d522 dissemination strategy - uninova.pt · partners/eu:technical research centre of finland (vtt),...
TRANSCRIPT
VOSTERVOSTERVOSTERVOSTERVOSTERVOSTERVOSTERVOSTER
VOSTER IST-2001-32031
Virtual Organisations Cluster
Deliverable D4.4 ICT support infrastructures and interoperability for
VOs
Prepared by UNINOVA
with contributions from ….
and inputs on projects from ALL partners Contract Start Date: 01.12.2001 Duration: 30months Project Co-ordinator: VTT Partners/EU:Technical Research Centre of Finland (VTT), Fraunhofer-Institute for Industrial
Engineering (FhG-IAO), Center for Technology & Innovation Management (CeTIM), Instituto de Desenvolvimento de Novas Tecnologias (Uninova), Research Institute for Rationalization and Operations Management at Aachen University of Technology (FIR), Loughborough University, Dresden University of Technology (TUD), University of Salford (Usal), Computas, Consortium for Advanced Manufacturing International (CAM-I), University of Amsterdam (UvA), Concurrent Engineering Consulting (CEC), Silesian University of Technology (SUT)
Project funded by the European Community under the "Information Society Technology" Programme
D4.2
VOSTER WP4 D4.4
ICT support infrastructures and interoperability for VOs
Deliverable
no & name : D4.4 ICT support infrastructures and interoperability for VOs
Version : 0.1 Date : 15.03.2003 Schedule :
Classification : Author(s) : (Compiled by)
L. M. Camarinha-Matos, K. Menzel Public
Contributions:
T. Cardoso
Short description: This document presents the results of the study of the contribution to the state of the art on general (horizontal) support infrastructures and interoperability mechanisms for Virtual Organizations / Virtual Enterprises (VOs / VEs).
Date : 15.03.2003 Version 0.1
By :
Date :
By :
Date :
By :
Date :
Document history : (Description of change from previous version)
By :
D4.4
Table of Contents
SUMMARY ............................................................................................................................ 3
1. INTRODUCTION ............................................................................................................... 4 1.1 OBJECTIVES .................................................................................................................. 4
2. THE INFORMATION BASIS ............................................................................................. 5 2.1 MOTIVATION .................................................................................................................. 5 2.2 LEVEL OF CONTRIBUTION TO INFRASTRUCTURES AND INTEROPERABILITY......................... 7 2.3 PROJECTS DURATION .................................................................................................... 1 2.4 PROJECTS CONSIDERED FOR ANALYSIS AND SYNTHESIS ................................................ 1
3. THE STATE OF ICT INFRASTRUCTURES...................................................................... 3 3.1 INTRODUCTION .............................................................................................................. 3 3.2 HORIZONTAL INFRASTRUCTURE...................................................................................... 4
3.2.1 Layer-based / transaction-oriented infrastructure ................................................. 4 3.2.2 Agent-based infrastructure.................................................................................... 6 3.2.3 Service federation infrastructure ......................................................................... 10 3.2.4 VO Life cycle support .......................................................................................... 12
3.3 SUPPORT FOR PROFESSIONAL VIRTUAL COMMUNITIES................................................... 15 3.4 SUPPORT FOR REMOTE OPERATION AND E-SCIENCE...................................................... 17 3.5 VO BREEDING ENVIRONMENTS SUPPORT ...................................................................... 20 3.6 IMPLEMENTATION TECHNOLOGIES AND STANDARDS....................................................... 21
4. FACILITY / INTERIOR SYSTEMS .................................................................................. 23 4.1 INTRODUCTION ............................................................................................................ 23 4.2 CLASSIFICATION .......................................................................................................... 23
4.2.1 VO-Life Cycle Phases and VO-Functions ........................................................... 23 4.2.2 VO Work Environments....................................................................................... 24 4.2.3 Facilities Classification ........................................................................................ 25 4.2.4 Interrelations: VO-functions - ICT - Facilities & Interiors ..................................... 26
4.3 SYNTHESIS .................................................................................................................. 28 4.3.1 Single Workspaces ............................................................................................. 29 4.3.2 Team Area .......................................................................................................... 30 4.3.3 Integrated CSCW-Lab......................................................................................... 32
4.4 SUMMARY ON FACILITIES AND INTERIORS ...................................................................... 34 5. CONCLUSIONS .............................................................................................................. 36
REFERENCES:.................................................................................................................... 36
Page 2 of 3
D4.4
Summary
This report presents a summary of the findings of the first phase of the VOSTER project regarding the contributions to the ICT support infrastructures to virtual enterprises / virtual organizations (VE/VO). These results are mainly based on the projects composing the VOSTER portfolio, i.e. projects represented by members of the VOSTER consortium. The main aspects considered in this report are:
Horizontal ICT infrastructures VO life cycle support services Infrastructures for Professional Virtual Communities Infrastructures for remote operation and e-science.
In terms of horizontal infrastructures three main approaches are analysed:
Layer-based / transaction-oriented infrastructure, Agent-based infrastructure, and Service Federation infrastructure.
This synthesis document also addresses the implementation technologies and standards. In addition to ICT infrastructures it is also necessary to consider their integration with the building and interior facilities (furniture, lightening, tele-conferencing, etc.) in order to create appropriate human support environments for VE/VO participation. Therefore one section of the report is devoted to facilities and interiors.
Page 3 of 4
D4.4
1. INTRODUCTION
1.1 Objectives The overall aim of VOSTER is to collect, analyse and synthesize the results from a number of leading European research projects on Virtual Organisation (VO), i.e. “geographically distributed, functionally and culturally diverse, dynamic and agile organisational entities linked through ICT”. In addition to the European projects, and although taking into account the limited resources available, VOSTER also intends to consider some relevant projects from other geographical areas (e.g. USA, Canada, Australia, Mexico, Brazil, Japan). Within this general scope, the Work Package 4 (Infrastructures and Interoperability) aims at defining the architecture, support services and interoperability principles for VO ICT infrastructures and suggest ways to implement them especially for SMEs. The work done during the first phase of the project was divided into three main steps: information acquisition, analysis synthesis. For this period the focus was put on the initial portfolio of European projects represented by partners in the VOSTER consortium. The following diagram shows the main facets of the information acquisition and analysis made:
Infrastructure architecture
Support technologies & paradigms
Infrastructure management
Infrastructure users & actors
Implantation approach
Functional blocks/componentsInfo & control flowsCoordination modelPhysical infrastructure
Main paradigmsMain support technologies
Initiation:Who / how Costs and risks sharingPolicies and regulations
Users, roles and usage modesOwnership and responsibilities
Integration of legacy systemsRe-organization needsTraining …
Figure 1 – Facets selected for information acquisition and analysis
This report presents a summary of the results of the Synthesis task, which is devoted to “define a common framework for relevant aspects of VO infrastructures and map the findings into it. Prepare consensus definition of the required functionalities of VO infrastructures.” For the next period non-European projects and also emerging products being offered by major software vendors will also be taken into account.
