International journal of computer integrated manufacturing , tập 23, số 3, 2010

Integrated approach to modelling human systems as reuseable components of manufacturingworkplacesJoseph Ajaefobi, Richard Weston*, Bilal Wahid and Aysin Rahimifard Manufacturing Systems

Integrated approach to modelling human systems as reuseable components of manufacturing

workplacesJoseph Ajaefobi, Richard Weston*, Bilal Wahid and Aysin Rahimifard

Manufacturing Systems Integration (MSI) Research Institute, Loughborough University, LE11 3TU, UK

(Received 31 May 2007; final version received 5 September 2009)

A new approach to modelling human systems as reusable components of manufacturing workplaces is described.Graphical and computer executable models of people competences and behaviours are created which arequalitatively and quantitatively matched to equivalent models of process networks, decomposed into roles anddependencies between roles To enable model creation and reuse, coherent sets of role, competence and dynamicproducer unit (DPU) modelling concepts have been defined and instrumented using enterprise modelling (EM),simulation modelling (SM) and causal loop modelling (CLM) techniques This paper reports on an application ofthe modelling approach to create related models of ‘process oriented roles’ and ‘candidate human systems’ so as tosystemise matching of role requirements to resource systems attributes and to inform aspects of strategic and tacticaldecision making in an SME making composite bearings

Keywords: dynamic producer unit; enterprise modelling; human systems modelling; process modelling; simulationmodelling

1 Introduction

Manufacturing enterprises (MEs) are designed and

engineered by people to achieve a wide range of goals

Normally, MEs comprise structured and technology

enabled systems of people who process physical and

informational workflows to add value to specific

pro-ducts in timely and cost effective ways Globalisation

of product markets, product customisation and

short-ening product lifetimes all impact in terms of increased

and more frequent customer requirement changes

(Krappe et al 2006) To cope with customer and other

environmentally induced product dynamics, MEs

operating in most industry sectors require enhanced

competences (provided by human systems) and

im-proved capabilities (from technical systems) to realise

their process-oriented roles effectively and in a timely

manner Generally, in MEs, people have collective and

ongoing responsibility for: (1) deciding what an

enterprise should do, (2) deciding how the enterprise

should be structured and use available supporting

technology to achieve those desired goals and

con-strain unwanted behaviours and (3) do most of those

product realising activities in a structured and

techni-cally-supported way as defined by (1) and (2) (Weston

et al 2004.) Essentially, people-centred organisations

such as MEs are complex in terms of their

compo-sition, structures and operations Consequently,

effec-tive and timely realisation of value adding activities

requires selection and matching of appropriate source system competences and capabilities to MErequirements, on an ongoing basis, to change resourcesystem solutions as ME requirement changes (Skyttner

re-2005, Swarz and DeRosa 2006) To achieve desiredresponses to ME requirement changes, enterprise sys-tems and their human and technical components areoften recomposed, reconfigured and reprogrammed

As necessary, enterprise system change can give rise toemergent behaviours in MEs with resultant changes

in operational scope and role requirements for peopleand their supporting technical systems To remaincompetitive however, MEs need constantly to developtheir (human and technical) systems so that their com-petences, capabilities and capacities remain aligned toemergent business and environmental requirements.With increased business fluidity comes a growingneed for change capable manufacturing organisations,namely organisation that possess an ability to ‘recom-pose’, ‘reconfigure’ and ‘reprogram’ their systemcomponents rapidly and effectively In turn this requiresimproved understandings about how business andenvironmental change can be realised via suitablechange to process structures and how this is related torequired resource systems structures, attributes andbehaviours (Zhen and Weston 2006, Weston et al.2007) Such understandings can be gained by studying

ME ‘requirements’ and related ‘resource components

*Corresponding author Email: R.H.Weston@lboro.ac.uk

International Journal of Computer Integrated Manufacturing

Vol 23, No 3, March 2010, 195–215

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

Ó 2010 Taylor & Francis

DOI: 10.1080/09511920903527846

and their configurations’ in model views Models of

MEs and their component parts can be captured in

different views, at different levels of granularity, via

alternative methods and by deploying appropriate

modelling tools and modelling languages; thereby

making it easier to represent, visualise, analyse,

under-stand and possibly predict behaviours of viable

config-urations of enterprise components and to inform

management decisions

Many ‘method-based’ approaches to engineering

change in MEs have been conceived and are becoming

widely adopted by industry including: just in time and lean

manufacture, agile manufacturing and postponement and

mass customisation (Womack et al 1990, Womack and

Jones 2003) However, it is observed that typically in

industry the application of these change methods:

(i) is ad hoc, constrained and piecemeal;

(ii) supports qualitative, rather than quantitative

analysis;

(iii) does not facilitate an ongoing externalisation

and reuse of organisational knowledge and data

(iv) is techno-centric, with limited characterisation

of impacts of people system roles,

compe-tences, behaviours and cultures

Hence the present authors propose the use of a

model-based approach to underpinning manufacturing

organisation design and change In conjunction with

enterprise modelling (EM), causal loop modelling

(CLM) and simulation modelling (SM) can usefully

be deployed to achieve ME requirements specifications

and capture, resource systems (solution) design, and

the ongoing matching of emerging requirements to

changes in solution design Since early 2000, the

authors modelling research have specified developed

and case tested systematic uses of state of the art

EM, CLM and SM technologies to develop virtual

models of large and small scale manufacturing systems

Essentially, their approach unifies the use of: (a)

decomposition principles defined by public domain

EM methodologies especially CIMOSA (AMICE

ESPRIT Consortium 1993), (b) causal and temporal

relationship modelling notations, provided by CLM

technologies, (c) discrete event and continuous SM

tools to computer exercise behaviours of selected

configurations of work loaded process segments and

their underpinning resource systems and (d) mixed

reality modelling based on the use of workflow

modelling techniques that enable interaction and

information interchange between simulation models

and real resource systems In that context, this paper

reports on progress made with respect to developing

models of people in their manufacturing work places

for the purpose of realising enhanced enterprise

behaviours and performances that continue to matchexplicitly defined but changing ME requirements

2 ‘Modelling people at work’ in MEsEvidently it is difficult for humans to model themselvesfor a number of reasons which include the following:(1) People are complex entities that generatevarious (individual and collective) behavioursthat are often context dependent (Ajaefobi

et al 2006);

(2) People acting as modellers often have strained understanding, knowledge and dataabout themselves, about the modelling contextand about related causal impacts

con-However every day we all generate and use simplemodels of ourselves, our fellows, colleagues, compa-nions, etc and of related situational impacts Researchreported in this paper is concerned with understandingand characterising problems and constraints associatedwith modelling people in ME workplaces Definitivefoci of reporting is on creating and using models of

‘human systems’ in relationship to common rolesperformed in MEs Here the term ‘human systems’ isused to infer either: competent individuals workingsystematically; loosely affiliated ‘workgroups’; orclosely coupled teams of people deployed to interact

in a structured work environment Generally though,resourcing value adding roles in MEs involves theuse of (a) systems of competent people, (b) suitabletechnical systems and (c) combination of (a) and (b).The choice and deployment of the above resourcesystem types depends on the nature of the require-ments, i.e the nature of the work to be done, whichoften dictates the extent of human involvement andpossible extent of automation, the costs of deployingparticular resource system types and the nature ofexpected outcomes To enable human systems tofunction effectively in MEs, various organising struc-tures that impact on their actions and behavioursare commonly deployed including: human organisingstructures (such as hierarchy, roles, responsibilitiesand authority) (Ashkenas 1995, Hendrick 1997),work organising structures (such as processing routes,batching and prioritising rules and ‘job’ and ‘task’assignments) (Bennis 1996, Vernadat, 1996, Medskerand Campion 1997) and enterprise cultures (includingcorporate beliefs and values)

Previous research findings by the present authorshad observed a key differentiation between human andtheir technical system counterparts which centre on acommon human ability to reflect on job performanceoutcomes and thereby as necessary to (1) develop new

competences and/or (2) develop new structures;

there-by modifying their behaviours and the behaviour of

the entire system leading to improved performance

(Weston et al 2003) People are therefore more flexible

than most technical systems because they have the

ability to reflect on (and develop) what they do, their

work patterns and behavioural relationships

To realise a prime objective of this research (i.e to

systemise and support with models aspects of matching

people to roles in the context of specific and changing

ME work places), it is assumed that people and

technical systems and the process-oriented roles they

realise in manufacturing workplaces need to be

modelled in a coherent manner Also it is assumed

that in conformance with established general systems

engineering practice, flexible ‘interconnection’ is

re-quired between developed models of process-oriented

roles (that explicitly define work requirements) and

developed models of solution configurations of

human and technical resource components To identify

common ME requirements (things MEs do to create

values for their customers), the authors have adopted a

process view of MEs thereby modelling specific ME

requirements as a specific network of dependent

processes and their derivative roles Previous authors

have classified and characterised processes commonly

found in most MEs (Pandya 1997, Chatha et al 2007)

A consensus view is that MEs typically deploy people

and technical systems to realise the following process

types:

(1) processes that realise products and services for

customers;

(2) processes that ensure that those product and

service realisations are well managed, such that

they remains aligned to established business

and manufacturing policies and strategic goals

of the ME; and

(3) processes that structure and enable ongoing

change as the ME systematically renews and

reconfigures itself, developing and

implement-ing new strategies, policies and processes in

response to external change

While conceiving, specifying, developing, realizing

and changing enterprise processes; people exercise

different role types, namely interpersonal,

informa-tional, decisional and operational roles (Mintzberg

1989, Steers and Black 1994) The term role can

further be described as: functions (tasks or activities)

that need to be performed by role incumbents; identity

created (or positions occupied) by incumbents in a

social structure while performing the role and

beha-viours that people or stereotypical people can/will

bring to roles and the management of role

dependencies (Wagner and Hollenbeck 1992, Steersand Black 1994, Ashfort 2000) The focus of this paper

is on functional roles; which represent sets of functionalactivities and operations that are resourced by peopleand their supporting technology during product(service) realisation (Zaidat et al 2004) To effectivelysatisfy role requirements, i.e attainment of specificresults required by the role through specific actionswhile maintaining or being consistent with the policies,procedures and conditions of the organisationalenvironment (Boyatzis 1982), role incumbents need

to bring to bear upon assigned roles work-relatedattributes especially competences The term compe-tence is presume to mean those work-related attributes,including: natural traits (underlying attributes), ac-quired traits (knowledge, skills, education, training,experience, etc) and consistent performance outcomesfor which a given resource system is known It followsthat in resourcing ME roles, two types of competencesare evident: (1) competences required by roles and (2)available competences possessed by potential roleincumbents (Harzallah and Vernadat 2002) Selectionand matching of people to roles involves ‘mapping’between available competences and required compe-tences Any such mapping is naturally constrained byfactors such as:

(a) Does the selected candidate system (person orpeople) have all the competences required bythe role(s)?;

(b) If the answer to (a) is affirmative, what is thecapacity i.e (how much in case of quantifiableoutcomes) will the system deliver in a giventime frame?;

(c) Can the selected candidate system cope withchanging workload requirements, includingchanges related to production volumes andproduct variances?;

(d) What aspects of the required competences arelacking in the solution provision?;

(e) Can such deficiencies (as identified in (d)) beremedied by training or upgrading the support-ing technology so as to enhance achievableperformance of the deployed candidate system?Addressing the above questions requires modellingconcepts with analytical and dynamic features to sup-port data capture on requirements, representation andanalysis of the available and required competences

In subsequent sections, data capture and modelling ofrequirements from such captured data, modelling ofcandidate solution and how simulation modelling can

be used to match specified requirements to alternativesolutions in both structural and dynamic behaviouralterms are discussed

International Journal of Computer Integrated Manufacturing 197

3 Need for new modelling concepts

To develop an approach to modelling human systems

as reusable components of MEs, it was observed to be

necessary to realise the specification and selection of a

modelling method with capabilities to:

(a) represent and abstract generic and specific ME

requirements, in terms of the required network

of processes used by any specific ME to realise

products and common workflows through

different segments of that process network

Here it was envisaged that such a modelling

method would facilitate and systemise the

decomposition of process-oriented

require-ments (explicitly modelled as process segrequire-ments

and their needed workflows) into well-defined

roles that are themselves can be decomposed to

enable their explicit modelling at different levels

of granularity; so that later the roles defined

can be flexibly matched to ‘work centres’ that

can be physically realised by suitable (human

and technical) resource systems;

