‘semantic’

communication channels only provides ‘traditional’

semantics-based integration, for example the

integra-tion based on the ‘standard data exchange formats’ In

that sense, the experimental platform enables

experi-mentations with both integration paradigms and, at

the same time, demonstrates that the ‘traditional’

semantics, or transaction-based integration paradigm

is embedded in the generative integration paradigm

(representing its special case)

Figure 15 presents implemented interfaces betweenthe Clients, Brokers and UMS cells that satisfy therequired features

5.3.1 Experimental set-up for research of somenonverbal factors

Additionally, it is also necessary to research theinfluence of nonverbal factors on semiotics-basedMSI For example different screen types and sizes areconsidered in order to evaluate their impact onpragmatic effects on MSI The examples of ‘desktop’size screens and very large screens, that are to provide

a virtual-presence-like environment, are presented inFigure 16 Also, future developments will considerimmersive VR and ‘Metaverse’ environments, as wellcollaborative architectures as the nonverbal instru-ments of semiotic-based MSI

The research of the influence of different screentypes and sizes and other environments should not beconfused with the research of ergonomic effects ofthese instruments, but exclusively their influence in thesemiotic/pragmatic context (of course, on some ‘re-mote level’ or context it could be considered influence

of ergonomics on semiotic effects )

6 ConclusionsThe discussion above leads us to a number ofconclusions

The new manufacturing systems paradigms thataim for sustainability in a dynamic, nonlinear,uncertain and ‘chaotic’ environment, should be cap-able of providing communication environments forenabling the full semiotic/pragmatic functions, i.e toemploy semiotics as the instrument, i.e to enable fullpragmatic functions, and consequently new semantics,that effectively represent the now hidden patterns ofbehaviour (that makes us understand the environment

Figure 13 (a) Initial experimental platform for the semiotic-based MSI, evolving (b) and (c), towards (d) ubiquitousmanufacturing system simulation network (Also, figurative presentation of virtual enterprise evolution: from (a) conservative,minimal network domain, towards (d) UMS ubiquitous network domain)

as ‘chaotic’) whether within the system/organisation or

of the environment

Concerning the possible critics, the semiotics-based

MSI could be questioned by claiming that the

‘traditional’ transaction-based integration approach

has proved efficient and that the semiotics-based MSI

did not (this is true from that standpoint) However,

the defenders of the ‘traditional’ approach to

integra-tion should take into account the dichotomy between

individual-community and society (see Section 3 and

Figure 7) Semiotics has proved that emphasising the

individual-community dimension, inherent to the

semiotics-based MSI, vs emphasising the society

dimension, inherent to the ‘traditional’ approach (e.g

the objective of omnipresence of international

stan-dards), directly enrich and ‘impoverish’, respectively,

the potential for innovation and novelty of the MSI

and the potential for MSI effectiveness or efficiency

Thus, the choice of the semiotics-based MSI or not, in

fact depends on choice of efficiency or effectiveness,

linearity or nonlinearity, unification or creativity,

controlling or freedom to create The choice, of course,

is in the context of expectation that the choice, whether

semiotics-based MSI or other, will bring us socialbenefits in whatever dimension we define

There are some other interesting questions.Probably one of the most interesting questions isdoes the semiotics-based MSI imply that the ‘tradi-tional’ approach to the MSI, i.e the ‘information

‘transaction’ MSI paradigm will disappear? Theanswer is no The ‘transaction-based integration’ isactually embedded in the semiotics-based MSI –generative integration, and for sure will have theexclusive role in specific scenarios (the discussion onthis question is already presented above)

Another interesting question might be: Do we

‘open the door’ of the traditional engineering discipline

to the psychologists, linguists, sociologists, psychiatrists, philosophers, etc., i.e the human sciencesresearchers? The answer is yes It is not realistic toexpect that the traditional engineering disciplines havethe responses to such growing complexity of our lives

neuro-as we are presenting today It would be irresponsible toabdicate in advance of the great knowledge created,and in the course of creation, in other disciplines thatfor sure can help in further development of our

Figure 14 Ubiquitous manufacturing system (UMS) experimental platform – an informal specification

engineering discipline It would be a ‘naı¨ve’ approach

to claim that human sciences are without a role once

the human relations are in fact recognised as the

legitimate MS domain, e.g., MSI domains of relation

with customers, collaboration, etc

Concerning future work, there are three large

domains, or three dimensions:

(1) The dimension/domain of the semiotic

instru-ments for MSI development and of the factors

of the semiotics-based MSI;

(2) The dimension/domain of identification of the

applicability domains, referring to various MSI

fields, such as CAD, CAPP, CAM, ERP, etc In

other words, the question is where are thelimits/frontiers, which are the barriers, whichare the ‘trade-offs’, where are the ‘break-even’points, for applications of the semiotic-basedMSI and/or of the traditional ‘transaction-based’ MSI However, besides the identification

of the application domains and their frontiers,there is an even more interesting question of thedynamicsof these frontiers

(3) The semiotic-based MSI meta-theoretical issue.Concerning this issue, the semiotics could beseen as having a double relation to the science

of manufacturing system integration (MSI)(paraphrasing Morris (1938) for the purpose

Figure 15 Interfaces between the Clients, Brokers and UMS Cells with both ‘pragmatic’ and ‘semantic’ communicationchannels

of MSI): it is both an instrument of MSI and a

meta-model of the science of MSI The semiotics

as the instrument of MSI means that the MSI as

a discipline uses the models, instruments and

procedures of the semiotics as an independent

science for the objective of improvement of the

MSI mechanisms, tools and processes The

semiotics as the meta-model, or a meta-theory,

of the science of MSI means that the semiotics of

the science, or discipline, of MSI is investigated

From this perspective the semiotics might be

used as an explanatory and unification

frame-work, or as an ‘organon’, of the MSI science –

and the sciences in general – (Morris 1938),

‘since every science makes use of and expresses

its results in terms of signs’

Finally, the above presentation of the semiotic

frame-work for manufacturing systems integration and its

generative integration model might conclude with a

comment on the semiotics-based manufacturing systems

integration (S-MSI) science by paraphrasing Saussure

(‘ Since it [Semiology] does not yet exist, one cannot

say for certain that it will exist But it has a right to exist,

a place ready for it in advance.’ (Saussure 1916)):

Since the semiotics-based manufacturing systems

integration (S-MSI) science does not yet exist,

one cannot say for certain that it will exist But it

has a right to exist, a place ready for it in

advance

Acknowledgements

The authors wish to acknowledge the support of: 1) The

Foundation for Science and Technology – FCT, Project

PTDC/EME-GIN/102143/2008, ‘Ubiquitous oriented

embedded systems for globally distributed factories ofmanufacturing enterprises’, 2) EUREKA, Project E!4177-Pro-Factory UES

Filipe, J., 2004 The organizational semiotics normativeparadigm In: L.M Camarinha-Matos and H Afsarma-nesh, eds Collaborative networked organizations Dor-drecht: Kluwer

Guiraud, Pierre, 1975 Semiology London: Routledge andKegan Paul Ltd (first published as: La Se´miologie,Presses Universitaires de France, 1971)

InterSystems, 2004 Ensemble, InterSystems Corporation.Available online http://www.intersystems.com/index.html

Katambwe, J.M and Taylor, J.R., 2006 Modes of tional integration In: F Cooren, J.R Taylor, and J.v.E.Every, eds Communication as Organizing Mahwah, NJ:Lawrence Erlbaum Associates, 55–77

organiza-Libes, D., Flater, D., Wallace, E., Steves, M., Feeney, A.B.,and Barkmeyer, E., 2004 The challenges of automatedmethods for integrating systems In: SE 2004 — IASTEDInternational Conference on Software Engineering Avail-able online http://www.mel.nist.gov/msidlibrary/doc/challenge.pdf

Liu, K., 2000 Semiotics in information systems engineering.New York: Cambridge University Press

Moor, A and Weigand, H., 2002 Towards a semioticcommunications quality model In: K Liu, R.J Clarke,P.B Andersen, R.K Stamper, and E.-S Abou-Zeid, eds.Organizational semiotics: Evolving a science of informa-tion systems Dordrecht: Kluwer, 275–285

Morris, C., 1938 Foundations of the theory of signs,International Encyclopedia of Unified Science, Vol 1,

No 2, 1-2, Chicago: University of Chicago Press.Figure 16 Different screen types and sizes for analysing of their impact on pragmatic effects during the communication

Morris, C., 1946 Signs, language and behavior New York:

Prentice-Hall Reprinted, New York: George Braziller,

1955 Reprinted in Charles Morris, Writings on the

General Theory of Signs(The Hague: Mouton, 1971), 73–

397

Nake, F and Grabowski, S., 2001 Human–computer

interaction viewed as pseudo-communication

Knowl-edge-Based Systems, 14 (2001), 441–447

Nell, J.G., 1998 Enterprise representation: An analysis of

standards issues Available online http://www.mel.nist

gov/msidlibrary/doc/jimnell95.pdf

Newman, S.T., et al., 2008 Strategic advantages of

interoperability for global manufacturing using CNC

technology Robotics and Computer-Integrated

Manufac-turing, 24, 699–708

Putnik, G.D., Cunha, M.M., Sousa, R., and A´vila, P., 2005

Virtual enterprise integration: challenges of a new

paradigm In: G Putnik and M.M Cunha, eds Virtual

enterprise integration: technological and organizational

perspectives Hershey, PA, USA: IDEA Group

Publish-ing, 1–30

Putnik, G., Gonc¸alves, P., Sluga, A., and Cunha, M.M.,

2008 Virtual environments for dynamically

reconfigur-able Concurrent/Collaborative Engineering ‘virtual’

team CIRP Annals, 57 (1), 171–174

Putnik, G., 2010 Ubiquitous manufacturing systems vs

ubiquitous manufacturing systems: Two paradigms In:

Proceedings of the CIRP ICME010 International

Con-ference, Capri, Italy

Recanati, F., 2004 Pragmatics and semantics In: L.R Horn

and G Ward, eds The Handbook of Pragmatics Oxford,

UK: Blackwell Publishing

Saludadez, J.A and Taylor, J.R., 2006 The structuring of

collaborative research networks in the stories

research-ers tell In: F Cooren, J.R Taylor, and E.J van Every,

eds Communication as Organizing Mahwah, NJ:

Lawrence Erlbaum Associates, 37–54

Saussure, Ferdinand de, 1916 Course in general linguistics.In: C Bally and A Sechehaye, eds Editions Payot andRivages, Paris Trans Roy Harris La Salle, IL: OpenCourt, 1986

Stamper, R., 1999 Organisational semiotics: Informaticswithout the computer? In: K Liu, R Clarke, and P.B.Andersen, eds Information, organisation, and technology:studies in organisational semiotics Dordrecht: Kluwer.Stamper, R., 2000 New directions for systems analysis anddesign In: J Filipe, ed Enterprise information systems.Dordrecht: Kluwer, 14–39

Stamper, R., Liu, K., Hafkamp, M., and Ades, Y., 2000.Understanding the roles of signs and norms in organiza-tions – a semiotic approach to information systemsdesign Behaviour and Information Technology, 19 (1), 15–27

Stamper, R., 2009 A Pragmatic Web Workshop, tion on I-SEMANTICS009, Graz, Austria

presenta-Suh, S.-H., Shin, S.-J., Yoon, J.-S., and Um, J.-M., 2008.UbiDM: A new paradigm for product design andmanufacturing via ubiquitous computing technology.International Journal of Computer Integrated Manufac-turing, 21 (5), 540–549

Suho, J., Hur, S.M., and Suh, S.-H., 2009 A conceptualframework for computer-aided ubiquitous system en-gineering: architecture and prototype InternationalJournal of Computer Integrated Manufacturing, 22 (7),671–685

The Standish Group Int, 2005 Chaos Rising: A ChaosExecutive Commentary Report, The Standish GroupInt., West Yarmouth, MA

An exploration of the integrative function of dialogue in manufacturing

Frans M van Eijnattenaand Goran D Putnikb*a

Eindhoven University of Technology, Department of Industrial Engineering and Innovation Sciences, Human PerformanceManagement Group, Eindhoven, The Netherlands;bUniversity of Minho, Campus of Azurem, Department of Production and

Systems Engineering, Guimara˜es, Portugal(Received 7 December 2009; final version received 7 March 2010)This paper is about the roles of dialogue as a generative mechanism in manufacturing system integration Itadvocates the integrative power of dialogue in the design and operation of manufacturing systems Dialogicalconversation is a powerful tool to create a learning organisation: it might be a valuable instrument to implement inmanufacturing companies that want to renew themselves, in order to help find assumptions behind the thinking, todiscover hidden patterns that block effective behaviours in all sorts of work contexts, and to develop openness,responsibility, and trust in both individual employees and work teams This paper re-analyses five studies about theintroduction of dialogue as a new mode of communication among managers, engineers, and workers in a Dutchmanufacturing company It explores the integrative functions, action and learning triggers, and collective learningprocesses that dialogue had in this particular case

Keywords: dialogue; multilogue; learning; integration; manufacturing systems

1 Introduction

In a contemporary manufacturing system, effective

human communication is vital, not only for its

operation, but also for its design and any further

developments and changes According to Bainbridge

(1983) an irony of automation is that the role of the

human factor will become more important rather than

less Therefore, further development of human

re-sources is crucial This paper explores the benefits that

dialogical conversation may have for manufacturing

system integration Important topics in this context are

(Tsoukas 2009): the fostering of new ideas;

under-standing of each other’s individual positions; respecting

each other’s personal points of view; and a balanced

development of the potentials of all professionals

involved, i.e., managers, engineers, and workers

This paper takes semiotics as its main point of

departure, and focuses on information processing at

the pragmatics level, i.e., the creation and

interpreta-tion of meaning by different professionals in a

manufacturing system The central thesis of this paper

can be stated as: the application of a very particular

mode of conversation – i.e., dialogical communication

between managers, engineers, and workers – serves as a

generative mechanism in manufacturing systems

inte-gration Engaging in dialogue will result in more

mutual understanding of the meanings of both actions

and interactions of the various professionals Also, it

will increase virtues like trust, appreciation, affection,and respect (Burbules 1993)

In this paper, this thesis is proved throughexecuting a secondary analysis of five empiricalresearch papers, most of which actually were pre-sented at academic conferences, and subsequentlywere published in various peer-reviewed journals oredited books, in the period 2001–2005 Especially,examples are sought of the integrative functions,action and learning triggers, and collective-learningprocesses that dialogue might effectuate in a manu-facturing system

2 Dialogue and discussion: two types of conversation

In this section, two basic conversation modes ofhuman interaction will be discussed: discussion anddialogue/multilogue As for semiotics, it is stressed thatboth the semantics (concepts underlying the basicmodes of human conversation) and the pragmatics(their functions and effects) are being discussed Theprimary focus will be on organisational learning (VanEijnatten and Putnik 2004): ‘by looking at the way thatpeople jointly construct maps’ (Argyris and Scho¨n

1978, p 19) in an organisation: ‘where peoplecontinually expand their capacity to create the resultsthey truly desire, where new and expansive patterns ofthinking are nurtured, where collective aspirations are

*Corresponding author Email: putnikgd@dps.uminho.pt

Vol 23, Nos 8–9, August–September 2010, 710–719

ISSN 0951-192X print/ISSN 1362-3052 online

Ó 2010 Taylor & Francis

DOI: 10.1080/09511921003767571

set free, and where people are continually learning to

see the whole together’ (Senge 1990, p 3)

Discussion or dispute is by far the most frequently

used type of conversation between managers, engineers,

and workers It can be described as a dominant process

of human interaction that is aimed at the reduction of

diversity in order to enable decision making The

method of discussion is used to efficiently arrive at one

single point of view from a multitude of different initial

positions A skilful discussion is based on arguments

Dialogue is another basic type of conversation

between managers and workers, which until now is not

very well known, and is much less frequently used in

organisations Dialogue can be described as a process

of human interaction in which the aim is to engage in

‘shared exploration towards greater understanding,

connection, or possibility’ (Co-Intelligence Institute

2009) A dialogue is a method of thinking together

(Gerard and Ellinor 1999, Isaacs 1999); a tool for

building team learning (Senge 1990) to overcome

individual and social barriers for sharing meaning,

values, and understanding (Slotte 2006) The ontology

of the dialogue is systemic (Buber 1947) Unlike

conventional means of conversing in modern life, the

objective of dialogue is to discover flaws and faulty

assumptions in one’s own thinking so that they might

be corrected (Bohm 1990) Being aware of these

assumptions may trigger individuals and groups to

change them, which may unleash the organisation’spotential for development and transformation Dialo-gue may promote collective learning and self-organisa-tion, the creation of a common culture, a shared vision,self-distantiation, and the development of new knowl-edge (Tsoukas 2009)

Multilogue is a dialogue process that takes placebetween groups of individuals Multilogue was in-vented in Russia back in the previous century as asimulation-game technique that creates a temporarylearning organisation According to Zaitsev (1998),and Zaitsev and Artemova (1998) it is a powerfulapproach to learning According to Hoogerwerf(1998), Van Eijnatten and Hoogerwerf (1999), andHoogerwerf and Poorthuis (2002), multilogues incompanies should include employees from all levelswho are highly involved in the theme of the conversa-tion, and who are willing to invest their time and energy.Dialogue is different from discussion in a number

of aspects, see Table 1

Discussion is a convergent deliberation process thatselects one single, final, definitive point of view from acollection of two or more individual positions.Reaching agreement is the basic goal of decisionmaking The process of discussion is aiming at exactlythat It may be characterised as a battle of argumentsthat are fired back and forth at full speed, by means ofwhich people defend their own positions vigorously, in