Page 4 of 5
D4.4
2. The information basis UNINOVA
2.1 Motivation During last decade, considerable investments have been made, namely in Europe and the USA, in a large number of research projects applying new organizational forms, which have produced an abundant variety of specific solutions and broad awareness for the necessary organizational changes. Particularly in the case of Europe, given both its cultural background in business and the current emerging development investments, Europe is placed in a key leading position for the development of the organizational forms fit for the digital age. However, the research in many of these cases is highly fragmented, being each project focused on solving specific problems and applying Information and Communication Technology (ICT) to partially design and develop a minimal B2B interaction mechanism to support its basic needs. As such, there is no effective consolidation/harmonization among them in order to have an effective impact. The availability of a large number of contributing elements (technologies, paradigms, best practices and models) constitutes a set of enabling factors for the opportunity of materializing concepts that although not new, were waiting for the enabling factors. However, most of these technologies and concepts are in their infancy and under development, mostly fragmented and non-interoperable. In this context some of the motivating elements for the VOSTER “consolidation” initiative are:
The implementation and configuration of a VO support infrastructure still requires considerable engineering effort, an obstacle to dynamic VOs.
Even the most advanced infrastructures coming out of leading R&D projects require complex configuration and customization processes hardly manageable by SMEs.
Almost all VO/VE projects are forced to create their own infrastructure due to the lack of common reference architectures and interoperability principles. This represents a deviation of some resources from its main focus, while generating something only applicable to each specific project.
Only a few projects if any, address the development of a comprehensive horizontal infrastructure.
There is a need to discern between enabling vs. disabling technologies; some efforts are too biased by short-term technologies, which might represent an obstacle for non-ICT SMEs.
TCP/IP, CORBA-IIOP, HTTP, RMI, SOAPJ2EE Framework, CORBA Framework, ActiveX Framework
EJBs, OAG and OMGs Business Objects and ComponentsUML, UEML, WfMC XML-based Business Language
WfMC, OMG-JointFlow, XML-WfMC standardsODBC, JDBC, FIPA, OMG-MASIF, Mobile Objects
JMS, MS-Message Server, MQSeries, FIPA-ACCBizTalk, CBL, OASIS, ICE, RosettaNET, OBI, WIDL
Servlets, JSP, MS-ASP, XSL, WSDL …!!!
Figure 2 – Proliferation of technologies – enabler or disabler ?
Page 5 of 6
D4.4
Although it is not a new subject but rather the central question in systems integration, interoperability remains as a critical topic in the agenda of VO/VE supporting infrastructures. Furthermore, the fast evolution of ICT technologies with reduced life cycles and the need to cope with technologies with different life cycles and at different stages of the corresponding life cycle represent a major difficulty.
On the other hand, it shall be noted that:
Emerging infrastructures induce new organisational forms, but emerging organisational forms will require new support infrastructures (a principle of co-evolution).
In addition to the infrastructure needs, some commonalities on basic functionalities start to emerge – requiring consolidation – but there is a lack of general support functions for collaborative distributed business processes. It is still the case that some stages of the VE/VO life cycle are mostly ignored by current projects.
Although the potential advantages of the Virtual Enterprises / Virtual Organizations are well known at the conceptual level, their practical implantation is still far from the expectations. In order to leverage the potential benefits of the agile VE/VO paradigm, there is a need for flexible and generic infrastructures to support the full life cycle of the VO/VE, i.e. creation, operation, evolution and dissolution. Achieving such infrastructures is still a major challenge. There is therefore an urgent and strong need to harmonise and consolidate:
o Current state of the art in terms of basic infrastructure support and interoperability principles.
o Contribution to a “de facto” reference architecture. o Identification and characterisation of emerging generic business support services for
collaborative environments. o Identification of EU strengths and gaps.
Page 6 of 7
D4.4
2.2 Level of contribution to infrastructures and interoperability As it could be expected, not all VE/VO projects within the VOSTER portfolio contribute directly to the VO/VE support infrastructures and interoperability area. Some are devoted to the legal or organizational facets, for instance. The following figure shows, in a simplified way, the estimated “level” of contribution (in a scale of 1 to 5) of each project to the main facets of the VO infrastructures. The considered facets are: “Horizontal” infrastructure– covering activities related to the design and development of generic ICT infrastructures to support collaborative businesses in networked environments, including interoperable and safe information exchange platforms, generic coordination and collaboration functionalities, legacy systems integration, etc. The considered sub-areas are:
Safe communications Interoperability and tools integration Information and knowledge sharing Repositories Coordination mechanisms Collaborative environments
VO creation functionalities - covering activities related to the design and development of functions to support the planning and launching of new VOs, such as partner registration and search, marketplace management, e-procurement and negotiation, etc. The considered sub-areas are:
VO planning Partners search / registry Marketplace / service centre management Negotiation support / VE/VO formation Management of catalogues of enterprises (enterprises profile / skills/services offered)
VO operation functionalities – ICT functions to support the operation phase of the VO, with particular focus on generic (i.e. domain independent) services. Sub-areas to consider include:
Distributed business process planning, scheduling and supervision Process modelling and simulation tools Distributed logistics management tools Performance assessment tools Specialized domain functions (those functions depending on the application domain of the
VO). VO dissolution functionalities – ICT functions to support the dissolution phase of the VO, such as:
Liability definition tools Product life cycle support tools (after the dissolution) Performance history data management (collecting and organizing information about past
performance of companies in VOs). Etc.
Main technologies – identifying the main implementation tools / technologies used in each project. Examples could be: Java, Multi-agent based environments, Jini, XML, SOAP, CORBA, Workflow engines etc.
Page 7 of 8
D4.2
"Horizontal" Infrastructures VO creation funtionalities VO operation functionalities VO dissolution functionaliti Main Technologies
PROJECTInfo & Knowl.
SharingCoord.support
Safety & security
Inetroperab. & legacy integr.
Collab. environm.s
VE planning
Partners search/select.
Enterprises catalog
Cluster / market place
Contract negotiation
Distrib. BP plan.& sched.
Distrib. BP supervision
Perform. Assess.
Process model./simul.
Specialized domain funct.
Liabilities definition
Prod/service lifecycle supp.