(b) represent, decompose, abstract and structure

models of human systems and their component

elements in terms of work-related attributes,

and especially functional competences;

(c) facilitate qualitative and flexible matching

between models of people (competences) and

models of process oriented roles;

(d) enable the selection and testing of alternative

role-people ‘couplings’ in a simulation

environ-ment, so that quantitative comparisons can be

made between the behaviours of alternative

candidate role-people couples when they are

subjected to historical and/or possible future

changing ME requirements

To satisfy the modelling requirements listed and to

achieve the envisaged benefits, a suitable modelling

approach needed to be specified and selected In

principle, any of the state of the art EM methods

that have been successfully tested and usefully applied

in industry (such as CIMOSA, IDEF, PERA and

ARIS) could have been chosen as the foundation

modelling method However, the authors chose to

deploy the open system architecture for computer

integrated manufacturing (CIMOSA) (Kosanke 1995,

Zelm et al 1995, Vernadat 1996) In spite of its known

modelling strengths, CIMOSA has some notably

weaknesses including (a) models developed using

CIMOSA are essentially static and hence cannot be

used to reason about changing requirements and the

impact of such changes on selected and alternative

resource systems and (b) CIMOSA does not have

specific modelling constructs to represent humansystems competences To address the limitationsobserved, a modelling framework incorporating rolemodelling concepts, competence modelling conceptsand the use of SM modelling tools to instrument role-people couplings was proposed The unified modellingframework proposed uses CIMOSA (AMICEESPRIT Consortium 1993) as the main modellingfoundation but it has extended the modellingcapabilities of CIMOSA (by exploiting its eclecticnature) to incorporate competence and SM modellingconcepts Furthermore, to reflect the fact that humansystems execute assigned roles while being supported

by technical systems, the dynamic producer unit(DPU) concept previously proposed by the authorsand their research colleges was employed to furthersystemise human systems modelling DPU modellingconstruct was proposed for the purpose of abstractdescription of enterprise resource units comprisingpeople, machines, computers and or a structuredcombination of those; that is a reconfigurable, reusableand interoperable component of complex organisationssuch as a manufacturing enterprise (Weston et al.2009)

4 Modelling methodology conceivedThe modelling framework shown in Table 1 wasproposed and developed to enable the capture ofcoherent models of ‘process-oriented roles’ and ‘hu-man systems commonly found in specific manufactur-ing work contexts’ Section 5 of this paper describes acase study application of this integrated approach tomodelling human systems, as reusable components ofmanufacturing workplaces

5 The case study company and its modellingrequirements

5.1 Company background

A composite bearing manufacturing company whichmakes to order a wide range of composite bearingproducts was chosen as the subject of a case study.For reasons of confidentiality the authors will refer tothe company as ComBear Ltd ComBear Ltd is arapidly growing UK based SME with a customer basewhich extends beyond Europe ComBear manufac-tures different composite bearing products suitable foragricultural, marine, mechanical, pharmaceutical andfood processing applications Essentially, all Com-Bear products are manufactured from reinforcedplastic laminates composed of synthetic fabricsimpregnated with resins and lubricant fillers Finalproducts are delivered to customers in tube andsheet forms as well as fully finished components

such as structural bearings, washers, wear rings,

sphericals, wear pads, wear strips, rollers, and bushes

Figure 1 shows some of ComBear’s current product

range

5.2 Reasons for modelling

The objectives of the research funded by the UK’s

EPSRC are described in detail in the EPSRC case for

support (Weston 2005) In the ComBear case morespecific modelling goals were agreed with the companymanagement and are listed as follows:

(1) to document ComBear’s current network ofprocesses formally, identifying who does whatand with what and at what time;

(2) for selected segments of ComBear’s currentnetwork of processes, to create computer

Table 1 The modelling stages of the ‘integrated approach to modelling human systems’

Stage

Purpose of each modelling stage and the

Stage 1: ‘Context Modelling’ Enterprise Modelling is used at this stage to

decompose and graphically represent relativelyenduring aspects of the specific network ofBusiness Processes (BPs) used by the ME understudy Stage 1 modelling is focused on characterisingproperties of the process logic currently used by thesubject ME

Network of BPs – used torealise products and services Segments of the processnetwork – that must beresourced by suitable human(and technical) systems Process segments aremodelled in terms of activity,information, material,control and exception flows.Stage 2: ‘Role Specification’ Various groupings of enterprise activities (that

constitute specific process segments and realisedependencies between process segments) areanalysed with respect to their ‘functional’ and

‘behavioural’ requirements Various groupingrules (based on research findings from workdesign, human science and the processmodularisation literature) are used to: identifyand specify viable roles and role relationshipsfor human systems

Required ‘functionaloperations’ and ‘functionalentities’

Viable role behaviours androle relationships

Functional and behaviouralspecifications for viable roles

Stage 3: ‘Shortlisting of

Candidate Systems’

A shortlist of candidate (human and technical orDPUs) resource system designs is established

in terms of their potential to match:

(1) competences and characters possessed bycandidate human systems to (2) required functionsand behaviours of viable roles and role relationshipsdefined during stage 2

Relative performance levelsand costs of resources aretabulated

A short list of candidateresource systems – withpotential to realise specifiedroles

Stage 4: ‘Modelling Dynamic

Behaviours’

Dynamic behaviour of process segments resourced byviable candidate human systems are modelled usingCLM and SM technologies in a unified way

The computer executable SMs so produced reusespecific structures and data about the businesscontent and ME process network defined previously

by the EM during stage 1 modelling Thereby the SMsencode: (1) specific process logic (and embedded rolerequirements); (2) alternative attributions of shortlisted resource systems (to embedded roles androles relationships) and (3) ME specific workflowsthrough viable [process logic – resource systems]

couples The purpose of so doing is to optimise thechoice of resource system and methods of achievingworkflow control based mainly on cost and lead-timecriteria

Process routes, embeddedroles, op times, etc

Alternative assignmentsand organisational groupings

of human resources to roles Work entry points,

inter-arrival times, workflowcontrols

Relative process segmentbehaviours

Comparative qualitymeasures

Motivational factors andmeasures

Overall throughput, valuestream and cost measures

International Journal of Computer Integrated Manufacturing 199

executable models that predict dynamic

im-pacts on (current and possible future) process

performances (e.g lead-times, throughput,

bot-tlenecks, inventory, value generation and

pro-cessing costs) of alternative work patterns and

workloads;

(3) use documented and computer executable

models to suggest potential beneficial changes

to organisational structures, management

phi-losophies and culture;

(4) use integrated models to identify where

pro-cesses can become more lean or agile so that the

firm can gain business benefits;

(5) use integrated models to suggest ways of

improving the deployment and performance

of (human and technical) resources;

(6) use integrated models to improve the

planning, scheduling and control of workloads

placed on primary ‘operational’ process

segments

To realise the stated modelling goals, the present

authors adopted use of the integrated approach to

modelling human systems, as reusable components of

manufacturing workplaces; the modelling stages of

which are described by Table 1

5.3 Formal documentation of ComBear’s process

network

Stage 1 of the integrated modelling approach involves

the capture of a specific CIMOSA conformant

ComBear’s enterprise model The main purposes herewere to:

(1) provide the university team with means ofexternalising and reusing knowledge (formerlyonly distributed amongst the minds of variouspersonnel) about the firm’s ‘operational’ (day

to day), ‘tactical’ (sometimes daily or weeklyand sometimes episodic) and ‘strategic’ (longerterm) activity flows;

(2) understand how ComBear’s operational ity flows enable the company to generate shortterm values and profit, whilst remaining com-petitive in the medium and longer term;(3) assist ComBear management and workforce todevelop a big picture of the firm’s activity flows,

activ-so that individuals can identify impacts of theirroles on the business performance of thecompany; and

(4) enable the University team (as directed byComBear managers) to use the knowledgeexternalised as base level of company specificdata which enables the development of reusablecomputer executable models of selected activityflows (so as to realise goals (2) to (6))

Before embarking on CIMOSA modelling the presentauthors spent approximately 5 man days (i) discussingactivity flows with its managers and (ii) observingtechnical and manufacturing personnel perform theirvarious roles Figure 2 provides an overview of the fullrange of ComBear processes identified, namely:

Figure 1 ComBear product samples

Strategic processes that operate as required to

envision, conceive and realise improved

competi-tiveness through day to day management,

leader-ship, financial and fiscal policy management

and control, and adapting business rules and

manufacturing policies in response to changing

customer requirements, environmental and

government regulations;

Tactical processes resourced by ‘technical and

mid-management teams’ that : (1) obtain and

process customer orders, (2) develop process

plans and job cards for product manufacture, (3)

design new products or improve the design of

existing products, (4) control and manage

production materials, and (5) plan, schedule

and control production operations;

Operate processes that produce and deliver

composite bearings and other products

manu-factured in three shops located within the

production facility namely: (a) raw material

processing shop, (b) sand and saw shop and (c)

machine shop

Although causal, temporal and structural links were

observed between most of the key processes, the

chosen focus of case study modelling was on the

‘operate processes’; instances of which need to

regularly be resourced by human and technical

resources, so that products are realised for customersand ComBear profit is generated Four types ofmodelling template were used to describe ComBear’s

EM in a graphical form, namely: context diagrams,interaction diagrams, structure diagrams and activitydiagrams

Context Diagram:describe in overview how various

‘domain actors’ (i.e departmental sections and theirsupply chain partners) work together within thebusiness context under study In this case the contextmodelled was the day to day production of compositebearings, of types and in quantities needed to satisfyorders from a variety of customers Figure 3 shows anexample context diagram captured in respect ofComBear

Interaction Diagrams: describe various (relativelyenduring) entity flows between processes (which inCIMOSA terms are modelled as Domain Processes(DPs) and their elemental Business Processes (BPs),Enterprise Activities (EAs) and Functional Operations(FOs))

Structure Diagrams: depict relatively enduringstructural relationships between DPs, BPs and EAs.This class of diagram is designed to code specificprocess decompositions consisting of ordered set ofactivities linked by precedence relationships, execution

of which is triggered by events such as arrival ofcustomer’s orders

Figure 2 Key processes identified in ComBear

International Journal of Computer Integrated Manufacturing 201

Activity Diagrams: are used to depict specific

process segments of concern, in terms of standard

activity flows needed to create products Activity flows

were represented for each product family tured by ComBear Figures 4 and 5 are examples ofstructure and activity diagrams

manufac-Figure 3 ComBear top-level context diagram

Figure 4 ComBear interaction diagram

Other kinds of modelled entity (such as events,

information flows and precedence relationships) can be

attributed to activity flows In Figure 5, the activity

flows illustrated explicitly depict processing activities

carried out in ComBear’s raw materials processing

shop to create so called ‘round products’

5.4 Roles identification and specification

Having represented ComBear production processes

using standard CIMOSA formalisms: domain

pro-cesses (DPs), business propro-cesses (BPs), enterprise

activities (EAs), and functional operations (FOs) (see

Figures 3–5 (section 5)), step 2 of the integrated

modelling method was implemented by specifying and

defining viable roles executed by ComBear resource

system elements at various work centres The term role

was used to refer to functions performed by role

incumbents Three classes of role were identified in

ComBear namely: (a) management roles representing

those management and coordination functions, (b)

technical/support roles that realise functions such as

specifying process plans and procedures for products

manufacture, planning and controlling production,

designing new products; etc, and (c) operational roles,

which represent direct products realising functions

Matching human systems to operational roles is the

focus of this paper and those roles in ComBear

comprised: (1) raw material processing-related roles;that produced materials for making different ComBearproducts, (2) machining roles; that shape processedmaterials into components and where applicableassemble them into products, (3) sanding roles; thatput finishing touches to outputs from machining rolesand (4) packaging and delivery roles Figure 6 showssome roles identified in raw materials processing andsanding operations In satisfying the raw materialsprocessing functions for instance, five different roles(R1.11–R1.15), which are groupings of operationsexecuted at different work centres, were identified.When grouping operations into roles, considerationswere made about:

(1) precedence relationships between activities andtheir functional operations (bearing in mindclosely coupled and non-separable activities);(2) dependencies between possible candidate

‘as-is’ and possible ‘to-be’ role incumbents(e.g availability of people and supportingmachine resources)