Table 1 Contrasts between discussion and dialogue

Basic conversation modes of human interaction

fragmented debate; either/or thinking

Conversation between two or more persons aboutthe context in which problems arise

Type of human interaction

Major goals Thinking alone, debating, fighting, beating

down, winning the argument, ‘ping pongmatch’

Thinking together, trying to better understand,dis-identify, unfolding meaning and newinsights

Resolving differences definitely Reflecting, exploring, rethinkingGaining agreement on one single meaning Creating shared meaning among many,

reordering knowledgeSeeing distinctions between the parts Seeing the connection between the partsTypical behaviours Persuading others, telling how, selling your

own point of view

Listening together without resistance, expressingprivate views

Imposing/‘making’ a decision, choosing one

order to persuade others To see distinctions,

discus-sion tends to analyse, and by doing so, decomposes

problems into smaller and smaller issues

Dialogue is a divergent deliberation process that

broadens the number and scopes of potential points of

view (Tsoukas 1999) Deepening understanding and

creating shared meaning are achieved by a slow,

nonjudgemental, open-ended process of personal

reflection, exchange of views, and inquiry into own

assumptions (Bohm 1990, Gerard and Teurfs 1995,

Ellinor and Gerard 1998, Gerard and Ellinor 1999,

Isaacs 1999) Investigation and integration of different

aspects is strived for, in order to arrive at re-ordered

knowledge and new insights As Table 1 shows,

discussion and dialogue are each other’s opposites

3 Methods

The research method basically is a secondary analysis:

i.e., five research papers were re-analysed that present

both the process and the results of the introduction of

dialogue in a Dutch manufacturing firm, over a period

of eight years (1999–2006) These studies are: Van

Eijnatten et al (2001), Van Eijnatten and Van Galen

(2002), Van Eijnatten and Van Galen (2003), Van

Eijnatten and Van Galen (2005), and Chen and Drost

(2006) The goal of the secondary analysis was to

assess the role of dialogue as a generative mechanism

in manufacturing system integration Some as yet

unpublished analyses of data were also added The

research studies used multiple methodologies: various

research designs (both cross-sectional and

longitudi-nal); various data-gathering methods (both

observa-tions, questionnaires, and administrative company

measures); and various data-analysis methods (both

qualitative and quantitative), see Table 1

The secondary analysis was structured by means ofthe following five themes:

(1) Company details, concise change history, andproblem statement

(2) The context of the dialogue project

(3) The actual training and practice of dialogue.(4) Short- and long-term effects of the training ofdialogue

(5) The roles of dialogue in manufacturing systemintegration

Since all research papers contain some informationabout all the mentioned themes with different em-phases, a summary will be compiled For themes 4 and

5, selected results for each individual paper will bepresented separately

4 MaterialThe company is a manufacturer of tailor-madecomplex processing systems At the start of thedialogue project, the company’s product portfoliocovered 250 functions and 2,500 variants Its marketshare at that time was 60% At the start of the dialogueproject the company employed 500 people in TheNetherlands (750 worldwide)

The company went through a whole series oforganisational developments, which started with theintroduction of self-managed work teams in PartsProduction in the late Eighties, which further diffusedinto Assembly, Stock & Shipping, Marketing, Sales &Installation, and R&D The company also restructuredits functional departments into the following fiveprocess sectors: Production & Procurement; Innova-tion; Parts & Services; Sales & Installation; and HRM,

Table 2 A specification of re-analysed research papers by topic

Van Eijnatten,Dijkstra andVan Galen 2001

Van Eijnatten andVan Galen 2002

Van Eijnatten andVan Galen 2003

Van Eijnatten andVan Galen 2005

Chen andDrost 2006General

Covered Period of the Actual

Research

Specific

Quantitative Evaluation of the

Finance & IT The company used a step-by-step

approach to change: the structural renewal of the

firm into a full team-based organisation took nearly

ten years Some associated performance improvements

were: Major reductions in throughput times and

rejects, improved set-up times and task times in

production, a make-to-order production regime, and

higher involvement and commitment of the company’s

personnel

Following the structural change, the company

embarked on cultural renewal, at the end of the 20th

century The reason for this was the general

observa-tion that personal interacobserva-tions within the company

were anything but smooth Managers complained

about employees not taking enough personal

initia-tives, and employees criticised management for not

listening to them What everybody experienced was

that important decisions were ‘pre-cooked’ by a small

elite group of dominant professionals, who pushed

their arguments in meetings, leaving others no further

room than just to follow Also, the use of praise as an

appreciation of good work was seldom used, and

major disagreements could easily develop into

pro-longed, disrupted personal relationships

In order to initiate cultural transformation,

management, workers, researchers and consultants

collaborated in an action research context to develop

and plan a series of process interventions, see Figure 1

The actual dialogue project was ‘embedded’ in a

cultural-change programme, in which managers and

workers went through a series of workshops and

conferences This programme started with an

intro-ductory workshop about complexity, chaos, and

leap-like changes in thinking This was followed by a full

day of dialogue training, given by an external

consultant in a conference resort to already existing

company project teams, consisting of both managers

and engineers Dialogue was further practiced ‘on the

job’ in actual meetings of these groups supported by a

facilitator After two to four months, off-the-job role

training was administered to the same project teams

(Lynch and Kordis 1988) After six months, a vision

conference was held, which lasted a full day, in which

four dialogue project teams dialogued/multiloguedabout personal visions The programme ended with arehearsal of the introductory workshop, approximatelyone year following its start The programme was firstadministered to 24 higher and middle managers in theperiod 1999–2000, and subsequently to 81 middle andlower managers and technical staff personnel in 2000–

2001, 160 employees in 2001–2002, and 65 employees

in 2002–2003 The vision conference was only nised for the higher and middle management Thecultural-change programme concluded with a confer-ence for the whole company This conference wasspontaneously organised by a group of employees,who gathered together as volunteers for its prepara-tion, in February 2003

orga-5 Intervention, the actual training of dialogueDialogue was trained in groups of 7–10 persons Forthe management, these groups were composed ofexisting (technical) project teams in the company.They were given half a day of instructions in the basicconcepts of dialogue, i.e., contrasts between discussionand dialogue, hearing versus listening, assumptionidentification, suspension of judgement, inquiry, andreflection, see Gerard and Ellinor (1999) Also, theylearned about the ‘ladder of abstraction’ (analysis ofthought process), ‘undiscussables’ and ‘defensive rou-tines’ (group errors and behaviours), ‘collusions’(secret agreements), ‘time out of time’ (systematicallyanalysing a conversation), and ‘check-in/check-out’(communicating personal expectations, intentions andevaluations at the start and end of a meeting), see VanEijnatten and Van Galen (2002) They also receivedhalf a day of intensive, guided dialogue practice Thewhole dialogue training session was organised off thejob, in a conference resort outside the company Allconversations were focused on the past, present, andfuture of the company

During the following months, the project teamswere supposed to use their regular meetings (onceevery three or four weeks) to practice dialogue in realcompany life, supported by a facilitator

Figure 1 The context of the dialogue project Sources: Van Eijnatten, Dijkstra and Van Galen (2001); Van Eijnatten and VanGalen (2002); Van Eijnatten and Van Galen (2003); Van Eijnatten and Van Galen (2005)

The job-role training was about different role

models (submissive, dominant, and personal

leader-ship, see Lynch and Kordis 1988), and had the same

basic structure as the dialogue-training sessions, thus

providing additional exercises in dialogue

6 Results, effects of the training of dialogue

The effects of dialogue, as measured in the company,

were diverse A summary is given per study

Study 1 (Van Eijnatten et al 2002) mainly is about

a comparison of two project teams consisting of seven

members – one of which was the change management

team that received repeated dialogue training since

September 1999, and the other team did not receive

any training at all All 14 individuals had filled out 130

items (Likert scale type) questionnaires in January,

April and September 2000 A longitudinal Q-mode

analysis of the data revealed that in both project teams

(dialogue and non-dialogue groups) some consensus

had developed about the fading away of ‘existing

behaviours’, based on distrust and control (the old

mindset), in the first year of the dialogue project

However, only in the dialogue group, some consensus

had developed about the wish to have more of the

‘desired behaviours’ based in consciousness,

connec-tivity and coincidence (new mindset), mainly in the first

nine months of the dialogue project

Detailed analyses of interviews, held with three

selected managers of the dialogue group, indicated that

they had felt that they behaved differently: they

reflected more on their own behaviours, had become

more open and honest to others, and increasingly let go

of control by delegation of authority Further analysis

of an interview with a member of the non-dialogue

group revealed that this person had heard about the

project, and had experienced different attitudes among

managers, i.e., listening more deeply, showing patience,

and accepting criticisms in meetings

Study 2 (Van Eijnatten and Van Galen 2002) is

about the dialogue and role-training sessions for three

different technical project teams, consisting of 8–10

persons All trainings were held in a conference centre

outside the company, lasted a full working day, and

were facilitated by an external consultant The study

describes the set up and the contents of both the

training and test sessions, and contains an analysis of

the resulting conversations, in considerable detail Two

researchers acted as independent observers, made

transcriptions of what was said, and coded the data

in units of 10–15 minutes, in terms of examples of

either dialogue or discussion Although there was no

preset programme, the same dynamic rhythms

devel-oped in all project teams: three conversations were held

during training sessions in the morning, and five

conversations were held in the test sessions, in the

afternoon All individual conversations lasted forapproximately half an hour

Qualitative analyses of the data showed that similardialogue trainings had differential effects on the threetechnical project teams The project team that had onlyrecently been formed in the company, had engaged indiscussions only, and had failed to show any instance

of dialogue in the test session The other two projectteams had shown combinations of both dialogues anddiscussions, in the test sessions However, for all threeteams, the learning curve had been rather flat: it takestime, much trial and error, and energy to masterdialogue, and the moment emotions had crept into theconversation, the teams had tended to slide back intodiscussions, promptly