Perform. Histot. data
BAP 1 4 1 1 4 4 4 4 3 1
BIDSAVER 3 3 3 3 4 4 3 2 2 2 2Internet, VPN, RDBMS, XML, search engines, PDM,
PM, ODBC, Remote Accesse-COGNOS 3 3 2 2 1 2 1 XML / SOAP / UDDI, Ontology managerE-COLLEG 3 2 3 5 3 3 3
ELEGAL 4 4 4 4 3 RDBMS, XML, Java, EJBeans, EXPRESSEXPIDE 4 1 2 1
EXTERNAL 5 5 3 3 3 3AKM, COMWorks, XML, SOAP, Webdav
...FETISH-ETF 3 4 4 3 4 4 3 3 Java, Jini
GENESIS 2 2
GLOBEMEN 4 3 4 3 2 3 2 2 3 3 3VERAM (extension of GERAM + set of other
modelling approaches), 7 layer info archoitecture
GNOSIS 3 3 4 2 2 2 4 3 3Optimization (Constraint Propagation), web &
internet based integration, XMLICSS 5 5 1 3 4 4 4 4 IFC 2.x, STEP, WfMC, distributed DB, RMI
ISTforCE 5 5 2 3 4 4 2 3 2 3 5 4 IFC 2.x, STEP, WfMC, XML, distributed DBMASSYVE 3 3 3 3 3 3 3 3 3 3 MAS, federated/distributed DB
NGMS 1 2 2 3 3 1 1 1
NIMCube 3 3 3 3ASP, JAVA, communic. security/authent.
standardsNIITM 4 4 3 4 4 3 2
OSMOS 3 3 3 Web based tools, XMLPRODCHAIN 5 3 1 1 1 2 5 5 4 ERP/SCM systems
PRODNET II 4 4 5 4 3 5 3 5 4 5 3STEP, EDI, workflow, DBPs, comm. security
/authent. standards, federated DB, RPCSYMPHONY 2 2 KM, BIS, EIS, DSS, …
VDA 4VL 5 4 5 5 5
ALIVECE-NET II 1 1 2
COVE 3 3 3 3 3 2 2 2 2 2 2 2 2 2 2ICCI 3
KM Forum 2 2 2 2 Web-basedProDAEC 1 1 1 5
THINKcreative 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
TeleCARE 5 2 5 5 5 1 4 MAS, mobile agents, federated DBUEML 1 4
1 2 3 4 5grade legend
Figure 3 – Overview of Projects’ contribution level to the ICT infrastructures
D4.2
2.3 Projects Duration The R&D projects within the VORTER portfolio considered for the information acquisition, analysis and the synthesis presented in this report are at distinct phases of their lifecycle. Therefore, the contribution from each project for this document is quite different for projects already finished (e.g. PRODNET II) or projects still in their early phase (e.g. SYNPHONY). Figure 4 shows a Projects’ duration table.
P
R
O
J
E
C T
TeleCARE
COVE
UEML
20042000 2001 2002 20031996 1997 1998 1999
VL
e-COGNOS
ICCI
ProDAEC
PRODCHAINOSMOS
PRODNET IISYMPHONY
VDA
ICSSISTforCE
MASSYVE
NIMCube
BAPBIDSAVER
NGMS
E-COLLEGELEGAL
EXTERNALFETISH-ETF
GENESISGLOBEMEN
GNOSIS
ALIVECE-NET II
THINKcreative
KM Forum
Figure 4 – Projects duration
2.4 Projects Considered for Analysis and Synthesis As a result of a preliminary information acquisition phase, it became clear that not all projects in the VOSTER portfolio have contributions to the ICT infrastructures and support services. Some projects are focused on non-technological issues (e.g. VIVE is focused on the legal issues of VOs), other initiatives are accompanying measures and therefore do not develop new infrastructures or tools (e.g. COVE, THINKcreative). Finally, some projects are still in their early phase and no substantial results are available yet (e.g. SYMPHONY), as seen in Figure 4. Based on these findings, the following projects were considered for the analysis phase and this resulting synthesis:
D4.4
PRODNET MASSYVE EKMF NIMCUBE OSMOS
ISTforCEICSS
EXTERNAL PRODCHAIN
COGNOS FETISH
GLOBEMEN EXPIDE
NIITM E-COLLEG
VL BIDSAVERTeleCARE
Contributions to:-Horizontal infrastructures
-Life-cycle support functions-Methodologies
-Facilities
Figure 5 – Projects considered for the current analysis
Page 2 of 11
D4.4
3. The state of ICT infrastructures UNINOVA
3.1 Introduction
Fig. 6 summarizes the various levels of required infrastructures to support not only inter-enterprise collaborative networks but also more human-oriented networks, i.e. professional virtual communities and virtual research communities that can emerge in the context of a VE/VO.
Enterprisenetworks
Human-orientedCollaborative networks
Basic horizontal infrastructures
Collaborative environments
Advanced collaborative environments
VO/VE
ProfessionalVirtual
Communities
VL /e-ScienceRemote
supervision
Safe communications, Information sharing & exchange,Basic coordination, Distributed business process
management
Shared spaces, Collaborative tools & assistants,Asynchronous collaboration, Flexible coordination
Remote access to equipment,Joint experiments management,
Data processing, visualization, mining
(transaction-oriented, agent-based, service federation)
Figure 6 – Main classes of VOs and corresponding infrastructures
The “typical” VOs (and Virtual Enterprises) have been mostly focused on the basic interactions to support business collaboration among enterprises, including: safe communications, distributed information management and information sharing (eventually using specific standards such as EDIFACT, STEP), coordination, and (minimal) distributed business process management, with little focus on the human collaboration. The second level of infrastructure is mostly dedicated to support collaboration among humans, although some of the projects also consider the organizations behind the professional virtual communities. Examples are the support for the case of Concurrent / Collaborative Engineering in networked organizations, networks of consultants, and other professional virtual communities. The third level is focused on a specialized human collaboration. It combines both inter-organizational and human collaboration, but including the access to remote equipment (e.g. machines, sensors, equipment for scientific experiments), collaborative experiments involving teams located close to the equipment and teams located remotely. Most of the VOSTER portfolio projects address the needs of inter-enterprise collaboration, but there are already a good number of projects addressing the specific needs of teams of professionals belonging to different organizations and operating in a networked environment.
Page 3 of 12
D4.4 Various projects address more than one type of infrastructure. So far, only a very limited number of activities have addressed the needs of the third level, which corresponds to a recent extension of the scope of virtual organizations.
3.2 Horizontal infrastructure
Three main approaches have been followed in the development of horizontal infrastructures for VO/VE:
Layer-based or transaction-oriented approach, which adds a cooperation layer to the existing IT platforms of the enterprises. Inter-enterprise cooperation is then performed via the interaction through these layers.