In general, roles specified were explicitly defined interms of functional competences that potential candi-date solutions need to bring to realise those roles Itwas observed that ComBear’s operators performdifferent roles while realising specified activity

Figure 5 Activity diagram for making materials for round products

International Journal of Computer Integrated Manufacturing 203

instances Roles performed during activity instances

depend on the product types passing through the

process structure In the RMP shop for instance,

the roles played by operators depend on whether a

particular operator is processing round, flat or strip

materials and their variants

Subsequent sections of this paper describe how

ComBear’s process-role oriented models (exemplified

by the templates shown in Figures 11) were

succes-sively reused (and as needed further detailed and

modified) as an explicit big picture of ComBear

processing requirements that needed to be resourced

by suitable human and technical systems

6 ComBear human systems modelling

To realise step 3 of the modelling methodology

depic-ted by Table 1, actual (‘as is’) shop floor operators’

data were elicited and collected in terms of numbers of

people deployed to specific operation areas, roles

assigned to such groupings of people and their

individual known skills The aim of so doing was to

(re)use this employee data to facilitate an ongoing

matching of available human resources to specified

instances of ComBear product realising processes

While collecting the operators’ data, it was observed

that most shop floor operations performed by

Com-Bear personnel have no formalised written standard

operating procedures Rather it is left to supervisors

and operators to flexibly prioritise and execute theirwork Hence performance outcomes vary amongstindividual operators with respect to quality (‘fitnessfor purpose’), quantity of work done and the rate ofachieving specified quantity of work within acceptablequality standards It follows therefore that ComBearmanagement depend heavily on the commitment(related to behavioural, ‘will do’ competences) andwork-related (functional, can do) competences of itsworkforce to achieve its overall business goals

To enable the capture and conceptual tion of the competences available among ComBearemployees, currently deployed in the production shop,the following assumptions were made:

representa- to realise the ‘make bearings to order’ processsegment, competent people supported by suitabletechnical systems need to realise roles associatedwith ordered sets of processes that relate to eachproduct type manufactured by ComBear Essen-tially when customer orders are received for eachproduct type the realisation of the related roleswill constitute different (multiple) value streams; a suitable match needs to be realised andmaintained between required competences (expli-citly defined by descriptions of competencyrequirements that can be associated to eachrole) and available competences (that people andsupporting technology can bring to bear on

Figure 6 Combined Use of CIMOSA and role-modelling constructs to represent ComBear processes

specified roles) to ensure that prescribed work

outcomes are not compromised nor that the

people deployed will be stressed;

‘unlike technical systems, human systems can

reflect upon their task and job performance out

comes and thereby as necessary (1) develop new

competences or (2) develop new structures (and

possibly behaviours) that can enhance future

performance outcomes

Bearing in mind the key process groupings identified in

ComBear, namely; (a) strategic, (b) tactical and (c)

operational process types, a competence classification

previously proposed (Ajaefobi and Weston 2003,

Ajaefobi 2004) was deployed However a relatively

minor adaptation of this classification was needed to

encompass all ComBear workforce roles identified

in relation to the firm’s processes The classification

presumes that when models of ‘requirements i.e roles’

are specified they can be compared with ‘coherently

defined models’ of candidate solution systems (people

and their competences organised by suitable

struc-tures) to draw up a first stage selection of suitable

candidate solutions thereby satisfying step 3 of the

‘integrated approach to modelling human

system-s’outlined in Table 1 Thus with respect to the specific

ComBear process network defined during stage 1 of

the modelling approach and described in outline in

sections 5.3 and 5.4, it was necessary to observe and

characterise strategic, tactical and operational

competence types that personnel should possess tosatisfy specific roles (and their elemental activities).Furthermore, it was observed that current (‘as-is’)human systems at ComBear (and in a number of otherindustry partners doing collaborative research with theauthors) are structured into a three level hierarchy,namely: (a) management group, (b) technical group (c)and shop floor operational group These three group-ings of workers interact (in a combined top-down andbottom-up manner) so that collectively they execute allstrategic, tactical and operational aspects of processes.Figure 7 is a simple conceptual representation of theprocesses required and corresponding competencespossessed by ComBear personnel

As mentioned previously the major focus (in thiscase study) was on ‘operate processes’ and thecompetences they required for their realisation frompeople In the ComBear raw material processing shop(RMP), the following ‘general competences’ wereidentified through detailed discussions between Com-Bear managers and the present authors as beingneeded by operators for them to be consideredcompetent:

ability to read, interpret, and execute workingand technical drawings and to follow job cardinstructions to produce required parts orproducts;

ability to set up and operate available machinetools;

Figure 7 Conceptual representation of required and available competences

International Journal of Computer Integrated Manufacturing 205

ability to identify, measure and mix appropriate

chemicals required for specified composite

manufacture;

being quality conscious and able to get assigned

jobs right during their first attempt and where

necessary being able to fix up minor defects on

completed products thereby restoring them to

acceptable quality standards fit for purpose);

ability to work under stress (if need be) to meet

deadlines;

demonstrate commitment in relation to assigned

roles, prioritising actions and being flexibly ready

to work on related jobs within his/her

compe-tence area;

ability to demonstrate an understanding of

industry safety rules

In ‘specific roles’ and for ‘their product type variances’

the specific nature and importance of these general

competences was known to vary Therefore to resource

value streams that realise different ComBear products

in the RMP shop, combinations of these competences

(possessed by particular operators) need to be selected

and matched to required competences of the roles

Also such a match must be maintained and changed as

requirements change Selecting and maintaining an

appropriate match between required and available

competences is therefore an ongoing tactical task so

that neither prescribed outcomes with respect to

product quality, quantity, lead time, operating

proce-dures, etc are compromised nor are assigned operators

are overly stressed In addition to possessing one or

more of the general competences in varying degrees,

operators were observed to differ in their experiences

and in their potential to deploy so called tacit

know-ledge, which enables some operators to (1) flexibly

adapt to changing work requirements, (2) re-set

machine tools to meet changes in job requirements

and (3) deliver promptly in regard to assigned

jobs keeping to quality and quantity demands in

comparison with their peers In view of the observed

competence variations, operators were grouped into

three categories, namely: (a) experts, (b) practitioners

and (c) trainees The observed competence variations

were found later to provide a logical thread of reasoning

to underpin role assignment when matching alternativegroupings of activities to operators and/or groupings ofoperators In the case instance being reported, i.e in theRMP shop, operators were rated in three key com-petence areas with respect to their: (a) abilities to set upand use relevant machines and related tools in the shop,(b) procedural knowledge of operations, and (c) actualprocess execution and average processing time onassigned job Numerical values of 1, 2 and 3 (corre-sponding to trainee, practitioner and expert operatorsrespectively) were used to rate the Operators according

to their performances in relation to specified reachablestates Table 2 shows ratings assigned to one of theexpert operators

7 Enhancing the re-usability of ComBear models usingDPU and role modelling concepts

Previous sections have described how CIMOSA andcompetence modelling concepts were used to captureand represent processes and human systems deployed

at ComBear This section discusses how the modelsdeveloped were semantically further enriched via theuse of DPU and role modelling concepts Essentially,DPUs are viewed as component entities that can beconfigured to form MEs via engineering processes thatinclude selection, structural and temporal confi-guration, programming and run-time interaction andinteroperation thereby producing desired outcomes.Notions about DPUs presume that at some level ofabstraction, enterprise components can be considered

to be ME modules However linked to this notion isthe fact that some ME components can be decomposedinto DPU subsystem units, so that their modular unitscan be recomposed differently to realise (changing)targeted behaviours; and by so doing organisationdesign and change can be engineered in a model drivenmanner Another way of viewing ME components isthat they have the ability to act individually and tointeract collectively as building blocks (whether theyare machines or modular units of machines, IT systemsand/or human resource systems) to achieve desiredbehaviours and therefore desired goals Suitable ME

Table 2 An expert operator’s rating

Operator’s identity Operations/work centres

Operators competence rating

Potential competenceefficiency (PCE)Basic (1) Practitioners (2) Experts (3)

components (resource system elements) with potential

to generate high performances when matched to

a specific network of processes constitutes a viable

candidate DPU Modelling ME components can

enhance understandings about their characters,

beha-viours, applications and change The DPU concept

was proposed by the present authors and their colleges

as a modelling construct which can be re-used in a

virtual environment to enable enterprise components

(especially active resource systems: namely humans,

machines and IT which can realise specified roles)

to be described coherently and explicitly as ‘reusable’,

‘change capable’ ‘components’ of MEs Thus, DPU

characterisation is designed to facilitate: (1) graphical

representation of resource systems, (2) explicit

specifi-cation of resource systems and (3) implementation

description of resource systems, such that DPUs can be

computer executed within simulation modelling

envir-onments (Weston, et al 2009) It is assumed that (1)

DPUs can function individually as a holder of one or

more assigned roles and (2) configurations of multiple

DPUs will interoperate so as to function collectively as

holders of one or more higher level (more abstract)

roles (i.e roles composed of lower level roles) Further,

depending on their working context, DPUs are

expec-ted to be explicitly defined in terms of recognisable

work-related attributes: i.e various types of

compe-tences (for human systems) and/or capabilities (for

machine and computer systems) Consequently,

de-pending on specifics of any work context and the

nature of work deliverables, viable candidate DPUs

should possess the abilities to produce identifiable,

measurable and observable outputs; so that for

exam-ple a DPU comprising of an operator and machine can

realise both product X and Y in specified quantities,

and with specified quality levels within specified cost

and time constraints Furthermore, for DPUs to

possess the competences and capabilities to cater for

changing work contexts and work deliverables, they

must also possess a competence to be re-programme

and/or re-configured as needed It follows that DPUs

have dynamic nature, which means that DPUs need

to be modelled in terms of both relatively enduring

structures and time dependent behaviours; so that

their model parameters can be adjusted to judge in

simulation environments their potential performances

in ‘Key Performance Indicator (KPI)’ terms such

as lead time, efficiency, utilisation, throughput and

operational flexibility

From the above description of DPU concepts, it

was decided that ComBear human operators and their

supporting tools and technology would be modelled as

DPUs To represent ComBear resource system

config-urations (using DPU concepts) in the process-oriented

roles they must realise, use of the CIMOSA EM

framework proved effective via its inherent separation

of ‘process-oriented requirements capture’ from based conceptual designs of resource system solutions’

‘DPU-By so doing ComBear resource systems were modelledseparately as solution units that on an on-going basiscan be matched to previously modelled process-oriented roles To facilitate formal capture, documen-tation, modelling and matching of DPUs to process-oriented roles, the unified modelling language (UML)was deployed UML was used to encode (1) DPUattributes and (2) process oriented roles This enablesreuse of information structures and informationentities related to the decomposition of specific net-works of processes Furthermore, the use of UMLprovided a flexible means of representing and structur-ing groupings of ComBear resource system compo-nents, modelled in terms of DPUs and groupings ofDPUs When creating models of the ‘as-is’ ComBearresource systems in UML, the entire ComBear work-force was modelled as a high level ‘DPU class’ which isreferred to as a ‘Human Systems Model DPU’ ThisDPU class is comprise of three subclasses; respectivelyrepresenting ComBear employee groupings that pos-sess the kinds of competences needed to execute one ofthe identified key processes, namely:

DPU1 representing the management and day today administrative group;

DPU2 representing the mid-management andtechnical group;

DPU3 representing the entirety of ComBearproduction operatives

Instances and objects that constitute the above DPUsubclasses were further modelled and their attri-butes documented Figure 8 is a detail illustration ofDPU3; which encodes ComBear’s employees classesand their instances responsible for executing the