The role training for the three technical projectteams, that had also been set up as dialogue sessions, hadproduced opposite results The project team that hadperformed the poorest in the dialogue training per-formed the best in the role training, and one project teamthat had performed well in the dialogue training per-formed the poorest in the role training Further analysisshowed that the project team that had made the largestprogress had exercised dialogue ‘on the job’ in regularmeetings most frequently, in between training sessions.Study 3 (Van Eijnatten and Van Galen 2003) gives

an in-depth description of how the dialogue trainingwas carefully embedded within the context of bothorganisational and cultural renewal of the firm Itshows the integrative use of different tools: workshops

to convey some prime concepts of organisationalchange, desired behaviour patterns; dedicated trainings

to master the competence of engaging in dialogue;instructing to better understand, and training teammembers to be able to deliberately take on personalleadership roles; the actual use of dialogue to commu-nicate personal values and visions, and creating anarena for novelty and spontaneity, in which a designteam of employees (not managers!) volunteered todevelop and present new company values

Also, the paper describes how the company hadfurther diffused dialogue up to the level of theshopfloor both by lowering the abstraction level ofthe theoretical workshop, by defining behaviours that

fit with the organisational concepts, and by trainingdozens of internal facilitators to stimulate dialoguethrough the whole company

Study 4 (Van Eijnatten and Van Galen 2005) gives

an overview of the whole change project as it developed

in the manufacturing company Special treatment wasgiven here to the initiatives of the change managementteam, which consisted of three members of thecompany’s management team, and four middle man-agers Together with the external consultant and thetwo researchers they helped to unfold the cultural-change programme, within an action research context

As a team they developed the workshops, trainings and

conferences as shown in Figure 1 The change

manage-ment team also explored the possibilities of a

whole-company search conference, which unfortunately never

took place because they could not agree about its

contents The change management team assisted in

helping to create a design network, consisting of 35

company employees (no managers), in February 2002

This network chose among itself ten persons to form a

committee in order to discover organisational values

They increasingly used multiple communication

chan-nels like the company’s intranet, home-made movies,

performances, and small questionnaires sent out to all

company employees It was these two groups with some

support of the change management team that

suc-ceeded in planning and organising a whole-company

meeting, in February 2003 The employees split up into

several teams to prepare for this major cultural

convention: a whole-company dialogue/multilogue

event It had never been seen before in the company

Study 5 (Chen and Drost 2006) reports on the effects

of dialogue over a period of six years (2001–2006), on the

basis of both Q- and R-mode analyses The same

questionnaire that had been used in study 1, was

administered to employees of the manufacturing

com-pany on three occasions (in April 2001, February 2003,

and February 2006) Longitudinal analyses were

per-formed on the data of 2001–2003, 2003–2006, and 2001–

2006 The 25 employees who had been trained in

dialogue before April 2001 (this group included all

earlier mentioned project-team members from study

2 who were still with the company in 2006), were

compared as a group with the employees who still had

not received dialogue training (only four persons

remaining from the original control group), in February2006

Longitudinal analyses revealed that the dialoguegroup (YD) had thought old-behaviour patterns haddecreased between 2001 and 2003, and stayed the same

in the period 2003–2006 The non-dialogue group(ND) had felt old-behaviour patterns had stayed thesame for the whole period 2001–2006 In the YDgroup, the wish to increase new-behaviour patternshad increased in the period 2001–2003, and decreasedagain in the period 2003–2006 In the ND group, thewish to develop new-behaviour patterns had stayed thesame in the period 2001–2003, and increased in theperiod 2003–2006

In 2001, both YD and ND groups had reflectedsimilarly on the existence of old-behaviour patterns In

2003 and 2006, the groups had increasingly thoughtdifferently about it In 2001 and 2003, YD and NDgroups had felt differently about the wish to developnew behaviour patterns; in 2006 they had thoughtsimilarly about it In the whole research period 2001–

2006 the consensus about opinions among employeeswho were trained in dialogue had been higher thanamong employees who were not trained in dialogue In

2001, personal initiatives had been evaluated moderate

to high for both the dialogue and non-dialogue groups

In 2003 and 2006, a minor contrast had developed infavour of the dialogue group

Additional analyses of the original research dataconfirmed that the different technical project teamstrained in dialogue had been far from mutuallyexclusive, see Figure 2 Four teams are shown in theFigure: 1) technical project team A, consisting of eightpersons; 2) technical project team B, consisting of nine

Figure 2 Staffing structure of the technical project teams

persons; 3) technical project team C, consisting of eight

persons; and 4) the company’s change management

team, consisting of seven persons However, the total

number of unique persons was only 21, not 35, because

three persons were members of two teams, four persons

were members of three teams, and one person was a

member of four teams Some 13 persons were members

of one single team only Thus, in the manufacturing

company, in total eight managers and technical staff

persons had participated in more than one project team,

so they were able to practice dialogue in multiple

contexts Also, they interacted in non-dialogue contexts

with many other persons outside the project teams, and

it is plausible to assume that they applied there what

they had learned

7 The roles of dialogue in manufacturing system

integration

The secondary analyses lead to the following

conclu-sions about the roles dialogue had in this particular

company: i.e., integrative functions, action and

learn-ing triggers, and collective-learnlearn-ing processes

Integrative functions:

(1) In the researched company, dialogue had a

stimulating effect on the development of

consensus in project teams, with respect to

both thinking and courses of action Dialogue

integrated new and more coherent forms of

action (study 1)

(2) In the researched company, dialogue had an

integrative effect on the day-to-day

collabora-tion of both managers, engineers, and

work-ers, not by means of technical integration, but

by human interaction The new management

style better integrated people throughout the

organisational structure (study 2)

(3) In the researched company, employees who

had been trained in dialogue shared more

common opinions about old and new

beha-viour patterns, than employees who had not

been trained in dialogue Dialogue indeed had

an integrative function in terms of developing

a shared set of values or ‘organisational mind’

(study 5)

(4) In the researched company, dialogue training

did not seem to have a real effect on personal

initiatives, but rather seemed to enhance

mutual understanding and respect (study 5)

Action and Learning Triggers:

(5) In the researched company, dialogue fitted

very well in a programme of organisational

and cultural renewal, in which novelty issought (study 3)

(6) In the researched company, the spreading ofdialogue through the whole organisation led

to unprecedented, but self-orchestrated tions by employees themselves (study 4).(7) In the researched company, the diffusion ofdialogue across the whole organisation trig-gered managers to allow employees to developtheir own initiatives (study 4)

ac-Collective Learning Processes:

(8) In the researched company, dialogue wassystematically trained and exercised, and wasnot learned in an instant To become compe-tent in dialogue apparently is a painstakingand time-consuming process (study 2).(9) In the researched company, there was nogolden route to dialogue Each team em-barked on its own journey, at its own pace.Patience and perseverance were key, especiallywhen it concerned new teams that had justbeen formed (study 2)

(10) In the researched company, training in gue started a joint thinking and learningprocess, which functioned as an integrativemechanism that connects people (study 2).Integrating Function and Collective LearningProcess:

dialo-(11) In the researched company, dialogue was used

as a powerful integration mechanism to spreadnew values through the company A small but

‘critical mass’ of people competent in dialogueinfluenced or ‘infected’ other persons in allsorts of work contexts, and by doing so, caused

a snowball effect with respect to the ment of new values such as openness, trust,and respect (additional analysis of researchdata)

develop-The thesis of this paper was that the application of avery particular mode of conversation – i.e., dialogicalcommunication between managers, engineers, andworkers – serves as a generative mechanism inmanufacturing systems integration Engaging in dialo-gue will result in more mutual understanding of themeanings of both actions and interactions of thevarious professionals Also, it will increase virtues liketrust, appreciation, affection, and respect (Burbules1993) Looking at the results of the secondary analyses,the conclusion is granted that the studies thesis isconfirmed in this particular case