Agent-based approach, including the cases that represent enterprises as agents and the inter-enterprise cooperation as interactions in a distributed multi-agent system.
Service-federation / service-market approach - according to which enterprises publish their services in service directories, representing their potential “offer” to the cooperation processes. By means of proper “standard” service interface, the interoperability with other (service requesting) enterprises, regardless of the heterogeneity associated with the actual implementation of the services themselves, is supported.
Although not all projects aim at developing horizontal infrastructures, the lack of a common and widely accepted reference model and reference architecture is forcing almost every project to design and implement its own mini-infrastructures, deviating some resources from its main focus, while generating something only applicable to that project. Although it is not a new subject but rather the central question in systems integration, interoperability remains as a critical topic in the agenda of VO supporting infrastructures. Furthermore, the fast evolution of ICT technologies with reduced life cycles and the need to cope with technologies with different life cycles and at different stages of the corresponding life cycle represent a major difficulty. There is a need to discern between enabling vs. disabling technologies; some efforts are too biased by short-term technologies, which might represent an obstacle for non-ICT SMEs. It is also important to notice that emerging infrastructures induce new organisational forms, but emerging organisational forms will require new support infrastructures (co-evolution principle).
3.2.1 Layer-based / transaction-oriented infrastructure
Typical infrastructures in this group have identified the need to exchange business (EDIFACT) and technical (STEP) data through the use of standards. An adequate coordination functionality and proper interaction with the enterprise’s ERP/PPC and PDM systems are necessary in each enterprise in order to “guarantee” its proper participation in the VO/VE-business process. Further functionality includes safe communication to guarantee the privacy and authentication of the business interactions, and federated / distributed information management to support the information visibility rights and sharing of production status data. Table 1 summarizes the main facets addressed by VOSTER projects and also the current limitations of the proposed solutions in the layer-based approach.
Page 4 of 13
D4.4 Table 1.
Transaction-oriented layer-based horizontal infrastructure
Key facets:Safe communications
Cryptography, symmetric & asymmetrickeys, digital signature, certificatesVPN
Information sharing and exchangeDistributed/federated information managementSpecification of access rights / visibility
Workflow-based coordinationStandards for exchange of some classes of information
EDIFACT, STEPMore recently XML based structures
Various approaches for remote objects& services access
RPC, CORBA, RMI, EJB, Jini
Current limitations :No common reference modelNeed to integrate different technologies (from different vendors)
Technical complexityUnclear responsibilities
Infrastructure is still complex, difficult to configure and poor interoperabilityLimited support for distributed business process managementLack of support for VE dissolutionLimited mechanisms for tracking and auditingPoor support for breeding environments management
Example 1.
PRODNET II
One of the earlier examples of a layer-based / transaction-oriented infrastructures is provided by the Esprit PRODNET II project. The infrastructure extends the functionalities of each VE member (represented by the enterprise applications such as ERP/PPC, PDM, CAD, etc.) with a Cooperation Layer responsible to handle all enterprises transactions and cooperation events.
Network
PCI
Coordination kernel
LCM DIMS
Service Service...
Configur.& Interface
PCI
Coordination kernel
LCM DIMS
Service Service...
Configur.& Interface
PCL - PRODNET Cooperation Layer
ECL- Enterprise Cooperation Layer
VMF- VE ManagementFunctionalities
FunctionalitiesEMF- Enterprise Management
CCL- Core Cooperation Layer(transaction protocols)
PCL - PRODNET Cooperation Layer
ECL- Enterprise Cooperation Layer
VMF- VE ManagementFunctionalities
FunctionalitiesEMF- Enterprise Management
CCL- Core Cooperation Layer(transaction protocols)
Figure E1.1 – PRODNET II multi-level architecture
A three-levels hierarchical coordination architecture is considered:
- Coordination level related to the transaction protocols (Core cooperation layer);
- Coordination level related to the enterprise’s duties regarding the involvement in the cooperation; and
- Coordination level related to the distributed business processes management.
Central to this coordination kernel are a Distributed/Federated Information Management System (DIMS) and a workflow-based Coordination Engine (LCM). A safe Communications Infrastructure (PCI) and a library of support services (EDIFACT, STEP/PDM) complete the Coordination Layer.
Page 5 of 14
D4.4
CoordinationEngine
(Workflow Engine)
CoordinationMonitor
Message Handler
Loader / WPDL Parser
LCM
GraphicalWorkflow
Editor
WPDL
DIMSQuery/Update Services
Federated
Schema
Manager
ViewHierarchyConfig
DIMSKernelConfig
FederatedQuery
Processor
PCI
SECURITYPICM
PRODNET Intelligent
Communication Manager
MCIMessage
Class Identifier
Multi Protocol Access Control
SMTP/POP3
Web Proxy
CGITCP/IP ...
API (RPC and DLL)PCI
SECURITYPICM
PRODNET Intelligent
Communication Manager
MCIMessage
Class Identifier
Multi Protocol Access Control
SMTP/POP3
Web Proxy
CGITCP/IP ...
API (RPC and DLL)
EDIFACTServices
STEP/PDMServices
LCM – LocalCoordination
Module
DIMS -Distributed Info.Manag. System
PCI – ProdnetCommunication
Infrastructure
Figure E1.2 – Main components of PRODNET II layer
Typical services for an industrial VE, as offered by the PRODNET platform, include:
• Exchange of commercial data via EDIFACT messages. • Exchange of technical product data using STEP. • Federated / distributed information management, supporting not only the administrative information about the
VE, but also the information the enterprise shares individually with other VE members. • Coordination module, handling all cooperation-related events (execution of a local activity flow plan). • Configuration module, allowing the definition and parametrization of the VE and the behavior of each particular
node. • Safe communications, including cryptography services, digital signature, certificates, auditing mechanisms, etc. • Monitoring of orders and production status. • Quality related information exchange. • Extended ERP/PPC system adapted to interact with a VE environment.
It shall be noted that infrastructures combining heterogeneous components from different vendors are potentially unstable being difficult to determine which component (or tool provider) is responsible when something goes wrong with such complex systems. In spite of the fast growing technological developments, lack of proper interoperability mechanisms among enterprise applications is a major obstacle to agile VO/VEs.
Setting up an infrastructure for VE/VO based on multiple technologies still requires a large engineering effort, which represents a major obstacle for the implantation of this new organizational paradigm. Furthermore, it shall be noted that the fast evolution of the information technologies often presents a disturbing factor for non-IT companies.