‘produce composite bearing products to order’ cesses and its derivative roles

pro-The use of UML formalisms semantically enrichedthe graphical representation and detailed structuraldocumentation of ComBear’s DPUs, their work-related attributes and relationships between thoseattributes Furthermore, the previously identified pro-duction processes modelled in terms of roles; (whereroles denote related operations and groupings ofoperations that are executed at distinct work centres

by one or more DPUs) were also documented usingUML formalism Here UML conformant ‘use casediagrams’ were created to document and depictprocess-oriented roles and corresponding DPUs ex-ecuting them DPUs were modelled as ‘BusinessWorkers’ (a role-based concept in UML use to captureand denote groupings of employees that perform rolesInternational Journal of Computer Integrated Manufacturing 207

specified in the ‘use cases’) along with the

responsi-bilities of the workers and the competences they bring

to those roles Figure 9 shows one of the use case

diagrams for a segment of ComBear production

processes; this flexibly links role 1 (modelled as an

exemplary activity flow diagram) for the ‘process raw

material’ use case

By means UML modelling constructs, elicited

work-related attributes of candidate human resources

were formally documented; in terms of competency

descriptions of individual and groups of DPUs and

their associated performance levels, as experts,

practi-tioners or trainees Similarly, roles were explicitly

defined in terms of required competences, i.e

compe-tences that viable candidate operators need to possess

to satisfy role requirements By so doing, explicit

descriptions of viable resource systems were matched

to explicit descriptions of roles, thereby forming

different Role-DPU couplings

However, it was observed that the dynamics of

work item flows have very significant impact on needed

competences and capabilities of DPUs; and therefore

on how these might best be configured to realiserequired product outputs on time and at an acceptablecost Furthermore, at this stage it was also observedthat selected configurations of Role-DPU couples werestatic; in the sense that they only describe relativelyenduring relationships between ‘work types andwork flows’ and ‘potential workers’ It follows that atthis stage (of modelling), the likely performances ofRole-DPU couples under varying work conditions(changing product volumes and variance) were yet to

be established

In the reality, however, customers’ changingrequirements impact significantly (in terms of reflectedworkloads placed) on roles and competences needed tofulfil such roles In principle also, human performance(be they competent individuals, groups or teams ofpeople deployed to realise predictable outcomes)can be measured in quantitative terms in order tocharacterise and rank impacts of (average or stochas-tic) system behaviours, such as lead time, quantity andquality of work, costs, etc Hence it was decided thatthe static competence model that encodes operators as

Figure 8 Using UML constructs to represent ComBear’s DPUs

experts, practitioners and trainees and their assigned

roles needed to be computer exercised via a simulation

modelling tool to replicate existing and when resultant

models have been validated to re-use them to predict

possible future performance outcomes when deploying

different operators under changing requirements

8 ComBear simulation models: dynamic views

8.1 Causal loop models (CLMs)

As a precursor to building dynamic models of

ComBear’s operational processes, the authors had a

brain storming session with ComBear management

during which causal and temporal dependencies

be-tween ME components and subsequently modelled

The brain storming exercise was facilitated using CLM

diagrams which were designed to enhance the

under-standing of causal relationships, and resultant

beha-viours, amongst common deployed ME system

components Figure 10 shows one of the causal loop

models created for this purpose depicting causal effects

on the rate of doing work; including expected impacts

of increased competences and capabilities from

de-ployed resource systems

Conventionally though, CLMs have been used as a

front end to the use of continuous simulation

technologies as they naturally provide a systemic way

of creating ‘stocks’ and ‘flows’ modelled in SMs of this

type But in the ComBear case, the authors chose to

use CLM to enable ‘front end’ thinking prior to use of

a discrete event simulator This choice of discrete event

simulation was made on technical grounds in that theComBear problem under study concerned multiplevalue flows with dynamic production volumes andmixes This necessitated the use of time dependentmodels of processing operations and related processingoperators and machine configurations, and relatedbottleneck and inventory analysis The need to modelindividual and mixes of product flows, subject toprobability was paramount Also it was necessary toevaluate key performance measures such as productionlead-times, resource utilisation, inventory levels, pri-cing, costs and value generation These and otherparameters needed to be predicted with differentiationmade when alternative resource configurations aredeployed and when different value flows are beinggenerated Therefore, a conscious choice was made todeploy proven and proprietary simulation technology(rather than research prototypes) with a view tosuggesting in the longer term decision-support toolsthat large and small companies can practically deploy

8.2 Dynamic modelling of selected ComBearprocesses

Having developed static models of ComBear processes,and considered the static matching of candidate humansystems to role requirements and discussed the nature ofcasual interactions between different ME components,the next step taken was to analyse the behaviours ofalternative human system configurations under differingworkload conditions This corresponds to stage 4 of

Figure 9 Examples of ComBear use case and activity diagrams

International Journal of Computer Integrated Manufacturing 209

modelling approach outlined by Table 1 The purpose

of simulation modelling (SM) was (1) to replicate

historical behaviours of the RMP section of the

production shop, so as to verify the correct operation

of the models created and to gain new insights in the

current operations performed in the RMP section and

(2) to predict possible future behaviours of alternative

Role-DPU configurations and (3) make suggestions

about possible future re-configurations of (human

and technical) resource systems, so as systemise and

quantify possible future outcomes should the changed

configurations be adopted

To help structure the design of the SMs,

stake-holders knowledge, structural relationships and data

previously coded into ComBear’s EM and current

DPU-Role couplings were reused Also understandings

generated from the CLMs were used to determine

which parameters of the SM needed to be experimental

variables when modelling from alternative points of

view Based on all these understandings, discrete event

simulation models of ComBear’s RMP shop were

designed and created using Simul81

With a view to realising specific ComBear

model-ling objectives listed in section 5.2, the RMP activities

(for each of the identified value streams) were coded as

elements of work to be realised at work centres of the

simulation models One key purpose of so doing was to

observe behavioural implications of interactions

be-tween the value streams and thereby key impacts of

sharing resource systems (i.e lower level DPUs); so as

to predict resultant throughputs and DPUs utilisations

as a result of such interactions and overall impacts ondownstream ‘Sanding and Machine’ shops; wherefurther value adding operations are carried out Initialobjective use of the simulation model was to investigatecapacity constraints of the RMP shop DPUs, as atimely supplier of requirements of its downstreamshops (sanding and machine shops) Throughputs

in the RMP shop were observed to be a largelypredictable function of the operators’ competences,including rate of working, flexibility and use of tacitknowledge Furthermore, apart from the ‘mixingbooths’ (there are two booths in the shop that areshared with other operators needing them) no othersignificant bottle necks were observed; as regards to theavailability of work centres to be used by the operators

as they move from one processing stage to anotheralong the value streams The modelling results pre-dicted that with the ‘as-is’ process and resource systemconfigurations and their workloads, an average opera-tor can make 4 or 5 flat sheets per day Similarly forstrip sheet and round products, daily operator’soutputs were predicted to be 3 to 4 strip sheets and

65 tubes per day respectively In terms of averageoperator utilisation, it was observed from the modelthat operators had a 85% utilisation for flat sheetmaking, a 67% utilisation for strip sheet productionand a 94% utilisation for round products Here thepercentage utilisation indicates the actual availability ofthat resource for the work In the case being reported, it

Figure 10 Exemplary CLM depicting causal interactions between common ME components

implies for instance that the operator was available for

only 85% of the total value adding time

Bearing in mind that useful value adding time (in

ComBear) is 426 min per day, it follows that on

average; an operator spends more than 1 h on

non-value adding operations per day, which other things

being equal, can be interpreted as waste To account

for the non value adding times observed, further

investigation sought to determine non value adding

(v) Sharing of people on other activities;

(vi) Operators doing just the volume of job

planned for the day i.e planned job is less

that operators’ capacity

It was observed that only factors (v and vi) above

impacted on the daily throughput of the operators and

none of (i–iv) really applied Most of the value adding

operations in raw material processing are essentially

manual, executed on ‘baths and work benches’, hence

changeovers with respect to tools do not necessarily

apply The distance travelled by operators was

considered negligible because the work centres were

basically adjacent to one another Further, materials

were promptly delivered and were readily accessible

and no breakdown of work centres was observed

during the investigation This leaves process

improve-ments (including lead time reduction, shortening of

Takt time, increasing throughput, and quality of work

done) or improving operators’ competences, including

aspects of competences such as flexibility, working

speed, tacit knowledge, etc that impact on performance

outcomes

The authors considered ways to rectify the

ob-served short comings in throughput and operator

utilisation by constructing possible ‘to be’ scenarios of

improved performances To achieve this, the roles that

constitute the identified value streams in RMP shop

were regrouped and then simulated under two different

conditions: (a) with increased work entrance and (b)

using the cycle times of the expert operators, instead

of the average operators’ cycle times The resultant

outcomes showed an improvement in both

through-puts and in the operators’ utilisations Throughput by

an average operator was raised from 5 to 8 units

of flat sheet per day and the operator’s utilisation

increased to 94% Similar improvements were

ob-served with respect to strip sheets and round products

value streams It follows that the simulation

experiments predicted that training the operators toimprove on their competences (thereby acquiring the

so called ‘expert competences), can improve theproduction system performance

Furthermore, to illustrate the impact of improvedavailable competences on throughput and timeliness

of ComBear products, a second dynamic model wascreated The CLM of Figure 10 illustrates how increase

in the number of competent operators will naturallyincrease available competences, which other thingsbeing equal will trigger increased rates of doing work,improved operational flexibility, economies of scaleand increased value generation These features weredemonstrated by the second simulation model usingthe RMP shop as a case instance It was observed thatapart from the ‘expert’ operators, (supervisors) whocan flexibly do any job realised by ComBear, otheroperators were narrow in their competence scope,being unable to work outside their operational areas.The implication is that operators cannot flexibly bedeployed to areas of bottleneck outside their corecompetence area In the second simulation model,assumption made was that all the operators canflexibly pick any job from the three identified valuestreams and competently deliver the prescribed outcome effectively and on time Based on this assump-tion, the average cycle times taken by the experts(supervisors) to execute value adding activities weremeasured for any of the three value streams Theexpert cycle times were then presumed to be standardoperation times for all RMP roles The purpose ofdoing so was to observe and compare the differences

in outputs when two different operation times i.estandard operation times (based on the performance ofaverage operators) and the experts’ operations timesare used in the model The results obtained by runningthe model showed: (i) improved lead times, (ii)increased throughputs and (iii) higher operatorsutilisations when compared with the ‘as is’ scenario;whereby all the operators work in their areas ofspeciality and with their existing competences andperformance levels In making raw materials for theflat sheet products for instance, the throughput showed

an increase of 30% while operator’s utilisation (whichwas previously 85%) rose to 100% Furthermore, jobwaiting times were reduced and jobs in the queue alsoreduced by 32% It follows that in non high techproduction system such as ComBear, investing onpeople to improve their competences and so called tacitknowledge will’ other things being equal’ significantlyimprove the over all system performance Table 3shows a performance comparison of two operatortypes in an ‘as-is’ and one of the possible ‘to-be’scenarios while Figure 12 is a screen shot of ‘as-is’ flatsheet materials making operations in the RMP shop.International Journal of Computer Integrated Manufacturing 211

Another factor that impacted on the throughput

and thereby the utilisation of the operators was the

capacity of the press (oven) and the minimum time a

work item spends during curing (in the oven) before it

is released The oven has a capacity of 6 flat sheets and

a processing time of 90 to 120 min However, the

simulation view predicted that some work items spend

more than 2 hours during curing Though this does not

affect the quality of the job, it will have some impact

on the daily throughput When a minimum curing time

was set at 90 min, a predicted throughput of 7 to 8

sheets per day per operator was obtained while

maintaining an average of 110 min oven curing time

From the foregoing discussions, it follows that the

weekly throughput of processed flat sheets materials

can in theory be increased to a weekly average of 30 to

40 units per operator contrary to the current observed

20 to 25 units Though currently, the operators can

be said to be performing below their capacities, asevidenced by the non value adding times observedespecially with respect to strip and flat sheets makingoperations, it was observed that the down streamshops were not in short of strips and flat sheets suppliesbecause of the defacto practice in ComBear whichallows the build up of much stock and large WIP Thispractice carries with it extra production costs, risksassociated with large WIP and general shop flooruntidiness In the short term, an approach using 5Sprinciples to cutting down wastes and improving shopfloor orderliness developed by the authors is beingtested in ComBear while in the long term, a full leanmanufacturing implementation has been proposed

Table 3 Performance comparison between average and expert operators

Operator type Product types

Averagedailyoutput

Operatorsutilisation %

Operator

Averagedailyoutput

Operatorsutilisation %Average operators

(standard cycle

times)

(expert cycle times)