8 Methods and techniques for dialogue and the role of

computers

In general, there are three types of dialogue methods:

the Bohmian method that is focused on process

(Bohm 1980, 1987, 1990); the Socratic method that is

focused on content (Nelson 1965); and the System

Sensitive Dialogue Intervention method that is

focused on both content and process (Slotte 2006)

The original Russian ‘open game’ multilogue method

is focused on both content and process (Zaitzev 1998,

Zaitsev and Artemova 1998, Hoogerwerf and

Poorthuis 2002)

There are a number of different

dialogue/multi-logue techniques A distinction is made between the

real-life dialogue/ multilogue sessions, as they were

used in the case illustrations in this study (Gerard and

Teurfs 1995, Ellinor and Gerard 1998, Gerard and

Ellinor 1999, Zaitsev and Artemova 1998, Isaacs 1999,

Hoogerwerf and Poorthuis 2002), and the virtual

dialogue/ multilogue sessions, called ‘world or

dialogue cafe´’ (Brown, Isaacs, and World Cafe´

Community 2005) For the application of dialogue/

multilogue in manufacturing systems the mixed

meth-od (both content and process) seems to be most

appropriate

With respect to the real-life variant versus the

virtual variant of dialogue/multilogue, also a balanced

mixture of the two might be most desirable, since

electronic communication is an indispensable and

integral part of modern business life The internet is

used as a platform for learning and deliberation

According to Shank (1993) net communication is

basically semiotic, and a good example of ‘abductive

multilogue’ It must be kept in mind, that face-to-face

conversations still score best in terms of channel

information richness (Daft and Noe 2001), and are a

necessity for the development of trust among team

members in the earlier stages of a dialogical process

However, further development of new organisational

concepts, based on inter-enterprise dynamic

network-ing, will make real-life, collocated (non-distributed),

meetings nearly impossible In such a case, it is a

necessity to rely on electronic communication, or

conversational support of geographically distributed

participants in the dialogues/multilogues

Resuming, with respect to the dichotomy between

the dialogue/multilogue as a ‘non-computer’

integra-tion mechanism and the tradiintegra-tional Computer

Inte-grated Manufacturing (CIM) that aims at computer

integration, the dichotomy exists only when

abstract-ing, i.e., not considering the possible supporting

technology for the dialogue/multilogue processes

However, when computer support for

dialogue/multi-logue processes is used, the manufacturing system

integration becomes true computer integrated facturing (CIM), dialogue/multilogue based CIM Thedifference with the traditional computer integration isthat in the traditional computer integration themessages are sent by a machine–computer (algorithm),and the interpreter of the messages is anothermachine–computer (algorithm), while in dialogue/multilogue-based CIM the sender of the messagesand the receiver interpreting them are humans But asthe communication between these humans is notpossible without computers, it is true CIM

manu-9 ConclusionsThis paper is about an exploration of the rolesdialogue might play in manufacturing systems integra-tion It explains the basic differences between discus-sion and dialogue, and provides a comprehensiveillustration of the training and use of dialogue in amanufacturing company It must be stated againexplicitly, that generalisation is not aimed for Theanalyses strictly and only show the integrative func-tions, action and learning triggers, and collectivelearning processes that dialogue had in this particularcase company, during the specified period of time.With this restriction in mind, this study may be used as

a practical illustration for managers, engineers, andworkers who want to explore or reflect on potentialpros and cons of the use of dialogue in manufacturingsystems

A next point of discussion is when to apply eitherdiscussion or dialogue This concerns the question inwhat types of situations discussion or dialogue can best

be used Discussion is best used in situations wherechoices are clear and there is a need for urgent andtimely decision making Some practical examples are:Day-to-day operational decisions about productionplanning, purchase levels, logistics, etc The basicallyconvergent process of discussion leads on to theefficient selection of one single best alternative orchoice Dialogue is best used in those situations inwhich choices are fuzzy, there is a need for clarification

of concepts, understanding mental models, or openinquiry into new opportunities, and in which thereexists almost no time pressure The process of dialoguecan be used to explore potential courses of action oralternative directions in the design and integration ofmanufacturing systems, and to discover root causesand types of solutions during operation The divergentdialogue process of thinking together can lead to awhole range of new options or solution spaces.Another situation in which dialogue may be helpful

is the process of building self-managed teams Because

of the non-judgemental process that dialogue initiates,team members are able to learn to know each other

better, and will develop trust and respect for each

other Ideally, in many manufacturing practices, both

dialogue and discussion could be competently used in

tandem: That is, in a dialogue process new possibilities

are searched for that will be evaluated and prioritised

in a discussion process to prepare for actual decision

making

As the manufacturing case illustrated, in order to

be effective, dialogue has to be embedded in the

day-to-day processes of organisational development and

change It also would be plausible to anticipate that the

more radical these changes are, the more essential

dialogue will become in order to achieve and maintain

a shared understanding

Future research might further explore the efficiency

of communicative events in manufacturing

environ-ments, in order to find a proper balance between

divergence, i.e., developing/exploring new ideas by

means of dialogue/multilogue, and convergence:

decid-ing about ‘good-enough’ courses of action by means of

discussion Concerning the role of computers in

dialogue and multilogue based integration of

manu-facturing systems, future research might address:

(1) Enhancement of traditional computer

sup-ported conversational and collaborative tools,

such as videoconferencing and Computer

Supported Cooperative Work (CSCW) tools

(2) The use of advanced technologies for

support-ing dialogue, e.g immersive Virtual Reality

based and metaverse-like environments for

dialogue sessions (see, e.g., Putnik et al 2008)

(3) Advanced organisational environments for

inter-enterprise dialogues and multilogues that

will provide ‘safe’ environments that eliminate

trust creation and trust management problems,

e.g., meta-organisations such as advanced

versions of ‘Market of Resources’

(4) Organisational and social implications of the

use of computers for the support of dialogues

and multilogues in manufacturing companies

References

Argyris, C and Scho¨n, D.A., 1978 Organizational learning:

A theory of action perspective Reading, Massachusetts:

Adisson-Wesley

Bainbridge, L., 1983 Ironies of automation Automatica, 19

(6), 775–779

Bohm, D., 1980 Wholeness and the implicate order 1st ed

London: Roudledge and Kegan Paul

Bohm, D., 1987 Unfolding meaning: A weekend of dialogue

with David Bohm London: Ark Paperbacks

Bohm, D., 1990 On dialogue Ojai, CA: David Bohm

Seminars Extended version 1997 by L Nichol, ed

London: Routledge

Brown, J., Isaacs, D., and World Cafe´ Community, 2005.The world cafe´ Shaping our futures through conversationsthat matter San Francisco, CA: Berrett-Koehler.Buber, M., 1947 Between man and man London: Roudledgeand Kegan Paul, translated by R.G Smith

Burbules, N.C., 1993 Dialogue in teaching: Theory andpractice New York: Teachers College Press

Chen, C.-L and Drost, V., 2006 Culture creation at amanufacturing company: A longitudinal evaluation re-search into the development of an organizational mind.Eindhoven: Eindhoven University of Technology, De-partment of IE&IS, internal research report (in Dutch).Co-Intelligence Institute, 2009 Dialogue [online] Availablefrom: http://www.co-intelligence.org [Assessed 6 Novem-ber 2009]

Daft, R.L and Noe, R.A., 2001 Organizational behavior.Fort Worth, TX: Harcourt, 311

Eijnatten, F.M van and Galen, M.C van, 2002 The dolphinattractor: Dialogue for emergent new order in a Dutchmanufacturing firm [online] International ScientificJournal of Methods for and Models of Complexity, 5 (1).June Available from: http://www.fss.uu.nl/ms/cvd/isj/[Assessed 6 November 2009]

Eijnatten, F.M van and Galen, M.C van, 2003 Towardsorganisational bifurcation? Navigating and path findingusing dialogue In: G.D Putnik and A Gunasekaran,eds Business Excellence 1: Performance measures,benchmarking and best practices in new economy Braga,Portugal: School of Engineering, University of MinhoPress, 243–248

Eijnatten, F.M van and Galen, M.C van, 2005.Provoking chaordic change in a Dutch manufacturingfirm In: K.A Richardson, ed Managing organizationalcomplexity Volume I: Philosophy, theory, and application.Greenwich, CT: Information Age Publishing, 521–556.Eijnatten, F.M van and Hoogerwerf, E.C., 1999 A shortintroduction to multilogue In: T Chase, ed Proceedings

of the 1999 STS Roundtable Meeting in Monterey,California Northwood, NH: STS Roundtable, M2–M9.Eijnatten, F.M van and Putnik, G.D., 2004 Introduction:Chaordic systems thinking for learning organizations.The Learning Organization, 11 (6), 415–417

Eijnatten, F.M van, Dijkstra, L., and Galen, M van, Dialoguefor emergent order: An empirical study of the development ofthe organizational mind in a Dutch manufacturing firm[online] Paper presented at the 17th EGOS Colloquium,Lyon, France, 7–8 July Available from: http://www.chaosforum.com/docs/nieuws/Emergent%20Dialogue.pdf [As-sessed 6 November 2009]