3.2.2 Agent-based infrastructure Although at an abstract level there are similarities between agent-based and layer-based approaches, from the underlying technology / software development point of view they are quite different. We are assuming here an AI notion of agent as an encapsulated computational system, that is situated in some environment, and that is capable of flexible, autonomous behaviour in order to meet its design objective. There are nowadays a large number of development platforms for multi-agents systems (MAS), most of them based on Java. Some of these platforms, e.g. FIPA OS, JADE, ZEUS, follow the FIPA (Foundation for Intelligent Physical Agents) specifications and several of them are open-source.
Page 6 of 15
D4.4 There are a number of characteristics in the VE/VO domain that make it a suitable candidate for
the application of multi-agent systems approaches. Examples of such characteristics include: – A VE is composed of distributed, heterogeneous and autonomous components, a situation
easily mapped into MAS. – Coordination and distributed problem solving also tackled by MAS are critical problems in
VE management. – Decision making with incomplete information, and involvement of network members as
autonomous entities, that although willing to cooperate in order to reach a common goal might be competitors regarding other business goals, is another common point.
– The effective execution and supervision of distributed business processes requires quick reactions from enterprise members. Computer networks being the privileged media for communication, there is a need for each company having a “representative” in (or “listening” to) the network. Agents can support this need.
– Recent developments in VE are changing the focus from information modeling and exchange to role modeling, addressing aspects of distribution of responsibilities, capabilities and knowledge.
– The phase of VE formation in which it is necessary to select partners and distribute tasks, shows market characteristics and negotiation needs that have been research issues for years in the MAS community (coalition formation).
– A VE consortium is a dynamic organization that might require re-configurations – e.g. replacement of partners, changes in partners’ roles, etc., for which a flexible modeling paradigm is necessary.
– VE supporting functionalities need to interact with the “local” environment (legacy applications and humans). Interaction with the environment is one of the defining attributes of agents.
– The scalability property of MAS seems particularly adequate to support dynamic VEs in which different levels of cooperation with different sets of partners might be established at different phases. On the other hand, each enterprise might itself be seen as composed by a network of semi-autonomous entities (departments).
– More flexibility than in a client-server model is required to support dynamic change of roles of the VE members.
– Continuous evolution of business models, technologies, organizational paradigms, and market conditions require effective support for evolution and a high level of modularity of the infrastructures.
– New forms of teamwork, namely cooperative concurrent engineering or Virtual Communities of Practice (VCP), are emerging in the context of VEs. Agents can play an important role as “assistants” to the human actors in such environments.
– There is a need to handle the requirements of autonomy vs. cooperative behavior for which federated MAS approaches may provide a balanced solution.
– On the other hand, as agents are designed and developed independently, it is quite difficult to guarantee coordination unless common rules (“social laws” and standards) are adopted. Theoretical foundations on agents’ sociability can be combined with current developments of a legal framework for VE/VOs.
It shall be noted that in spite of these strong motivating elements, the application of MAS technology to VE/VO infrastructures is still limited to research projects. Table 2 summarizes the main facets addressed by VOSTER projects and also the current limitations of the proposed solutions in the agent-based approach.
Page 7 of 16
D4.4 Table 2.
Agent-based horizontal infrastructure
Key facets:Support for VE creation
Partner search and selection based on negotiation Virtual market places and brokersPreliminary steps towards e-contracting
Some support for VE operationDynamic scheduling functionsCombination of inter-agent communication and federated information managementFirst steps in contract management
Current limitations :There are many development platformsfor MAS, namely some FIPA compliant (e.g. JADE, FIPA OS) but they are not robust enough when operating over InternetSecurity and persistence mechanisms are not yet well integrated with MASLack of integration between AI and BP communities (e.g. there is a need to integrate ACL with BP languages)Developments mostly at prototype level; real demonstration cases missing
A growing number of research prototypes on the application of multi-agent systems and market-oriented negotiation mechanisms for the VO/VE formation are being developed. Early MAS applications to VO/VE are mainly focused on the creation phase. In many cases it is assumed that simple mechanisms of inter-agent cooperation are sufficient to support the operation phase of VO/VE. The decision-making in a VO/VE is however a complex process where it is important to combine human decision with some automatic functionalities. It is even likely that the level of automatic decision-making will evolve as the trust of humans in the systems increases. But independently of the ultimate decision making center, there is a need to provide mechanisms to support process coordination, supervision, and controlled information exchange and sharing.
Example 2.
MASSYVE
The INCO MASSYVE project is an example of application of a multi-agent approach in VO/VE infrastructures.
Fig. E2.1 illustrates a practical example of the application of MASSYVE to VE creation in the context of an industry cluster formed by twelve companies in the domain of moulds and die-casting. The cluster is legally represented by a broker entity that supports a human expert responsible for getting and analyzing business opportunities. By means of a broker agent an opportunity is transformed into a distributed business process that is then distributed (through a contract-net protocol) to the (potentially interested) enterprises within the cluster. In the end of the whole process, a set of possible teams of enterprises (“potential” VEs) that can carry out that business opportunity is formed and the most suitable team is proposed (but the ultimate decision is made by the human expert). In this example, there are three VEs capable of accomplishing the business process but VE1 was the selected team.
Moulds industry cluster [Rabelo, Camarinha-Matos, Vallejos 2000]MASSYVE Project
E1
E9
E5E11
E3
E2
E8
E12
E10
E7
E4
E6
BPa
BPb
BPc
BPd
VE2
DBPClient
Broker
VE1
VE3
Based on a real scenario: cluster of 12 mould and die industries in the south of Brazil
• The broker agent is an assistantto the human broker
• The broker “builds” a set of possible consortia.
• The final selection of thewining consortium is made bythe human.
Human broker
Figure E2.1 – MAS-based brokerage in VE formation
One of the distinctive aspects of the MASSYVE approach is its hybrid / semi-automatic philosophy in which agents are
Page 8 of 17
D4.4 used as human assistants. The fact that the MASSYVE approach is focused on clusters of enterprises, i.e., a controlled multi-agent universe, makes it a feasible solution from the implantation point of view, since a common modelling framework can be agreed upon and adopted by all members of the cluster.
The Contract-net Protocol coordination mechanism is used to support the task assignment among agents, and the Negotiation method is used to overcome conflicts taking place during planning or execution phases, both at intra-enterprise and inter-enterprise levels.
mould_sizemould_typemould_materialdue_datemax_price
Mould_i2 tonsmouldaluminum28/12/0090.000
I can do it only if ...
I knowhow to do
that !
I will be inmaintenance
I cannotmake that !
I’m free inthat period
I accept it too !I accept
it !
Me too
I accept it too !I accept
it !