Figure 11 ‘AS-IS’ average operators’ utilisation

9 Conclusions and ongoing research

ME managers and developers require much improved

analytical means of deploying key and scarce resources

including people, IT systems and machines This need

is growing because of increasing workplace dynamics

Generally also state of the art enterprise engineering

methods are techno centric and do not account for

significant behavioural differences between human

and technical resources This paper introduces an

approach to modelling human and technical resources

coherently but distinctively, which builds upon a

unification of best in class EM and SM frameworks,

methods and tools This unification has been achieved

by defining and developing the use of integrating

modelling concepts, centred on ‘role’, ‘DPU’ and

‘competence’ modelling This first part of this paper

explains how the unifying concepts chosen have built

upon a complementary use of CIMOSA enterprise

modelling principles Subsequent paper parts describe

a case study application in an SME and the lessons

learned

The authors observed that great benefit was

obtained from combining the use of EM, SM and

CLM to conceive and analyse alternative futures for

ComBear Use of EM has benefited as follows

(i) It has organised the capture of complex

knowledge about ComBear processes,

struc-tures, resources and workflows in a virtual

manner which ComBear managers can readily

understand and verify as being representative

of how the firm actually operates;

(ii) The same EM has provided a ‘static’ (processoriented) big picture of the company and itssupply chain which has been reused manytimes and updated and developed as newunderstandings about ComBear were devel-oped Via use of the embedded EM decom-position technique it has broken downComBear’s big picture into sets of lower level(more detailed) requirements descriptions foroperations (job, tasks and activities) that need

to be performed by that firm Also keytemporal and causal relationships betweenthe decomposed process segments have beenexplicitly modelled;

(iii) Multiple value flows have been overlappedonto the big picture of ComBear provided bythe EM This has allowed differentiationbetween different product classes to be estab-lished Also it has enabled reasoning aboutalternative flow rates for different valuestreams and the loads they place on differentoperational sequences (or process segments);(iv) ComBear’s EM also provides an overarchingstructural framework and pool of reusableknowledge and data which supports the designand development of any number of lower level,more focused SMs In this way the context ofsimulation modelling is explicitly defined in

Figure 12 Screen shot of ‘as-is’ flat sheet materials making operations

International Journal of Computer Integrated Manufacturing 213

terms of static requirements and alternative

viable resource system candidates;

(v) The structural framework provided by

Com-Bear’s EM is currently being used to organise

alternative viable configurations of SMs, so

that their combined interoperation and

per-formances (as a configuration of DPUs) can be

judged in the light of context dependent and

dynamic business and environment

require-ments faced by ComBear today and possibly

into their future

Because of space limitations and the focus of concerns

herein, beneficial use of the CLMs has not really been

illustrated in this paper However significant benefit has

been gained by using the EM in combination with

various CLMs to (a) consider how changes in product

dynamics (e.g product orders, production volume

variation and production mix variation) impact on

production system performance requirements, and

thereby on human resourcing requirements and (b) as

a front end for designing SM experiments, by helping to

identify possible control, controlled and likely causal

impacted variables in different ComBear production

shops, and thereby enabling the construction of

well-engineered simulation experiments The systematic

method proposed has many potential uses in

Enter-prises that are subject to ongoing change in product

types and quantities realised Particularly it can help to

avoid inappropriate and risky change engineering such

as by supporting: short and medium term planning;

production system value and cost analysis; production

due date and cost estimation; the management of

information about human competences possessed by

specific Enterprise; new manufacturing paradigm

selec-tion and analytical design; and factory design and

investment planning However within a single paper

and for a single case study only limited illustration of

these purposes has been possible Other purposes will

be reported in subsequent journal papers

References

Ajaefobi, J.O and Weston, R.H., 2003 Modelling human

systems in support of process engineering In: Zulch Gert,

et al., eds Integrating human aspects in production

management Heidelbery: Springer, 3–15

Ajaefobi, J.O., 2004 Human systems modelling in support of

enhanced process realisation Thesis (PhD)

Loughbor-ough University

Ajaefobi, J.O., Weston, R.H., and Chatha, K.A., 2006

Predicting the performance of candidate human systems

within specific work contexts International Journal of

Computer Integrated Manufacturing, 19 (7), 651–662

AMICE ESPRIT Consortium, 1993 Open system

architec-ture for CIM, 2nd revised and extended edition New

York: Springer

Ashfort, B.E., 2000 Role transition in organisational life: Anidentity based perspective Mahwah, New York: Lawr-ence Erlbaum Associates

Ashkenas, R.N., 1995 The boundaryless organization: ing the chains of organizational structure San Francisco:Jossey-Bass Publishers

break-Bennis, W., 1996 Changing organisations New York:McGraw Hill

Boyatzis, R.E., 1982 The competent manager: a model foreffective performance New York: Wiley

Chatha, K.A., Ajaefobi, J.O., and Weston, R.H., 2007.Enriched multi-process modelling in support of the lifecycle engineering of business processes IJPR, 45 (1),103–141

Harzallah, H and Vernadat, F.B., 2002 IT-based tency modelling and management: from theory topractice in enterprise engineering operations Computers

compe-in Industry, 48, 157–179

Hendrick, H.W., 1997 Organisational design and ergonomics In: G Salvendy, ed Handbook ofhuman factors and ergonomics New York: Wiley, 595–637

macro-Kosanke, K., 1995 CIMOSA overview and status ters in Industry, 27, 101–109

Compu-Krappe, H., Rogalski, S., and Sander, M., 2006 Changes forhandling flexibility in the change management process ofmanufacturing systems Proceedings of the IEEE Inter-national Conference on Automation Science, Shanghai,China, pp 551–557

Medsker, G.J and Campion, M.A., 1997 Job and teamdesign In: G Salvendy, ed Handbook of human factorsand ergonomics New York: Wiley, 450–489

Mintzberg, H., 1989 Mintzberg on management: insideour strange world of organizations London: ColliarMacmillan

Pandya, K.V., Karlsson, A., Sega, S., and Carrie, A., 1997.Towards the manufacturing enterprises of the future.International Journal of Operations and ProductionManagement, 17 (5), 502–521

Skyttner, L., 2005 General systems theory, problems,perspectives and practices Singapore: World Scientific.Steers, R.M and Black, S.J., 1994 Organisational behaviour.New York: HarperCollins

Swarz, R.S and DeRosa, J.K., 2006 A framework forenterprise systems engineering processes 19th Interna-tional Conference of Systems and Software Engineering(ICSSEA), Paris

Vernadat, F.B., 1996 Enterprise modeling and integration:principles and applications London: Chapman and Hall.Wagner, J.A and Hollenbeck, J.R., 1992 Management oforganisational behaviour New Jersey: Prentice Hall.Weston, R.H., 2005 Unified modeling of complex systems tofacilitate organisation design and change EPSRC PartA: Case for support MSI Research Institute, Loughbor-ough University

Weston, R.H., Rahimifard, A., Ajaefobi, J.O., and Cui, Z.,

2009 On modelling reuseable components of capable manufacturing systems Journal of EngineeringManufacture, Proceedings of IMechE, Part B, 223, 313–336

change-Weston, R.H., Byer, N., and Ajaefobi, J.O., 2003

EM in support of team system engineering In:

R Jardim-Goncalves, et al., eds Proceedings of 10thISPE International Conference on Concurrent Engineer-ing: Research and Applications Lisse: Swets andZeitilinger

Weston, R.H., Chatha, K.A., and Ajaefobi, J.O., 2004.

Process thinking in support of system specification and

selection Advanced Engineering Informatics, 18, 217–229

Weston, R.H., Guererro, A., and Chatha, K.A., 2007

Process classes deployed in manufacturing enterprises

IJCIM, 20 (6), 505–523

Womack, J.P and Jones, D.T., 2003 Lean thinking: banish

waste and create wealth in your corporation New York:

Free Press

Womack, J.P., Jones, D.T., and Roos, D., 1990 The machine

that changed the world New York: Rawson Associates

Zaidat, A., Vincent, L., and Boucher, X., 2004 Role and

processes based modelling approach for organisation

network In: L.M Camarinha-Matos, ed Virtual

en-terprises and collaborative networks Dordrecht: Kluwer

Adademic, 35–42

Zelm, M., Vernadat, F.B., and Kosanke, K., 1995 TheCIMOSA business modelling process Computers inIndustry, 27, 123–142

Zhen, M and Weston, R.H., 2006 Simulating dynamicbehaviours in complex organisations: case study applica-tion of a well structured modelling approach In:

T Nketsa, et al., eds Proceedings of European Simulationand Modelling Conference, Toulouse, France, EUROSIS-ETI, pp 390–395

International Journal of Computer Integrated Manufacturing 215

XML-based neutral file and PLM integrator for PPR information exchange between heterogeneous

PLM systemsSang Su Choia,b, Tae Hyuck Yooncand Sang Do Nohb*

a

Korea Institute of Industrial Technology, 35-3, Hongcheon-ri, Ipchang-myun, Seobuk-gu, Cheonan-si, Chungnam, 331-825, Korea;b

Department of Systems Management Engineering, Sungkyunkwan University 300 Cheoncheon-dong, Jangan-gu, Suwon,

Gyeonggi-do, 440-746, Korea;cProduction System R&D Team, Daewoo Shipbuilding & Marine Engineering Co., LTD, 1, Aju-dong, Geoje-si,

Gyeongsangnam-do, 656-714, Korea(Received 25 September 2008; final version received 28 October 2009)PLM (product lifecycle management) is an innovative manufacturing paradigm which allows company’s engineeringcontents to be developed and integrated with all business processes in the extended enterprise throughout theproduct lifecycle This allows engineering decisions to be made with a full understanding of the product and itsportfolio, including processes, resources, and plants For today’s manufacturing industries, support from softwaresystems is essential for the creation, management, and coordination of all manufacturing-related information.Generally, PDM (product data management) systems are extended to PLM systems to manage all PPR (product,process, resource) information throughout the product lifecycle

To implement PLM in an environment where many different information management systems are being used, anapplication to exchange data, files, and information between heterogeneous PLM systems is essential, and a standard fileformat for the exchanges is extremely desirable In this paper, PPRX is defined as a standard data format using XML for

a neutral file containing PPR information, and PLM integrator, which supports PPR information exchanges betweencommercial heterogeneous PLM systems and other systems is developed With the proposed implementation and a casestudy for an automotive company, exchanges of PPR information can be made without loss, which reduces unnecessaryeffort and helps successful implementation of PLM by supporting effective integration and information sharing

Keywords: PLM (product lifecycle management); PDM (product data management); PPR (product, process,resource); XML (eXtensible markup language); neutral file; PLM integrator

1 Introduction

PLM (product lifecycle management) is an innovative

manufacturing paradigm which leverages e-business

technologies to allow company’s engineering contents

to be developed and integrated with the company’s

business processes in the extended enterprise

through-out the product lifecycle This allows engineering

decisions to be made with a full understanding of the

product and its portfolio including processes,

re-sources, and plants (Siemens PLM Software 2003)

PLM extends PPR (product, process and resource)

content and knowledge to other enterprise processes by

coupling e-business technologies with engineering

applications focused on product development and

manufacturing Before PLM, applications such as

CAD (computer-aided design), CAM (computer-aided

manufacturing) and CAE (computer-aided

engineer-ing) were somewhat independent from the enterprise

mainstream (Shoaf 2000) Although PLM emerged

from tools such as CAD, CAM, and PDM (product

data management), it is necessary to understand PLM

as the integration of these applications throughout theproduct lifecycle

The core of PLM is the integrated management ofall PPR-related engineering data and the technology toaccess and utilise this information, so most companiesfocus on integrating and managing PPR informationfor successful PLM Generally, a company alreadyuses many different information management systems,

so one serious obstacle to PLM is the informationexchange problem (Shoaf 2001) To implement PLMwith many different information management systems,

an application to exchange data, files, and informationbetween heterogeneous PLM systems is essential, and astandard data file format for PPR informationexchange is extremely desirable

In this paper, the standard data format for PPRinformation using XML (eXtensible Markup Lan-guage) is define, and develop the PLM integratorwhich supports PPR information exchanges betweencommercial heterogeneous PLM systems and othersystems Also, the proposed system is implemented,