Ellinor, L and Gerard, G., 1998 Creating and sustainingcollaborative partnerships at work: Dialogue, rediscoverthe transforming power of conversation New York: JohnWiley

Gerard, G and Ellinor, L., 1999 Dialogue: Something old,something new; dialogue contrasted with discussion; thebuilding blocks of dialogue: A living technology; Beha-viours that support dialogue [online] Available from:http://www.thedialoguegrouponline.com/whatsdialogue.html [Assessed 6 November 2009]

Gerard, G and Teurfs, L., 1995 Dialogue and tional transformation In: K Gozdz, ed Communitybuilding: Renewing spirit and learning in business SanFrancisco, CA: Sterling and Stone/New Leaders Press,142–153, summary by T Glaser [online] Available from:http://webinator.crinfo.org/articlesummary/10232/ [As-sessed 6 November 2009]

organiza-Hoogerwerf, E.C., 1998 Searching for new grounds through

Russian Open Gaming Nijmegen: Radboud University,

Nijmegen Business School, The Netherlands, internal

report

Hoogerwerf, E.C and Poorthuis, A.M., 2002 The network

multilogue: A chaos approach Journal of Organizational

Change Management, 15 (4), 382–390

Isaacs, W., 1999 Dialogue and the art of thinking together: A

pioneering approach to communicating in business and in

life New York: Doubleday

Jones, A., 2007 Speaking together: Applying the principles

and practice of dialogue[online] Available from: http://

www.spaceforlearning.com/docs/Speaking%20Together

%20-%20Alison%20Jones%20Sep%2007.pdf [Assessed

6 November 2009]

Lynch, D and Kordis, P., 1988 Strategy of the dolphin:

Scoring a win in a chaotic world New York: William

Morrow

Nelson, L., 1965 Socratic method and critical philosophy:

Selected essays New York: Dover (translation by T

Brown)

Putnik, G., Gonc¸alves, P., Sluga, A., and Cunha, M.M.,

2008 Virtual environments for dynamically

reconfigur-able Concurrent/Collaborative Engineering ‘virtual’

team CIRP Annals, 57 (1), 171–174

Senge, P.M., 1990 The fifth discipline: The art and practice

of the learning organization New York: Currently/Doubleday

Shank, G., 1993 Abductive multiloguing: The semioticdynamics of navigating the net [online] The ArachnetElectronic Journal on Virtual Culture, 1 (1) Available from:http://www.ibiblio.org/pub/academic/communications/papers/ejvc/SHANK.V1N1 [Assessed 8 April 2009].Slotte, S., 2006 Systems sensitive dialogue intervention.Systems Research and Behavioral Science, 23, 793802.Smith, M.K., 2001 Dialogue and conversation [online] In:The encyclopaedia of informal education London:YMCA George Williams College Available from:http://www.infed.org/biblio/b-dialog.htm [Assessed 6November 2009]

Tsoukas, T., 2009 A dialogical approach to the creation ofnew knowledge in organizations Organization Science,

20 (6), 941–957

Zaitsev, A.K., ed 1998 Self-learning organization: A globalperspective Kaluga: Kaluga Institute of Sociology(KaIS), internal report

Zaitsev, A.K and Artemova, T., 1998 Russian open game as

a tool for organization strategy development Kaluga:Kaluga Institute of Sociology (KaIS), internal report

Structure arguments for collaborative negotiation of group decisions in manufacturing systems

integration

Nan Jing* and Stephen C.-Y Lu

The IMPACT Research Laboratory, Viterbi School of Engineering, University of Southern California, Los Angeles, California

90089 USA(Received 18 May 2009; final version received 21 February 2010)

To support the collaboration negotiation of group decision-making activities in manufacturing systems integration,the authors have developed an approach to structure the negotiation argument with identified and organisedobjectives and preferences This paper reviews relevant research work and presents the framework and methods ofthis approach To overcome the limitations of these works in structuring argument for effective collaborativenegotiation, our approach synthesises a value-focused objective hierarchy with the generic argument structure tospecify how the arguments can be generated based on stakeholders’ proposals, objectives, criteria and preferences,and then exchanged amongst multiple stakeholders This synthesis framework also helps us suggest operationalmethods to evaluate the arguments based on the level of the objectives achievement In addition, this paper describes

a prototype system which implements our approach using the advanced web-based software technologies with thegoal of demonstrating how to systematically carry out effective collaborative negotiation of group decisions inmanufacturing systems integration

Keywords: argument; collaborative negotiation; group decision; manufacturing systems

1 Introduction and overview

The integration of the manufacturing systems often

involves a process beginning with the product design,

going through the production cycle and completing

the product deliverables to the customer needs The

process may be taken by multiple and distributed

enterprises in the phases of requirement analysis,

product design, material purchasing, manufacturing,

quality assurance, inventory control, and distribution

In each phase, one or multiple manufacturing systems

(or subsystems) may be deployed The ideal integration

of these systems should handle the design,

manufac-turing, test, and distribution for each part of the

product without disruption, from the raw materials to

the completed entirety It is clear that the integration

of manufacturing systems heavily impacts almost

every division of a manufacturing enterprise, including

product design and development, materials processing,

quality control, inventory management, and customer

service It usually draws the information from all these

divisions and performs an information integration

based on an enterprise-wide data warehouse It has

been said that the manufacturing systems integration

(MSI) helps to reduce design and production time and

create more market value by improving the

productiv-ity and profitabilproductiv-ity (Canada and Sullivan 1989)

Despite its known advantages, researchers alsoargue about the feasibility and effectiveness of theMSI, especially around how the decisions of designingand implementing the MSI process are made in thegroup that is often from various departments withconflicting interests (Kahraman et al 2000, Kahraman2001) In reality, many proposals for how to designand implement the manufacturing systems integrationhave been rejected because of different reasons,notably one of which is the lack of strong argumentsfor justifying the short-term returns of manufacturingsystems In fact, many of these proposals provideintangible benefits, e.g better product quality, greaterprocess flexibility, reduced inventory space and betteradoption experience with new technology However,without well-structured arguments with clearly identi-fied and presented values to the enterprise, all of theseintangible benefits are difficult to be noticed, nego-tiated and evaluated by the stakeholders in the groupdecision-making process during the MSI process.This difficulty is further compounded by the recentindustry globalisation and Internet revolution, wheremost of these group decision meetings are carried out

by stakeholders across geographical, disciplinary andtemporal boundaries Without thorough consideration

of every value brought by the MSI or structured

*Corresponding author Email: jingnan@gmail.com

Vol 23, Nos 8–9, August–September 2010, 720–738

ISSN 0951-192X print/ISSN 1362-3052 online

Ó 2010 Taylor & Francis

DOI: 10.1080/09511921003730819

arguments as the common ground, their negotiation

and decision making cannot achieve satisfactory

results, which will then hinder the quality of integrated

manufacturing system to a significant level

Researchers have tried to advance the development

of MSI by helping to justify the value of these

integrated systems Kahraman et al (2000) use a fuzzy

benefit/cost ratio analysis for the justification of

manufacturing technologies Kahraman (2001)

devel-op capital budgeting techniques using discounted fuzzy

cash flows However, these works only focused on

analysing the tangible benefits, i.e the financial value,

of the manufacturing systems integration, instead of

directly supporting the MSI stakeholders to explicitly

identify and include the intangible benefits in their

negotiation with other decision makers In these group

decision-making processes, the MSI stakeholders

should be able to collaboratively negotiate with other

stakeholders by providing well-structured arguments

and clearly presenting the value of the MSI related to

the enterprise Therefore, one of the critical challenges

in supporting the MSI stakeholders to more effectively

carry out collaborative negotiation in group

decision-making processes is to help them structure their

arguments with clearly identified and presented values

of the integration of manufacturing systems and guide

them to how to utilise these structured arguments for

more effective collaborative negotiation

Our work is motivated by this challenge We have

recognised the manufacturing systems integration

(MSI) brings in both tangible and intangible value

of the manufacturing enterprise All these values

should be clearly identified and presented, as much

as possible, in the arguments of the stakeholders who

participate in the collaborative negotiation of group

decision-making processes for MSI These arguments

should be commonly structured so that all the MSI

stakeholders, who are often geographically distributed

and cross-disciplinary, have a common ground for

reference and can effectively make well-informed

group decisions In addressing this challenge, the

authors have reviewed a variety of school of studies

related to group decisions Most of the existing

appro-aches of supporting group decisions have not provided

sufficient support to decision makers who have to

structure their negotiation argument with clearly

identified and organised value Therefore, although

these works have developed some theoretical

ap-proaches and systematic methods to support group

decision in general, their applications are very limited

when applied in the integration of manufacturing

systems to address the challenge we have identified

On the other hand, previously, a generic argument

structure has been defined by Toulmin (1958) It

has been widely utilised to build argument-based

negotiation process models in many studies of borative negotiation and proved to facilitate commu-nication However, these works do not effectivelysupport collaborative negotiation of group decision

colla-in modern MSI processes, such as identifycolla-ing andorganising stakeholders’ value (e.g their objectives fordesigning the integrated manufacturing system andtheir preferences for design solutions)