Me too
Selected Enterprise_Agent for
the mould
Selected Enterprise_Agent for
the mould
mould_sizemould_typemould_materialoccupation
Capability_Enterprise_j[1-3 tons]mouldplastic10/12/00
For each task in the Business Opportunity (i.e. each mould)a possible agent is selected using a standard contract-net protocol
Figure E2.2 – Contract-net based protocol for partners selection
In MASSYVE an integration of MAS and federated information management is proposed. Each agent is enhanced with a Federated Information Management System (FIMS), through which it seamlessly interoperates and exchanges information with other agents. However, considering the autonomy of agents, the access to information is strongly controlled by the information visibility rights defined among them that in turn preserve their autonomy. Therefore, a MASSYVE agent is seen as a kind of tandem architecture composed of a “normal” agent and its FIMS. An essential concept introduced in this architecture is that the data elements are not sent from one agent to the other via a high-level protocol (e.g. ACL language), as in the traditional push strategy case, but rather through a pull strategy, via accessing to the respective agents’ FIMSs. Thus, the high-level protocol is only used for the control/coordination purposes. Figure E2.3 illustrates this approach. Consider an example case where a given agent (B) processes some information and generates some results (for example the “actual end of the production date for a part P and its termination status”) that are needed to be accessed by another agent (A), according to some predefined supervision clauses specified in the VE’s contract. Following the contract, then B sends a message to A (represented by “1” in this Figure), communicating that the data item on “part P’s actual end of the production date” is at this enterprise. This control message sent from B to A informs A that now this data item is available and can be accessed by A (through its FIMS’ import schema). Please notice that the access rights for the shared data among nodes are dynamically and bilaterally configured and preserved by their import/export schemas according to agents’ roles in the collaboration and their needs.
Internet
Enterprise BEnterprise A
1
4
32 5
Exportschema
Integratedschema
Importedschema
EnterpriseAgent A
EnterpriseAgent B
FIMS_A FIMS_B
Figure E2.3 – Combination of MAS and federated information management
Page 9 of 18
D4.4 With deeper studies of VO application domains however this paradigm reveals many specific aspects that cannot be simply supported by basic MAS approaches. In the VO community interoperation / cooperation must be regulated by the following requirements that require further extensions to the basic MAS technologies:
- Cooperation agreements and contracts that establish a framework for the general operating conditions must be established.
- Distributed business process models and mechanisms that establish the allocation and sequence of tasks to be performed by the community must exist.
- Efficient data exchange and communication services, distributed service management functionalities, support for nodes autonomy / privacy, high level of service quality, auditability, and accountability, etc., have to be guaranteed.
A number of recent research works have addressed the issues of contract modeling and electronic contracting processes.
3.2.3 Service federation infrastructure Figure 7 illustrates the basic principles behind the service federation approach. Regardless the different implementation approaches, the general three steps – publish, discover, invoke – are usually considered.
Serviceimplementation
Serviceadapter
ServiceClient
(Application)
Service proxy
Service proxy
Service spec
Service Catalog
Service interface
Serviceprovider space
Client space
Servicepromoter space
Service
discovery
andProxy
Copy
Remote service invocation
Serviceimplementation
Serviceadapter
ServiceClient
(Application)
Service proxy
Service proxy
Service spec
Service Catalog
Service interface
Serviceprovider space
Client space
Service CatalogManager
Publ
ish
or Regi
stra
tion
Remote service invocation
1 2
3
WSDL ?
SOAP ?
UDDI ?
Figure 7 – Service federation approach
According to this model, companies (potential members of the virtual organization) are considered as “service providers”, i.e. the potential collaborative behavior of each company is “materialized” by a set of services. The approach assumes the existence of one entity that keeps a catalog of services where service provider companies publish their service offers. This entity is sometimes called a “service market”, a “service promoter node”, or even “service portal”. Services are published in the catalog through a “service specification” record that also includes a “proxy” or representative of the service to allow transparent remote invocation of the service. Client companies can then use the functionality provided by the catalog manager to “discover” existing services (and their providers). Once a service is selected, the client gets a copy of the
Page 10 of 19
D4.4 corresponding proxy and can invoke it remotely in a transparent way, i.e. without caring about the actual location of the service or even its implementation language. An early example of this approach is represented by the JINI technology. More recent vendor-independent efforts have been trying to establish a number of de facto standards to facilitate the creation of such infrastructures: UDDI, WSDL, SOAP. The implementation of a harmonized representation of services in the service catalog does not necessarily mean that all members have access to all services all the time. Service providers shall keep their autonomy and the right to specify whom and under which conditions, has access rights to their registered services. Table 3 summarizes the main facets addressed by VOSTER projects and also the current limitations of the proposed solutions in the service federation approach. Table 3.
Service federation horizontal infrastructure
Key facets:Basic architectures for service federationMechanisms for remote service invocationPreliminary standards for servicedescription and cataloguing: UDDI, WSDL, SOAP, …Basic service search mechanismsPreliminary mechanisms for Value Added Service composition
Current limitations :Poor integration of service federation andVO conceptPoor integration of security / privacy mechanisms Search mechanisms still too basicAd-hoc concept of portal
Example 3.
FETISH-ETF
FETISH-ETF is an example of application of a service federation approach to establish a horizontal infrastructure to support VOs in tourism. Clustering is a typical phenomenon in this sector, especially at regional level. Regional tourism promotion organizations try to offer an integrated view of the local resources (service promoter node). In order to reduce the fragmentation and dispersion of information some trends to form networks of service promoter nodes are emerging.
SPN
SPNSPN
Service Promoter Nodes
Catalog ToolsTools
SPN
ServiceService
Service provider
ServiceService
Service provider
ServiceService
Service provider
Tourism service providers
Access rights manager
Access rights manager
LookupServices
LookupServicesServices
Registrationmanager
ServicesRegistration
manager
BP Manager /VE planner
BP Manager /VE planner
MembersRegistration
ManagerMembers
RegistrationManager
......
Service Catalog Management
Catalog ToolsTools
SPN
ServiceService
Service provider
ServiceService
Service provider
ServiceService
Service provider
Tourism service providers
Access rights manager
Access rights manager
LookupServices
LookupServicesServices
Registrationmanager
ServicesRegistration
manager
BP Manager /VE planner
BP Manager /VE planner
MembersRegistration
ManagerMembers
RegistrationManager
......
Service Catalog Management
Figure E3.1 – Network of service promoters
The developed infrastructure is based on Java/JINI, extending the service federation mechanisms of JINI to a wide area network. Services are specified using WSDL.
Services implementation can be done in any language provided that a wrapper (proxy) for remote access transparency is developed in Java.