*Corresponding author Email: sdnoh@skku.edu

Vol 23, No 3, March 2010, 216–228

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

Ó 2010 Taylor & Francis

DOI: 10.1080/09511920903443234

and applied to practical examples of an automotive

shop as a case study

2 Related work

Most enterprises implement PDM systems because they

have problems with information management and

exchange between existing heterogeneous CAD, PDM,

and EDM (Engineering Data Management) systems In

the 1980s, many countries proposed a standard neutral

format for exchanging geometric information between

various CAD systems Moeck suggested exchange of

technological data via IGES (Initial Graphics Exchange

Specification) which was developed in the USA in the

late 1970s (Moeck 1990) It is now the most widely used

format for CAx data exchanges (NIST 1996) Zhang

et al proposed the STEP data exchange framework for

virtual enterprises (Zhang et al 2000), and Sun et al

proposed the Data Exchange System which provides the

underlying support for data exchange and data

main-tenance in integrated building design systems using an

object-oriented database and ISO-STEP technologies

(Sun and Lockley 1997) The actual designation of

‘STEP’ is ‘ISO 10303; Industrial automation systems –

Product data representation and exchange’ (STEP

2007) In the 2000s, STEP-Compliant NC was

devel-oped as a common model and language for CAD,

CAPP, CAM, and CNC which integrated and

trans-lated knowledge among CNC machine tools, vendors,

and users This included a new NC data interface called

‘ISO 14649; Industrial automation systems – Physical

device control – Data model for computerised

numer-ical controllers’ (International Organization for

Stan-dardization 2000) Gong et al suggested an XML-based

heterogeneous information processing and integration

method for enterprise management information

sys-tems They proposed a new XML schema, XSLT, to

effectively describe and integrate heterogeneous

infor-mation (Gong et al 2007) Lou et al performed research

on ontology-based integrated product modeling for

multidisciplinary collaborations They suggested the

concepts of a master model and reusing data from other

domains in product development, and implemented a

framework for integration (Lou et al 2007)

PSL (process specification language) is the

interna-tional standard for exchanging information about the

manufacturing process PSL defines a neutral

repre-sentation of manufacturing processes that supports

automated reasoning These process data are used

throughout the life cycle of a product, from the early

stages of the manufacturing process during design,

process planning, validation, production scheduling,

and control (Schlenoff and Gruninger 2001) In 2001,

Burkett presented PDML (product data markup

language), which is a new paradigm for product data

exchange and integration (Burkett 2001) PDML is aset of XML vocabularies and a usage structure fordeploying product data on the internet by making itvisible to the DoD (Department of Defense) weaponsystem support personnel PDML offers a new para-digm for product data exchange based on existingtechnology that facilitates the integration and inter-operability of business processes across the DoD andcontracting organisations, particularly among PDMsystems used for process control and product datamanagement (Borgman et al 2002) In 2004, Park et al.presented FeatureML which is a neutral file based onXML for exchanging CAD data (Park et al 2004).Siemens PLM Software, one of the world’s largestPLM software and service providers, has developedPLM XML which is an emerging format for facilitat-ing product lifecycle interoperability using XML PLMXML provides a lightweight, extensible and flexiblemechanism for transporting high-content product dataover the internet, and aims to form the basis of a richinteroperability pipeline connecting Siemens PLMSoftware products and third party adopter applica-tions (Siemens PLM Software 2008) PLMXMLconsists of more than ten kinds of independent schema,including PDM and MPM schema In addition to thebasic schema used for product information, thePLMXML provides an extensible and flexible mechan-ism to handle diverse information But mostPLMXML applications focus on product information,and deal with manufacturing process and resourceinformation as part of the non-geometric information

of product Though the PLMXML schema is open,there are limitations in developing PLMXML applica-tions without the use of commercial software such asPLMXML SDK So far, the PLMXML SDK Toolkitmainly supports product information, including pro-duct structure and geometry information

Most previous researches have focused on themanagement or exchange of product information, andthis exchange is from the perspective of informationmanagement of product geometry rather than from theperspective of PLM Therefore, the problem needs to

be addressed from the perspective of PLM, mostimportantly in the aspects of the efficient managementand exchange of product, process, and resourceinformation Moreover, so many systems currentlyexist, research on the methodology of informationexchange and the definition of a standard should bedone to combine the different systems In 2004, Choi

et al developed data exchange middleware whichintegrates the PLM system and a virtual engineeringsystem (Choi et al 2009) The paper describes only theexchange of product information, and mentions thenecessity of extending their basic model to the PPRmodel

International Journal of Computer Integrated Manufacturing 217

In reality, even if various standards are suggested for

PDM data exchange, it will be difficult to apply the

current standards due to the absence of a system which

supports the standards A one-by-one re-presentation of

the information according to defined standards is not a

simple job, and so is not worth the enormous cost and

time needed to conform to the standards This paper

defines a neutral file which provides a more efficient way

to manage and exchange PPR information In addition,

a PLM integrator is developed to supports the exchange

of PLM data by applying this neutral file

3 PPRX, a neutral file for PPR information exchange

3.1 PLM services

In this paper, an XML-based neutral file is defined by

referring to PLM services, which is an international

standard for exchanging product and related information

PLM Services is an Object Management Group (OMG)

specification for the exchange of product lifecycle data

using Web-service technologies, and it was developed by

the Extended Product Data Integration (XPDI) taskforce

of the ProSTEP iViP association in April 2004

Figure 1 shows properties and sources of PLM

Services 1.0

Here, a Web-service is a software system designed

to support interoperable machine-to-machine

interac-tion over a network It has an interface described in a

machine processable format such as WSDL (Web

Services Description Language) Other systems

inter-act with the Web-service in a manner prescribed by its

description using SOAP (Simple Object Access

Proto-col) messages, typically conveyed using HTTP (Hyper

Text Transfer Protocol) with XML (eXtensible

Mark-up Language), serialisation in conjunction with other

Web-related standards (Seely 2001, W3C 2004)

The specification of the PLM Services 1.0 defines a

Platform Independent Model (PIM) for Product Lifecycle

Management Services Its informational model is derived

from the ISO 10303-214 STEP model by an EXPRESS-X

mapping specification and an EXPRESS-to-XMI

map-ping process The functional model is derived from the

OMG PDM Enablers V1.3, and the selected scope of the

information model was chosen based on the requirement

analysis of the PDTnet project The specification defines a

Platform Specific Model (PSM) applicable to a

Web-services implementation defined by a WSDL specification,

with a SOAP binding, and an XML schema specification

(Feltes et al 2002, Lukas and Nowacki 2005, Object

Management Group 2005)

3.2 Definition of the neutral file (PPRX)

An XML based-neutral file for the management and

exchange of PPR information is defined by referring to

PLM Services in this paper However, PLM Servicesare inflexible because they include unnecessarilydetailed information which is not essential for infor-mation exchange between PDM systems, so this paperdefines the neutral file by referring only to the structureessential for PPR information exchange Importantaspects of the neutral file definition are:

(a) Flexibility: Easy and extensible modification ofthe neutral file is essential because informationexchange and modification in a PDM systemare frequent;

(b) Independence: In all systems, the neutral filemust be defined independently to supportinformation exchange between heterogeneousPDM systems, and it should be defined as aform that can interpret and exchange any kind

of information on the system;

(c) Suitability: The PDM system should be defined

in a suitable format to present all of the PPRinformation since this includes informationabout the product, process, and resource.This paper defines a neutral file based on XML thatcan be extended, and which redefines the structure ofthe neutral file so that it supports the total manage-ment of PPR information based on the structure of thePLM Services The name of the neutral file defined inthis paper has been abbreviated from ‘PPR eXchange’

to ‘PPRX’

The top-level neutral file defines the relationshipbetween the neutral files for each PPR, and therelationship between the neutral files within the PPRinformation Under the top-level, another level ofneutral file exists with each file containing PPRinformation

As shown in Figure 2, each PPR neutral fileaddresses some part of the PLM Services, and consists

of three parts: ‘Information’ presents the structure ofthe PPR information (BOM), ‘Document’ links therelated geometry files, and ‘Document Property’presents the detailed properties of the geometry files

Figure 1 Properties and sources of PLM Services 1.0.(Object Management Group 2005)

The method of defining the structure of the

‘Information’ section is identical to the method used

to define PLM Services, and in this part, the PPR

structure is presented In particular, the subordinate

structures of all three neutral files under the data on

‘Information’ have the same format (Figure 3) The

‘Item’ tag represents the ‘Information’ section within

the neutral file of the product, and in the neutral file of

the process and resources, the tag is changed to

‘Process’ and ‘Resource’ ‘Id’, ‘Name’, and ‘Version’

are the subordinate attribute values of ‘Item’, and the

structure of ‘Item’ is defined in ‘Assembly Component

Relationship’

In the ‘Document’ section, the related PPR

geome-try files are defined and represented in the same format

As shown in Figure 4, the ‘Document’ section presents

the location of the geometry files rather than the other

attribute values, and has a structure that links these

attribute values to ‘Document Property’

In the ‘Document Property’ section, the attribute

values related to the geometry files are defined, and

presented differently depending on the attribute for each

geometry file Hence, ‘Document Property’ will be

introduced in the explanation for each neutral file

3.2.1 Relation definition of the neutral fileThe neutral file for the PPR relation definition definesand presents the neutral files of the product, process,and resources which are closely connected Therefore,this paper only defines the relationship between PPRinformation rather than addressing the aforemen-tioned schema or PLM Services Figure 5 is anexample of the defined neutral file for the PPR relation.The file extension is ‘.lop’ which is an abbreviation of

‘Link of PPR’

In this paper, process-centred management isapplied for efficient PPR information management inPDM or PLM systems In the neutral file, PPRinformation is produced under the process tag accord-ing to the process procedure and represents thisrelationship

3.2.2 Product definition of the neutral file

In the product neutral file, the ‘Information’ sectionconsists of the continuation of the ‘Item’ tags, witheach ‘Item’ tag representing one product The assemblystructure of the product is shown under the tag The

Figure 2 Structure of the neutral file for PPR information exchange, PPRX

International Journal of Computer Integrated Manufacturing 219

Figure 3 Structure of ‘Information’.

Figure 4 Structure of ‘Document’

product represents not only the completed product,

but also each component that would be assembled as a

part of the product

The ‘Document’ section defines the location of the

geometry file related to the product, and in ‘Document

property’, the specified attribute values related to the

product are defined Figure 6 shows the structure of

‘Document Property’ The

‘Document_location_prop-erty’ section represents the location of the geometry

file, such as the CAD file or STEP file, and contains the

detailed format of that file The

‘Document_size_prop-erty’ section represents the size of the geometry file and

contains information of the LOD (Level of Detail)

‘Value_with_unit’ contains the product provider and

the own number for the product The rest of the

process and resources described below

3.2.3 Process definition of the neutral file

The ‘Information’ section of the neutral file for the

process consists of the ‘Process’ tags Each specified

process makes up one tag A process represents each step

of the production of the product, in which the process

steps are usually ‘factory-line’, ‘one´s work place’, ‘the

progress of work’ and ‘unit work’ Since no geometry file

exists for the process in the ‘Document’ section, thesections only links to ‘Document Property’ and in the

‘Document Property’ section, the neutral file showsthe specific attribute values that the process involves

As shown in Figure 7, the attributes for the neutralfile of the resources and the product are the sameexcept for ‘Valu_with_unit’ In ‘Valu_with_unit’, theset up time for the process is defined and includes thestandard working time for the process and the unit oftime used

This information has been referenced in ‘SDX(Simulation Data Exchange)’ (Sly and Moorthy 2001),and the necessary information related to the processcan now be shown in the neutral file as input values for

a process or work simulation

3.2.4 Resource definition of the neutral file

In the ‘Information’ section of the neutral file for theresources, each resource is defined and the level ofresources is presented under the ‘Resource’ tag Theresource represents not only the machinery and humanlabor, but also the working location, working desk,and everything else that is necessary to perform theprocess The ‘Document’ section presents the location

Figure 5 Example of the neutral file for PPR relationships

International Journal of Computer Integrated Manufacturing 221

of the geometry information related to the resources.