To overcome all these aforementioned limitationsand resolve these issues to effectively support colla-borative negotiation of group decision in MSI, ourapproach is to develop a new framework to helpstakeholders identify and organise the value of MSIand use these values to structure negotiation argu-ments for group decision-making activities By utilisingthese structured arguments, an operational process isdevised to guide the stakeholders to generate, ex-change, and evaluate the negotiation arguments inMSI To illustrate such as a framework developed inour work, the rest of this paper is structured as follows:Section 2 reviews a few schools of study which arerelevant to this work The research foundations of thiswork are introduced in Section 3 Based on thesefoundations, our framework to structure negotiationargument with organised value for MSI is presented inSection 4 Section 5 discusses how to implement thisframework using computer technologies Finally,Section 6 concludes this paper and outlines the openissues that are to be addressed to extend and improvethis work

2 Reviews of related works

As mentioned in Section 1, the manufacturing systemsintegration (MSI) is often a group decision-makingprocess based on the information from all the depart-ments in an enterprise This process is involved withmultiple stakeholders who might be geographicallydistributed, cross-disciplinary and asynchronous Ourchallenge is to identify and organise these stakeholders’value and then structure their negotiation argumentswith the organised value To put the discussion inperspective, this section reviews a variety of disciplines

in relation to group decision, collaborative negotiationand structuring arguments

2.1 Group decisionDecision scientists interested in group decisions haveinvestigated various negotiation models and decisionanalysis functionalities that help to achieve group con-sensus among multiple interests and competing posi-tions of stakeholders However, these models andfunctionalities have not provided full support todecision makers who have to identify, organise and

integrate their multi-disciplinary objectives and

per-spectives from distributed locations and asynchronous

communications The schools of study in this field

include game-theoretic analysis (Rosenschein and

Zlotkin 1994, Kraus 2001b, Sandholm 2002);

heuris-tic-based approaches (Faratin 2000, Kowalczyk and

Bui 2001, Fatima et al 2002, Klein 1995); and

argumentation-based approaches (Kraus et al 1998,

Parsons et al 1998, Sierra et al 1998) The details of

each study in relation to our work are explained below

Rooted in economics, game theory studies

interac-tions between self-interested agents The objective of

game theory is to determine the best (i.e most rational)

decision a player can make, using mathematical

modelling In order to do so, the player must take

into account the decisions that other agents can make

and must assume that they will act rationally as well

Game theory based approaches can help decision

makers understand and predict the outcome of a

negotiation and then help them make strategic

deci-sions in group decision-making process (Nagarajan and

Sosic 2008) A frequently mentioned drawback of game

theoretic approaches is the perfect rationality

assump-tion In order to select the best strategy, the player must

know the entire environment as well as the opponent’s

knowledge Unfortunately, in real world situations,

players have private information hidden from their

partners in the decision-making process

A way to overcome the game theory limitations

described previously is to use heuristic approaches

Heuristic-based negotiation is based on search strategies

where the objective is, instead of finding the optimal

solution, to find a good solution in a reasonable time

Stakeholders do not need to be perfectly rational and

information can be kept private While heuristic

methods do indeed overcome some of the shortcomings

of game-theoretic approaches, they also have a number

of disadvantages (Jennings et al 2001) First, the models

often lead to outcomes that are sub-optimal, because the

information and space that the negotiation team can

explore is always limited by the design of the heuristics

method, which is usually ad hoc Second, because of the

ad hocdesign of the heuristic method, it is very difficult

to predict precisely how the team and stakeholders will

behave and there is usually no guaranteed solution at

the end of the execution of the heuristics Consequently,

these approaches usually need extensive evaluation

through simulations and empirical analysis which is

not often available owing to the resource limitation in

engineering processes

Another category is argument-based negotiation

approach that follows a generic structure of arguments

defined in Toulmin (1958) which helps stakeholders

lay out their negotiation information and

meta-information into various components and this

information includes major claims, support data andadditional persuasive perspectives such as justification,degree of desire and rebuttal condition In the nego-tiation approaches presented previously, stakeholderscannot justify to their partner why they refuse an offer

or what part of the offer was problematic Proposals

do not include the explanations of the positions andconsiderably limit the potential of negotiation Theidea behind argumentation-based negotiations is pre-cisely to give this additional information (e.g ajustification about why the partner should accept aproposal) to stakeholders, helping the negotiationprocess by identifying part of the support data andbackground information that does not get exploredotherwise Different authors have presented applica-tions of argumentation-based negotiation models(Jennings et al 2001, Atkinson et al 2005, Capobianco

et al 2005, Buttner 2006) These approaches canincrease the efficiency of negotiation process by addinginformation that was not used before By revealingnew information, the partner can be persuaded that acertain proposal is better than it thought Based on theadvantage of these approaches in systematically andeffectively organising and conveying stakeholders’perspective, One of the main limitations of theseapproaches falls in that the stakeholder must be able toevaluate the arguments and estimate their performance

in order to choose one best argument This is because,although Toulmin defined the generic and well-adopted argument structure, he proposed his views

on argumentation informally in loosely specifying howarguments relate to other arguments and providinglittle guidance as how to evaluate the best one(Zeleznikow 2002) It is still more intended as a way

of checking and arranging arguments for overlookedflaws (Houp et al 1998) instead of directly supportinggroup decision-making, such as specifying how toevaluate the structured argument based on thegoverning factors in decision making (e.g stake-holders’ objectives and perspectives)

2.2 Semiotics for collaborative negotiation

In order to design an effective collaborative tion approach to support group decisions, an interest-ing school of study the authors have investigated issemiotics, which investigates how human use of signsand symbols, and their reaction to the interpretation ofsigns, are employed in communication and coordina-tion A general definition for semiotics is ‘a generalphilosophical theory of signs and symbols that dealsespecially with their function in both artificiallyconstructed and natural languages and comprises ofsyntactics, semantics, and pragmatics’ (Miriam 2007).Organisational semiotics (Filipe and Liu 2000), a

negotia-subdivision of semiotics, focuses on the analysis of

organised decision activities in business settings and is

thus more related to our research in this domain There

are two concepts very popularly used in this

organisa-tional semiotics, namely ‘sign’ and ‘norm’ A sign is

something that is utilised by someone for something

else to indicate certain respect or capacity A norm is a

generalised disposition to the information shared by

stakeholders of the organisation (Liu 2004)

Semiotics has been widely used in computer

supported cooperative work (CSCW) and human

computer interaction (HCI) when handling the

inter-facing between the stakeholders and the system

Their research proves that the structure and

inter-pretation of stakeholders’ activities (sign) should be

analysed as they indicate the value that stakeholders

hold through their logical deduction (norm) (Andersen

2001) This implies to us that, in our work of

develop-ing a collaborative negotiation approach referencdevelop-ing

semiotics, stakeholders’ objectives and preferences in

certain values of the manufacturing products (norm),

should be included in stakeholders’ structured

argu-ments to make the analysis of stakeholders’

negotia-tion activities (sign) more effective for exploring the

stakeholders’ value (norm)

Semiotics also strongly holds that the interface

amongst stakeholders, or between stakeholders and the

computer systems, should be not only interpretable,

but also verbalisable (Andersen 2001) The same

principle is applied in our research as well regarding

how to structure the negotiation arguments of the MSI

stakeholders All the negotiation arguments should

explicitly explain the stakeholders’ values (e.g

objec-tives and preferences) and the way they are structured

should help stakeholders more easily describe their

argument verbally In fact, it will be more helpful

for the communication between the stakeholders, if

their arguments can be visualised Janssen and Sage’s

study shows it is advantageous in certain cases to a

verbalised or structured depiction than natural

lan-guage description (Janssen and Sage 1996) They have

stated the reasons as follows: first, visualisation eases

comprehension The components of the argument are

explicitly represented, meaning that it is easier to

identify the particular elements of an argument, and

the structure comprising these elements facilitates the

elicitation of the elements Second, it is easy for the

person filing the boxes to see what is missing as well as

the reasoning that has been put forth In this regard, it

is easier to compare arguments between multiple

experts and between claim and counterclaims than

between statements in generally unstructured discourse

(Janssen and Sage 1996)

In a summary, the research of Semiotics, especially

organisational semiotics, from the pragmatic level,

justifies the necessities of structuring arguments withstakeholders’ objectives and preferences in a waywhich can make them easily verbalised and visualised.This necessity will be addressed by the research that ispresented in this paper

2.3 Engineering collaboration via negotiationparadigm and socio-technical framework

In order to resolve the challenge of our work instructuring negotiation arguments with identified andorganised objectives and preferences, in this section

we will review our previous work in an engineeringcollaboration via negotiation (ECN) paradigm and asocio-technical framework (STF) This work helps uslay out key concepts and components in defining anegotiation process utilising structured arguments fordecision tasks This section will explain these concepts(i.e ECN and STF) in detail