Page 11 of 20
D4.4
Hotel Palma
HotelReservation
Node A
ServiceImplementation
Wrapper_HRP
HotelReservationspecParametersAttributes…
Service Catalogue
HotelReservationspecParametersAttributes…
Service Catalogue
HotelReservationspecParametersAttributes…
Service Catalogue
HotelReservationspecParametersAttributes…
Service Catalogue
Figure E3.2 – “Publishing” a service in the Services Promoter Node (Catalogue)
Hotel Palma
HotelReservation
Node A
ServiceImplementation
Wrapper_HRP
HotelReservationspecParametersAttributes…
Service Catalog
Node B
Travel AgencyParadiseTours
Copy
ServiceInvocation
Wrapper_HRPcopy of
Figure E3.3 – Discovering and invoking a service
The tourism services can be provided in two levels. The lower level contains the basic tourism services, which represent atomic services created by a tourism operator. On top of this level, it is possible to create another level in which the basic services are used as components of a Value Added Service (VAS). An example of a VAS is a holiday package. The “materialization” of such a value-added service involves a process – business process (BP) – requiring the invocation and coordination of other services that may be provided by different service providers. In such a case of distributed execution the BP becomes a distributed business process (DBP). The contributing members can therefore form a temporary organization or virtual enterprise. FETISH-ETF provides a tool for composing VAS, specifying the corresponding workflow and selecting service providers from the catalogue. A selection of a set of service providers to satisfy the needs of a particular VAS leads to the formation of a particular VO to attend the business opportunity given by each instantiation of that VAS.
A “popular” term - portal – could be considered a special case of service federation infrastructure. This term is often used in the literature to represent rather different structures, from (i) a simple web entry point to a number of resources without any other form of collaboration among the resource providers, to (ii) some form of virtual organization. In this latter case, the basic idea is the existence of a “centralized access point”, where the various VO members are “represented” and that might offer a “joint representation” to the outside, as well as a platform for interactions among the VO members.
3.2.4 VO Life cycle support In terms of VO life cycle support there is a clear focus on the “creation” and “operation” phases. While in earlier projects the emphasis was mostly on supporting the operation phase, it can be noticed that in running projects the creation phase is becoming as important.
Page 12 of 21
D4.4 The “dissolution” phase remains almost untouched. Even though a few projects claim they address this phase, the actual functionality supported is rather limited.
Operation
Evolution
DissolutionOperation
Evolution
Dissolution
Partners search and selectionContract negotiationDefinition of rights
Join/Leave procedureInfrastructure configur.
Secure data exchangeSharing and visibility rights
Orders managementDBP planning & scheduling
High level coordinationCollaborative engineering
....
Liabilities definitionAccess rights definition
...Add / remove partnerChange roles
...
CreationCreation
Figure 8 – Functions in VO life-cycle support
VO creation. Some of the functionalities addressed in different projects include: VO planning, partners search and selection (perhaps the most addressed area), enterprise catalogues, contract negotiation, etc. It shall be noted that the fact that several projects address one topic this does not mean the topic is extensively treated. For instance, the issue of partners search is, in some cases, meant as a facility to help search by humans based on HTML pages. Only a few projects progressed towards computer-assisted (e.g. based on agents) partners search. One of the problems in partners search for VO creation is the availability of directories of companies / organizations where their profile (skills, resources, performance history, etc.) are represented in a standard format. The unavailability of a standard to represent these profiles has been a major obstacle. Recent developments (UDDI, WSDL, SOAP) represent promising steps in this direction. Nevertheless if the catalog is confined to a cluster or VO breeding environment it is easier to reach a common representation of profiles. VO operation. Coordination of distributed business processes (DBP) and activities is an important element in the VO operation. Most of the early approaches took a workflow-based approach, starting with the WfMC reference architecture and experimenting with extensions for supervision of distributed processes (cross-organizational workflow) including some preliminary but very limited works on exception handling, multi-level coordination and its relationship to coordination roles of the VO members, and flexible workflow models to support less structured processes. In terms of DBP planning and modeling some graphical languages have been suggested but a standard is still necessary in order to allow effective distribution / sharing of business processes. Proposals like PIF and PSL have been discussed in the USA but there is still more work to do. A recent European initiative, the UEML network, might contribute to some harmonization in the modeling area. In terms of business support functions, most of the projects have addressed application specific cases. ERP vendors have been extending their monolithic single-enterprise-centric systems in order to comply with more dynamic supply chains and networks, but this is an area requiring more investment on generic functionalities.
Page 13 of 22
D4.4 Example 4.
ISTforCE
Some examples of specific functions for the operation phase can be found in the ISTforCE project:
Personal Planning System module: enables the management of multi project participation by interacting with distributed project management servers. Methods have been developed for merging different project workflows into a personalized workflow that supports the user in organizing his work more efficiently. The Personal Planning Server is based on a relational database according to the IFC 2x process model. The Personal Planning Client is developed as a Java application.
Data
Data
Workflow 1
Project 1
Engineer
Data
Workflow 2
Project 2
Data
Workflow 3
Project 3
WF 1WF 2WF 3
Personal Planning Service
Data
Data
Workflow 1
Project 1
Engineer
Data
Workflow 2
Project 2
Data
Workflow 3
Project 3
WF 1WF 2WF 3
Personal Planning Service
Figure E4.1 – Personal planning system
e-Commerce Services (ECS):
Figure E4.2 – Electronic payment service
VO evolution and dissolution. During the life cycle of a VO it is natural that some partner leaves the consortium and be replaced by a new organization. The termination of this collaboration process or even the ending of a VO, are subjects not properly addressed yet. The consequences of the operation of a VO/VE cannot be simply discarded when the VO/VE dissolves. Most of these consequences are of a legal nature and shall be regulated by the cooperation agreements. That is the case, for instance, of the responsibility of customer support / product maintenance during the life cycle of the product / service generated by a VO/VE. Environment regulations are also forcing companies to plan provisions regarding the product disposal and recycling after its end of life. Recent regulations in some countries also state that the liabilities regarding each component of a
Page 14 of 23
D4.4 product may ultimately lie with the component’s supplier. In the case of a network chain type of manufacturing this forces each node in the chain to keep track of the history of each component/sub-product that “passed by” this node. This is a functionality that is properly supported by most of the more advanced ERP systems, but not by most of the scheduling systems, for instance. There are, however, several other less “material” issues which are more difficult to handle. One of these issues is the Intellectual Property Rights policy, namely for the post-dissolution phase and its consequences in terms of information accesses by the VE members. In some cases there is also the possibility that the VE evolves into a more permanent organization, a joint venture enterprise created by the VE members, to exploit the intellectual and industrial property results developed in cooperation. There is also considerable knowledge that can be elicited from the ending cooperation experience, namely the knowledge about what went right, what went wrong, partners performance / reliability, jointly defined business process templates, etc. Defining the ownership and access rights to this knowledge is not an easy task and requires further investigation. Example 5.