In the ‘Document Property’ section, more attributes of

the geometry information are stored As in the neutral

file for the product and the process, the rest of the parts

are the same except for ‘Value_with_unit’ which is

constructed differently to define the specified

attri-butes ‘Serial_Number’ represents the product number,

and ‘Scale’ is the size of the resource expressed asWidth, Depth, and Height ‘Position’ expresses thelocation of the resource as X, Y, Z ‘Weight’ representsthe weight of the resource, and ‘Moving_Style’represents whether the object is fixed or moving, and

if it is moving, whether it moves vertically orhorizontally (see Figure 8)

Figure 6 Structure of ‘Document_Property’ for the product

Figure 7 Structure of ‘Document_Property’ for the process

4 PLM integrator for PPR information exchange

between heterogeneous PLM systems

4.1 Concepts of the PLM integrator

Generally, a standard schema made by a company

requires a specific software development toolkit, and

many proposed standard neutral files are not

asso-ciated with an application that can support those

standards Users have to make a neutral file manually,

or they have to find an application which supports the

standard only partially This is largely why previous

proposals have not been widely adopted The PLM

integrator developed in this paper is application

software that supports the XML-based neutral file

using PPRX, and can exchange PPR information

between commercial heterogeneous PLM systems The

PLM integrator uses PPRX in every step of the PPR

information exchange After PPR information is

extracted from a commercial PDM or PLM system,

it is converted to XML files in PPRX Also, the PLM

integrator can import all PPR information into an

other PDM or PLM system

The proposed PLM integrator was developed on

the NET Framework using MS Managed Cþþ as a

programming language It was developed using API

functions and a development toolkit which were

provided by commercial PDM or PLM systems Two

PLM adapters were developed, one for Teamcenter

Engineering by Siemens PLM Software, and the otherfor SmarTeam by Dassault System

The PLM integrator consists of the XML adapterand the PLM adapter as shown in Figure 9 TheXML adapter manages and interfaces with the XML-based neutral file in PPRX, and it is not necessary tomodify the program until the definition and structure

of the neutral file are changed The PLM adaptersupports exchanges between commercial PDM orPLM systems, and should be developed individuallyfor each system using PPR information because eachcommercial system has a different developmentenvironment and method of information representa-tion Figure 10 shows a basic GUI (Graphical UserInterface) for the PLM integrator developed in thispaper

4.2 Main functions of the PLM integratorThe PLM integrator developed in this paper has threemain functions One is the ‘Download’ function, whichconverts PPR information from a commercial PDM orPLM system to neutral files in the PPRX schema The

‘Upload’ function registers PPR information in theneutral files to a commercial PDM or PLM system.The ‘View’ function views the PPR structure, proper-ties, and relationships, which helps to understand eachproduct, process, and resource along with their

Figure 8 Structure of ‘Document_Property’ for the resource

International Journal of Computer Integrated Manufacturing 223

relationships Figure 11 shows these main functions

and a basic flowchart of the PLM integrator

5 Application of PPRX and the PLM integrator to

PPR information exchange in an automotive company

PPRX, the XML-based neutral file, and the PLM

integrator were applied to the practical case of PPR

information exchange between commercial

heteroge-neous PLM systems The scope of the case study was

Processes #1-#7 on Workstation #1, which is the ‘Trim

A’ line in a general assembly shop of an automotive

company, consisting of 70 products, 90 processes, and

113 resources Figure 12 shows the line, and Tables 1,

2, and 3 summarise the product, process, and resourceinformation for this case

At first, all PPR information and data wereregistered to Teamcenter Engineering by SiemensUGS PLM Solutions, which is a popular commercialPLM system used in the automotive industry Theassembly structure, geometric information, and rela-tions for the product, process, and resource were alsoincluded Using the PLM integrator developed in thispaper, all information was extracted from the PLMsystems, and four independent XML-based neutralfiles using PPRX were generated automatically Thefirst neutral file included relationships for the PPR andthe other files which included the PPR information.This means the neutral file in PPRX represented theproperties and assembly structures of each product,process, and resource, which were pre-defined in thePLM system, along with a new definition of therelationships between them Geometric files and otherdata in the PLM system can also be translated withoutany loss

Whether the neutral file is generated from PPRinformation in the same PLM system by the PLMintegrator or not, PPR information in the XML-basedneutral file using PPRX can be uploaded automatically

by the PLM integrator In this case study, the ‘Upload’function of the PLM integrator was applied to twocases In the first case, the neutral file was generated bythe PLM integrator from SmarTeam, which is anotherpopular commercial PDM system by Dassault System

Figure 9 Concepts of PLM integrator

Figure 10 Basic GUI of the PLM integrator

In the second case, the file was edited manually by a

general text editor All PPR information and data

from the objective line in the automotive general

assembly shop were registered successfully in both

cases by the PLM integrator without any loss

Figure 13 shows the procedures and results of the

case study, including ‘Download’ and ‘Upload’ using

the PLM integrator and the XML-based neutral file

using PPRX

Using PPRX and the PLM integrator developed inthis paper, it is possible to manage PPR informationeasily and reduce the effort needed to prepare data fordiverse engineering activities The case study of theautomotive general assembly shop showed that itusually takes more than 32 h to collect and arrangePPR information and data Using the neutral file withPPRX and the PLM integrator, the same tasks tookonly two hours

Figure 11 Main functions and basic flowchart of the PLM integrator

Figure 12 Workstation #1, ‘Trim A’ line of an automotive general assembly shop

International Journal of Computer Integrated Manufacturing 225

Table 1 A portion of the summary of product information for the case study.

Product information

BUMPER-LOWER_LID_T (10012)SEAT-ROD HOOD_HOLD (10013)SEAT-ROD HOOD_HOLD (10014)

CLIP-BUTTON_N10 (10016)

LATCH A1-TRUNK LID (10018)

Table 2 A portion of the summary of process information for the case study

Process InformationBP_GA1_TR1_PR (20000) Process01 (20001) 01_LH (20011) Check manufacturing instruction (20030)

Joint ball (20031)Assemble trunk handle (20032)Assemble rear bumper (20033)Exchange hold rack (20034)Process01 (20001) 01_RH (20012) Check manufacturing instruction (20035)

Remove door hold (20036)Remove vinyl (20037)Exchange hood/door hold rack (20038)Process02 (20002) 02_LH (20013) Check specification (20039)

Insert key/card (20040)Remove door hold(NB) (20041)Joint trunk clip (20042)02_RH (20014) Check specification (20043)

Assemble harness-tailgate wire(NB) (20044)Assemble hinge bumper(NB) (20045)Process03 (20003) 03_RH (20015) Check manufacturing specification (20046)

Connect harness-tailgate handle wire(NB) (20047)Minor-assemble trunk lid handle(NB) (20048)

Table 3 A portion of the summary of resource information for the case study

Resource Information

Tool2 (30034)Knife1 (30039)Hammer1 (30040)Hammer2 (30041)Door_remover (30046)Striker1 (30047)Striker2 (30048)AirTool1 (30049)

Human2 (30010)Human3 (30011)Human4 (30012)

Column2 (30051)

WorkBench1 (30054)WorkBench2 (30055)Furniture1 (30059)

Fence1 (30061)Fence2 (30062)

SlidingRack2 (30078)

6 Conclusion

Most companies are trying to implement PLM

systems that support diverse engineering activities

with effective management of PPR information

during the entire product lifecycle Differences

be-tween commercial PDM or PLM systems in

informa-tion definiinforma-tion and structure cause difficulty in

exchanging information In this paper, PPRX (PPR

eXchange), which is a standard data format for PPR

information using XML was defined, and the PLM

integrator was developed to support PPR information

exchanges between commercial heterogeneous PLM

systems using PPRX PPRX refers to PLM Services,

which is an OMG specification, and consists of fourXML files representing product, process, resource,and PPR relationships The PLM integrator consists

of the XML adapter which interfaces with the based neutral file in PPRX, and the PLM adapterwhich supports exchanges between commercial PLMsystems By managing and exchanging the PPRinformation from the general assembly shop in anautomotive company, the PPR information exchangebetween commercial heterogeneous PDM or PLMsystems without any loss is possible The PPRX andthe PLM integrator developed in this paper are usefulfor managing and exchanging PPR information, andfor the successful implementation of PLM throughout

XML-Figure 13 Procedures and results of PPRX and the PLM integrator for the automotive case study

International Journal of Computer Integrated Manufacturing 227

the product lifecycle in many manufacturing

industries

Acknowledgement

This work was supported by a grant from Virtual

Construc-tion System Development Program funded by Ministry of

Construction & Transportation of Korean (06 high

technol-ogy fusion E01), and performed under supports by a grant

from Ministry of Knowledge Economy of Korea (Project

No 10011475-2008-23) The authors would like to

acknowl-edge the contributions of their colleagues and project

partners

References

Borgman, J., Kekkonen, J., and Reikko, L., 2002 How B2B

standards support product data exchange in product

development phase [online] Available from: http://

www.ebrc.fi/ [Accessed 3 March 2004]

Burkett, W.C., 2001 Product data markup language: a new

paradigm for product data exchange and integration

Computer-Aided Design, 33, 489–500

Choi, S.S., et al., 2009 MEMPHIS: new framework for

realistic virtual engineering Concurrent Engineering, 17

(1), 21–33

Feltes, M., et al., 2002 Product Lifecycle Management Service

Revised Submission (OMG document number

mantis/2004-04-01)[online] Available from: http://www.prostep.org/

file/13580.plm-serv10 [Accessed 5 May 2004]

Gong, P., Yang, C., and Shi, D., 2007 Heterogeneous ££

Proceedings of the 11th International Conference on

Computer Supported Cooperative Works in Design, 26–

28 April 2007, Melbourne, 951–955

International Organization for Standardization, 2000 ISO/

DIS 14649-1 Industrial automation systems and

integra-tion – Physical device control – Data model for

computer-ized numerical controllers – Overview and fundamental

principles [online] Available from: http://www.iso.org

[Accessed 10 June 2007]

Lou, Y., Lin, L., and Dong, J., 2007 Research on ££

Proceedings of the 2007 11th International Conference on

Computer Supported Cooperative Works in Design, 26–28

April 2007, Melbourne, 698–702

Lukas, U and Nowacki, S., 2005 High Level Integration

based on the PLM Services Standard ProSTEP iViP

Science Days 2005, 28–29 September 2005, Darmstadt:

Cross-Domain Engineering, 50–61

Moeck, H., 1990 Exchange of technological data via IGES.Computer Standards and Interfaces archive, 9 (2), 149–151.NIST, 1996 Initial Graphics Exchange Specification (IGES)[online] Available from: http://www.itl.nist.gov/fipspubs/fip177-1.htm [Accessed 30 August 2007]

Object Management Group, 2005 Product Lifecycle agement Services Convenience Document [online] Avail-able from: http://www.omg.org/cgi-bin/doc?dtc/2005-03-

Man-08 [Accessed 5 March 20Man-08]

Park, S.H., Choi, E.S., and Chong, T.H., 2004 TheExchange of Feature Data among CAD Systems UsingXML Transactions of the Korean Society of MachineTool Engineers, 12 (3)

Schlenoff, C and Gruninger, M., 2001 The ProcessSpecification Language (PSL) Overview and Version 1.0Specification [online] Available from: http://www.mel.nist.gov/msidlibrary/doc/nistir6459.pdf [Accessed 28 Oc-tober 2009]

Seely, S., 2001 SOAP – Cross Platform Web ServiceDevelopment Using XML Prentice Hall

Shoaf, S., 2000 Innovation that moves markets [online] IBMPLM Solutions Available from: http://www.ibm.com/software/plm [Accessed 20 June 2007]

Shoaf, S., 2001 IBM PLM: Build new markets [online] IBMPLM Solutions Available from: http://www.ibm.com/software/plm [Accessed 20 June 2007]

Siemens PLM Software, 2003 P3R and PLM 2003’ UGSPLM User Conference, 6–7 November 2003, Gyeonju.Siemens PLM Software, 2008 PLMXML Schema Resources[online] Available from: http://www.plm.automation.siemens.com/en_us/products/open/plmxml/schemas.shtml[Accessed 4 February 2008]

Sly, D and Moorthy, S Simulation Data Exchange (SDX)implementation and use Proceedings of the 2001 WinterSimulation Conference, 9–12 December 2001, Arlington,1473–1477

STEP, 2007 STEP (ISO 10303) Overview [online] Availablefrom: www.tc184-sc4.org [Accessed 2 March 2008].Sun, M and Lockley, S.R., 1997 Data exchange system for

an integrated building design system Automation inConstruction, 6, 147–155

W3C, 2004 Web Services Architecture W3C Working GroupNote [online] Available from: http://www.w3.org/TR/ws-arch [Accessed 4 February 2008]

Zhang, Y., Zhang, C., and Wang, H.P., 2000 An Internetbased STEP data exchange framework for virtualenterprises Computers in Industry, 41, 51–63