Real-world negotiation tasks undertaken by gineering teams are always driven by many conflictingsocial, economical and technical (SET) factors Tradi-tional engineering research has mainly focused ontechnical factors, with some recent efforts beingextended to consider the economic factors Whilerecognising the importance of both technical andeconomical considerations, our past research has beenfocusing on social factors, and more specifically, ontheir interactions with technical factors The authorsview an engineering team activity, such as developingsoftware, as a technical activity with a human purpose.Therefore, when a team of engineers (i.e multiplestakeholders) with differing life cycle concerns cometogether to develop new engineering solutions, it canlead to a complex socio-technical campaign To resolvethis challenge, an engineering collaboration via nego-tiation (ECN) paradigm was developed in our pastwork to define the key concepts and lay out thebackground for this type of socio-technical campaign inthe field of collaborative engineering

en-Based on this ECN paradigm, the authors havedeveloped a socio-technical framework (STF) as afoundation to pursue basic research in collaborativeengineering The STF has its roots in the SocialConstruction Theory proposed by Peter Berger andThomas Luckmann in 1966 (Berger and Luckmann1967), which states that meaning and institutions (e.g

a collaborative decision during software development)are a jointly negotiated and agreed constructionbetween the participatory stakeholders More specifi-cally, our STF uses perspective models of thesestakeholders to integrate social interactions withtechnical decisions, and then uses these models tomanage decision conflicts during a co-constructionprocess taken by the engineering team

As stated, there are three components in this STF.

The first component is the technical decision process,

which serves as the baseline process for socio-technical

co-construction process The ‘baseline process’ refers

to a series of required activities, consisting of tasks and

states, which must be performed by stakeholders

according to some pre-established steps adapted from

the specific domain practices or mandated by

corpo-rate policies The second component of our STF is the

perspective model of team stakeholders In our

research, we define a perspective as the stakeholders’

viewpoints towards specific concepts of a particular

campaign In this way, we can analyse these

perspec-tive models to estimate their differences as a measure of

the distance between the viewpoints of stakeholders

These analysis results can provide explanations for the

stakeholder decisions and/or offer rationales for

conflict management by the team The final component

of our STF is conflict management In our research,

conflicts are the result of perspective model analysis

and are managed systematically based on these

analysis results In our research, we detect and analyse

conflicts according to the composition of both

stakeholders’ concepts and their perspective models

(i.e the sources of conflicts) And we manage conflicts

by managing the negotiation processes where

stake-holders’ perspectives are co-constructed

3 Research foundations

3.1 A socio-technical co-construction process

Based on the ECN paradigm and the STF framework

that are discussed in Section 2, the authors take

the next step to develop a specific process through

which this conceptual framework can be further

detailed and made ‘operational’ for computer

imple-mentations and engineering applications This

pro-cess, which is called socio-technical co-construction

(STC) process, specifies an operational procedure

to address the three key components of the

socio-technical framework

3.1.1 The eight steps of the socio-technical

co-construction (STC) process

(I) Define a starting ‘baseline process’ for the

chosen application domains, as the basis to

be co-constructed (i.e changed) later, upon

the agreement by all involved stakeholders,

as a means to resolve conflicts The defined

baseline process can be further broken down

into a series of tasks and states

(II) Identify a group of ‘stakeholders’ who have

an interest in the outcomes of, and will

directly or indirectly participate in, the

co-construction process of a particularcollaborative campaign (e.g MSI design).(III) Propose an initial ‘concept structure’ (CS)for a particular engineering process toorganise the concepts provided by the team.(IV) Establish the initial ‘perspective model’ forall participating stakeholders to expressopinions for each concept in the conceptstructure

(V) Build the ‘perspective model state diagram’(PMSD) for each concept in the conceptstructure A PMSD is a research apparatus

in STF to depict the explicit relationshipsamong stakeholders’ concepts (includingboth shared and individual concepts) inaddition to their purpose and contextinformation

(VI) Perform the ‘perspective analysis’ on thecurrent PMSD Since PMSDs link the CSwith perspectives and have references tothe baseline process, they provide integratedinformation to conveniently analyse thecloseness of, or distance among, differentstakeholders’ perspectives at that particularmoment

(VII) Conduct the ‘conflict management’ tasksaccording to the results perspective analysis.(VIII) Obtain a ‘shared reality’ as a result of theco-construction process This final product

of the STC process is a shared reality, which

is a broader concept than traditional proaches (e.g., a finished design in terms of aproduct model)

ap-The socio-technical co-construction process(STCP) provides us with research context and grounds

of building a new negotiation process, as STCPspecifies eight steps with sufficient operational details

to guide the stakeholders through the process of constructing group decisions However, stakeholderswho works in STCP are required to fully share theirperspectives (e.g proposals, objectives, preferences,and justifications) yet they have not been guided abouthow to organise their perspectives and structure thenegotiation arguments with these perspectives duringthe process

co-3.2 Argument structure: Toulmin’s structure ofarguments

Practicing collaborative design and negotiation gue has been found to be positively linked withargument development and critical thinking skills(Smith 1977, Hart 1990, Parsons et al 1998, Jin et al

dialo-2005, Marttunen 1992) Furthermore, the work of

Buckingham and his colleagues argue that

standardis-ing an argument’s structure facilitates its subsequent

communication since important meta-information

and relationships can be more easily perceived and

analysed by others (Buckingham et al 1997) Stephen

E Toulmin’s 1958 work, ‘Uses of Argument’ has

become commonplace in structuring arguments –

Toulmin acknowledges as much in the preface to his

1984 text, ‘An Introduction to Reasoning’ (Toulmin

et al 1984) For example, Houp, Pearsall and

Teheaux’s textbook, Reporting Technical Information,

introduces Toulmin logic as providing ‘a way of

checking your own arguments for those overlooked

flaws It can also help you arrange your argument’

(Houp et al 1998)

The goal of developing arguments in negotiation is

to persuade or convince others that one’s reasoning is

more valid or appropriate Toulmin’s structure of

argument provides a visual way of combining the

language symbols and data structures that support

the argumentation process (see Figure 1) Toulmin’s

structure is procedural and the layout of this structure

focuses on the movement of accepted data to the

claim through a warrant Toulmin also recognises

three secondary elements that may be present in an

argument: backing, qualifier, and rebuttal Backing is

the authority for a warrant, provides credibility for the

warrant, and may be introduced when the audience is

unwilling to accept the warrant A qualifier indicates

the degree of force or certainty that a claim possesses

Finally, rebuttal represents a certain condition or

exception under which the claim will fail and hence

anticipates objections that might be advanced against

the argument to refute the claim (Toulmin 1958)

As such, Toulmin’s argument structure becomes a

popular mechanism for structuring arguments betweennegotiating stakeholders It aims to clarify the reason-ing process by encouraging parties to make explicitimportant assumptions, distinctions, and relationships

as they construct and rationalise ideas (Buckingham

et al 1997)

Using the Toulmin’s argument structure, which isgenerally more objective than implicit arguments, it ishard for stakeholders to hide bias because the groundsand backing of an argument are clearly listed anddescribed to support the claims Therefore, all stake-holders’ perspectives are generally relatively easy to befully observed by others through examination of theground and warrants that the stakeholder expresses(Janssen and Sage 1996) The process of preparing theargument components has been proven to be anobjective way to organise and convey stakeholders’perspectives In addition, the logic flow between thesecomponents inherently helps us specify the order ofexchanging perspectives in the group negotiationbetween the multiple stakeholders As mentionedearlier in Section 2.1, argument structure has beenused to build an argument-based negotiation processmodel in many previous studies including collaborativedesign and proved to facilitate more objective and faircommunication (Chang and Han 1995, Sierra et al

1998, Sillince and Saeedi 1999, Amgoud et al 2000,Avery et al 2001, Kraus 2001a, Rong et al 2002).However, there are remaining unresolved issues inmost of the above work, such as a systematic guide toevaluate the arguments for the best in an operationalnegotiation process Our research pushes this forwards

by developing a collaborative negotiation frameworkthat synthesises this generic structure with an objectivehierarchy in the context of the socio-technical co-construction process in order to structure the argu-ments with stakeholders’ objectives and preferences,which can be used for arguments evaluation Moredetails about the objective hierarchy are discussed

in Section 3.3 and the synthesis is introduced inSection 3.4

3.3 Value-focused objective hierarchy

It is our belief that decision conflicts have roots instakeholders who have different perspectives (e.g.understandings and expectations) of the engineeringtask, such as the group decision-making in design anMSI solution, our approach uses an objective hier-archy as an organisation of the objectives on which thestakeholders can declare their perspectives Further-more, these objectives should also help stakeholdersreveal the real ‘values’ the stakeholders want to get out

of the design task These values are the expectationsfor why the stakeholders join in a negotiation process

Figure 1 Toulmin’s argument structure