Automated discrete-pin adjustment for reconfigurable moulding machine

Y Wanga*, Z Wanga, N Gindya, R Tangband X.-J Gub

a

Faculty of Engineering, Ningbo Campus, University of Nottingham, UK;bSchool of Mechanical, Materials and ManufacturingEngineering, University of Nottingham, UK;cCollege of Mechanical and Energy Engineering, Zhejian University, Hangzhou, China

(Received 31 May 2007; final version received 5 September 2009)

To stay competitive, the moulding industry must possess manufacturing systems that are responsive to rapid marketchanges Reconfigurable moulding is therefore much in demand After the development of a vacuum formingmachine, in which the pins are configured to a near-net shape and machined to final shape with minimal materialremoval, this paper discusses in detail the support software development that enables the automatic adjustment ofdiscrete-pins to represent different component geometry In this paper, the methodology is first explained, and theimplementation of the system within the software is then demonstrated The software system includes three parts: (1)component discretisation, (2) discrete-pin construction, adjustment, and G code generation (SCAG) and (3) displayand verification Vacuum forming of a car model is used as case study

Keywords: discrete pin tooling; reconfigurable moulding; automated pin adjustment

Dp The nominal diameter of the discrete pin

Lp Length of the discrete pin and pitch (Pp)

Nijk(x,y,z) The kth node within the patch Sij, k2 [1,t]

Pij(x,y,z) The pin in the ith row and jth column of

the discrete-pin matrix

Pp The pitch of the discrete pin

Rp The number of row of the discrete-in

matrix

Sc A general 3D component surface Sc

Sij (x,y,z) The (i,j)th NURBS patch of the Scin

response to the Pij

St The tooling space constructed by the

discrete-pin matrix

Xp Distance between the central axes of the

two discrete pins next to each other in the

X direction

Yp Distance between the central axes of the

two discrete pins next to each other in the

Y direction

1 Introduction

Manufacturing companies in the 21st century face

unpredictable market changes These changes include

the frequent introduction of new products, ing product demand and mix, the manufacture oflegacy parts for existing products, new governmentregulations, and new material and process technology

chang-To stay competitive, manufacturing companies mustpossess manufacturing systems that are responsive toall these variables

The moulding industry is a fundamental industry

in many countries and moulding itself is regarded asone of the most important manufacturing processes.Currently, moulding machines are overwhelminglydedicated, expensive and time-consuming and havebecome the bottle neck of the industry Reconfigurablemoulding is now much in demand as the mouldingindustry has to become more responsive to unpredict-able market changes

A new reconfigurable moulding machine utilisingdiscrete-pin tooling is therefore proposed The reconfi-gurable tooling of such a machine is composed of anarray of identical discrete-pins which are engaged witheach other within a frame By adjusting the verticaldisplacement of the discrete pins, a wider variety ofcomponent geometry can be formed, thus activecontrol over the tool shape for the use with differentcomponents becomes possible

Having developed a machine for reconfigurablemoulding, the paper focuses on the software pro-gramme that enables automatic G code generationfor discrete pin adjustment that in turn enables itsreconfigurability for different components

*Corresponding author Email: yan.wang@nottingham.ac.uk

International Journal of Computer Integrated Manufacturing

Vol 23, No 3, March 2010, 229–236

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

Ó 2010 Taylor & Francis

DOI: 10.1080/09511920903527853

2 Prior art

As defined by Munro and Walczyk (2007), a

reconfi-gurable tool herein is defined as a machine that can

be repeatedly configured by a user to shape different

mechanical parts in a manufacturing setting Although

most machines are reconfigurable by virtue of their

replaceable cutting bits, dies, moulds, rollers, and the

like, the machines discussed in this paper are

reconfi-gurable pin-type tools with a variable surface similar

to the popular three-dimensional pin art or

PinPres-sionsTMtoy (Fleming 1987) Reconfigurable pin tools

are often used for machining, forming, moulding or

casting of complex curvature parts from metal,

poly-mers, ceramics, wood and laminated composites from

metal, polymers, ceramics, wood and laminated

com-posites Generally, reconfigurable pin tooling includes:

reconfigurable work holding and reconfigurable

moulding Reconfigurable tooling is both economical

and ecologically advantageous for the industry as one

reconfigurable device serves many components,

sig-nificantly reducing tooling costs and lead time in

the manufacture of new components Waste of material

is also reduced both in the manufacturing process

and in the disposal of workholding devices that are

no longer needed

Reconfigurable work holdings normally utilise an

array of pins, which are pushed into contact with

components and gently conformed to the component

surface and lock into position The support pins can

be used to reduce machining deformation and

vibra-tion In terms of the driving mechanism for work

holding devices, two possibilities are spring advancing/

mechanical locking (MATRIX GmbH Stuttgart) and

pneumatic advancing/hydraulic locking (KOSTYRKA

GmbH) Another approach to conformable fixtures

has been the encapsulation of at least part of the

component in phase changing materials, e.g a low

melting alloy containing lead (Camp 1978) or

magne-torheological fluids (Rong et al 2000)

In terms of reconfigurable moulding, having

reviewed the history of reconfigurable tooling patents

from 1863 to 2003 and related research from the late

1960s to the present, Munro and Walczyk (2007) have

suggested that although small-scale reconfigurable pin

tool prototypes have been built for research and for

one-off manufacturing projects, only limited designs

have actually led to commercialisation Two examples

of the commercialisation of reconfigurable moulding

machines are the multi-point forming (MPF) system

and reconfigurable pin tooling technology (Surface

Generation)

MPF is a flexible forming method for sheet metal

parts developed by a research group led by Professor

Li at Jilin University, China It uses two reconfigurable

pin groups which function as the stamping dies to form

a sheet The reconfigurability of the MPF machine

is realised by a matrix of pins in the punches, each

of which are controlled by a servo motor and areindependently adjusted to an appropriate position toform different component geometry The application ofthe MPF machine has been used for the manufacture ofpanels for high speed trains (Avant-Garde Commu-nication), sheet metal forming (Sun et al 2007) andmedical engineering applications (Chen et al 2006, Tan

et al 2007) The research undertaken into the MPFmachine has included the minimisation of spring back(Liu et al 2008), the digital manufacture of titaniumprosphesis for cranioplasty (Chen et al 2006, Tan et al.2007), the study of forming rules and characteristics ofMPF for dish heads (Qian et al 2007), the finite elementsimulation of MPF sheet forming process (Cai and Liu2005) and the surface design for a blank holder in theMPF (Sun et al 2007)

Reconfigurable Pin ToolingTM technology wasdeveloped by Surface Generation plc using patentedpin technology (Halford 2005a, Halford 2005b, Half-ord 2006) The pin tooling system developed by SurfaceGeneration plc has a matrix of square-shaped pinsmade of a consumable tool material When adjusting apin, the pin rows next to the pin being adjusted can beseparated automatically to allow individual pins, whichare mounted on discretes, to be adjusted vertically byrotating them around the central axis After adjust-ment, all the square pins are oriented at 45 degrees toallow for efficient packing with adjacent rows Finally,the rough upper surface of adjusted pin ends is CNCmachined to produce the final tool shape without theneed for excessive material waste

Although advanced research has been conductedwith MPF technology, since every pin in the MPFsystem is individually controlled by a servo motor,the machine is very expensive, which prohibits thetechnology from being employed widely Similar pro-blems exist in the reconfigurable pin tooling technol-ogy developed by Surface Generation plc

The reconfigurable moulding machine, utilisingdiscrete-pin tooling, can move pins automatically todesired positions by a CNC discrete driver, therefore,eliminating the need for employing servo motors toadjust individual pins, significantly reducing the cost

of servo motors and the complexity of the machinesystem, and being more affordable for industry, e.g.SME companies

Having developed a vacuum forming machineshown in Figure 1, in which the pins are configured

to a near-net shape and machined to final shapewith minimal material removal, this paper discusses

in detail the development of support software thatenables an automatic adjustment of discrete-pins in

order to represent different component geometry.

This software enabling system has not been discussed

within the remit of previous reconfigurable moulding

systems

3 Methodology

Identical standard discrete pins are used There are

two possible pin engagement methods as shown in

Figure 2 Obviously, the gap between pins is smaller

in Figure 2(a) than of that in Figure 2(b), thus the

engagement method in Figure 2(a) is selected, leading

to one less pin in the even row than those in the odd

row Assuming that the discrete-pin matrix is

com-posed of m 6 n pins, let Pij(x,y,z) denote the pin in

row i and column j, Pij (x,y,z) is represented by the

intersection point of the top surface and the central

axis of the pin, then tooling space Stcan be expressed

as a set of these discrete points as

St ¼ P11 [ [ Pij [ Pmn; 2 ½1; m; j 2 ½1; n ð1Þ

The parameters of a discrete pin tooling include rows(Rp) and columns (Cp) of discrete pins (Cp is thecolumn number in the odd row), pin diameter (Dp),length (Lp) and pitch (Pp) The distance between thetwo discrete-pin central axes in the x direction and inthe Y direction, designated as Xpand Ypas shown inFigure 2, is needed to identify the position of Pij, and isdetermined by the pin diameter (Dp) and the Pitch (Pp).When two identical external discretes are meshedtogether, the distance of the discrete pin in X direction

Xpis

Xp ¼ Dp 0:649519  Pp ð2Þ

Yp¼

ffiffiffi3p

Figure 1 Frame of reconfigurable screw pin tooling system

Figure 2 Screw-pin engagement methods

International Journal of Computer Integrated Manufacturing 231

matrix, the position of Pijis a function of i and j The

function of the Y coordinate value of Pijis

Pijy ¼ i  Yp R p 1

Since there is one pin less in even rows, the x, y, z

positions of the last pin in even row is assigned as zero

Otherwise, the function of the x coordinate value of

Pij(x,y,z) is dependent on whether i is odd or even:

Pijx ¼ j  Xp C p 1

=  Xp if in even row ð5Þ

Pijx ¼ j  Xp C p 2

=  Xp if in odd row ð6Þwhere i2 [0,Rp7 1] and j2 [1, Cp7 1]

A general three-dimensional component surface Sc

can be meshed as m 6 n NURBS patches

Sc¼ S11[ ::: [ Sij:::[ Smn ð7Þ

where Sij (x,y,z) denotes the (i,j)th NURBS patch of

the Scin response to the Pij,

Pij:x Xp=  Sijx  PijX þ Xp= ð8Þ

Pij:y Yp=  Sijy  Pijy þ Yp= ð9Þ

Each of the patches can be discretised into manydiscrete points, and assuming that point cloud is welldefined and can replace the surface model, Sij can befurther represented by t discrete nodes, then

Sij Nij1[ ::: [ Nijk:::[ Nijt k2 1; t½  ð10Þwhere Nijk(x,y,z) denotes the kth node within the patch

Sij

In order for the tooling space St to represent thenet-shape of the component geometry surface Sc,whilst having minimal material removed, then

4 Software implementation

Discrete-pin tooling control is conducted in three steps

and the detail of the framework is shown Figure 4

Different software platforms including CAD/CAM

(Unigraphics), Finite Element Analysis software

(ABAQUS) and scanning software (Gom) are

inte-grated into the interface and developed using Visual

Basic The system is composed of three parts: (1)

component discretisation, (2) discrete-pin

construc-tion, adjustment, evaluation and G code generation

(SCAG) and (3) display and verification

4.1 Component discretisation

In order to obtain the component position information

in response to the discrete-pin matrix, it is essential to

discretise the 3D component surfaces There are two

types of discretisation: discretisation of 3D CAD

model and discretisation of the 3D physical model

The discretisation of 3D CAD model is also called

mesh generation, and is the same as that of finite

element analysis, and is therefore conducted usingFEA software (ABAQUS/CAE) Discretisation of a3D physical model, employed when CAD models arenot available, is conducted through scanning by usingthe Gom scanner

To avoid meshing the component surface with thefinite element analysis environment, it is important

to obtain a good quality of nodes that are evenlydistributed over the component surface Also, whentransferring geometry from a CAD environment to anFEA environment, information loss occurs frequentlyowing to the disparity between the interpretation anddata expression of different software, therefore it is firstimportant to repair the geometry before the discretisa-tion takes place The detailed steps for the geometrydiscretisation are shown in Figure 5

4.2 Discrete-pin construction, adjustment, evaluationand G code generation (SCAG)

SCAG is a user-friendly interface developed within

a visual basic environment The purpose of the

Figure 4 Software framework

International Journal of Computer Integrated Manufacturing 233