BIDSAVER
••TThhee VVEE ccoonnttrraacctt eessttaabblliisshheess,, tthhrroouugghh aa ddeeddiiccaatteedd tteecchhnniiccaall sseeccttiioonn,, rruulleess ffoorr:: –Computing partners’ rights on the final product and on related profits: reference algorithm id based on pro-rata ownership of profits, based on a development mortgaging mechanism with asymptotic royalty and on the subdivision of profit according to effort and financial risk
–Assessing liability of individual partners on the final product, through the reference to responsibility on product parts and processes.
••TTHHEE IICCTT eennvviirroonnmmeenntt ooff BBIIDDSSAAVVEERR aaccccoommmmooddaatteess:: –Management of dynamic composition of partnership
–Establishment of responsibilities
–Tracking of activities and costs (effort + expenditures).
••IItt ssuuppppoorrttss ffoorr VVEE tteerrmmiinnaattiioonn::
–The enacting of principles for product liability allocation to partners down to the end of liability period
–The computation of rights and due profits to individual partners upon VE termination, through the valorisation of assets and of residual industrial risks.
3.3 Support for professional virtual communities When a proper cooperation environment is in place, professional virtual communities (PVC) / virtual teams emerge within such environment, constituting a fundamental element of value creation, innovation and sustainability. Virtual Communities and Communities of Practice are not new concepts but they acquire specific characteristics and increased importance when considered in the context of the collaborative networks of organizations. These communities, although spontaneously created, are bound to certain social rules resulting from the commitment of their members to the underlying organizations (new concept of social-bound PVCs). This is the case, for instance, in concurrent or collaborative engineering where teams of engineers, possibly located in different enterprises, cooperate in a joint project such as the co-design of a new product. A large number of computer supported cooperative tools are becoming widely available for synchronous cooperation. Some examples are teleconference, and chat tools combined with application-sharing mechanisms. Considering the geographical distribution, the autonomy of the VE members, the local corporate cultures, and also the individual working preferences of the team members, it is likely that most of the activities will be carried out in an asynchronous way, which requires new assisting tools.
Page 15 of 24
D4.4 In terms of coordination, several approaches to develop flexible workflow systems have been proposed. In the case of processes mainly executed by humans, rigid forms of procedural control are not adequate. People like to keep their freedom regarding the way they work. Product design, like any other creative process evolves according to a kind of “arbitrary” workflow. It is therefore necessary to also support loosely constrained sets of business processes. The trend is followed by other communities of professionals (e.g. consultants) that share the body of knowledge of their professions such as similar working cultures, problem perceptions, problem-solving techniques, professional values, and behavior. Another special area of application, as for instance represented by the TeleCARE project, is the creation of virtual communities for providing care services to elderly. These communities involve organizations and people such as care providers, health care professionals, relatives of elderly, and the elderly.
Table 4 summarizes the main facets addressed by VOSTER projects and also the current limitations of the proposed solutions in the support for professional virtual communities.
Table 4.
VCP-support infrastructure
Key facets:First steps towards “shared working spaces”Large number of “small” tools (e.g. chat, instant messaging, teleconference, CSCW)
Some application sharing mechanismsFirst steps towards flexible workflow and asynchronous coordination of activities, notification mechanismsBasic VCP management
Current limitations :Very limited integration of tools andmechanismsLimited coordination facilitiesNo adequate VCP management forprofessional communities (Virtual Communities of Practice - VCP) able to capture multi-level relationshipsNo integration of IPR issues in the VCPmanagement servicesLimited support for mobile contexts
Example 6.
TeleCARE
The IST TeleCARE project aims at designing and developing a configurable framework focused on virtual communities for elderly support. Different actors – e.g. care workers, health workers, elderly relatives, and elderly – cooperate in order to provide care assistance to elderly living alone at home.
Leisure site
Care center 3
Emergency mobile site
Mobility for elderly
Relative’smonitoring site 2
Virtual shop 1
Special doctor site
Elderly home sites
Care center 2
Virtual shop 2
Relative’smonitoring site 1
Police StationPolice Station
Care center 1
Figure E6.1 – Virtual community for elderly care
Page 16 of 25
D4.4
A mobile-agent approach (based on AGLETS open-source platform) is adopted.
Figure E6.2 – Mobile agents based approach
Core MAS Platform Level
Inter-platform Mobility Federated Information Manager System
Enter-tainmentservice
Safe Communicationsinfrastructure Devices abstraction layer
Tele
CAR
E B
asic
Pla
tform
Spec
ializ
edC
ompo
nent
s
Inference engine
Basic Multi-Agent Platform
OntologyManager
Inter-agent communication
Agent Factory Platform Manager
External Enabler Level
Vertical Services Level
Virtual Community Support
StatusMonitoring
service
TimeBank
service
Specialized Interfaces for elderly
Resources Management
System
AgentReception &Registration
AgentExit
Control
PersistenceSupport
Module
Agendaservice
Core MAS Platform Level
Inter-platform Mobility Federated Information Manager System
Enter-tainmentservice
Safe Communicationsinfrastructure Devices abstraction layer
Tele
CAR
E B
asic
Pla
tform
Spec
ializ
edC
ompo
nent
s
Inference engine
Basic Multi-Agent Platform
OntologyManager
Inter-agent communication
Agent Factory Platform Manager
External Enabler Level
Vertical Services Level
Virtual Community Support
StatusMonitoring
service
TimeBank
service
Specialized Interfaces for elderly
Resources Management
System
AgentReception &Registration
AgentExit
Control
PersistenceSupport
Module
Agendaservice
Figure E6.3 – TeleCARE node’s architecture
Security and privacy is of a major importance in this domain. VPN approach is used at the communications level, agent passport and federated information management mechanism are adopted for information privacy, and biometric and Java smart cards are being evaluated for user identification.
3.4 Support for remote operation and e-science
Page 17 of 26
Advanced forms of cooperation mostly in the area of design and manufacturing require mechanisms to support a controlled “intrusion” of a company, for instance the VE coordinator, into the “territory” of its partners. An initial example of this “intrusion”, which is properly supported by the federated database paradigm, is the access to selected (authorized by the cooperation agreements) subsets of the information (for instance, the orders’ status, stock levels, etc.) [4]. But this process may assume more extensive forms. Consider the case that a company wishes to
D4.4 “open a window” over the shop-floor of its partner (according to contractual rules) to monitor the manufacturing process of the ordered parts and even have an interference on the shop-floor processes, i.e. supervise these processes from distance and in cooperation with the local people. The local supervision functionalities installed in each production site shall interact / cooperate with the global VE (network-wide) supervision, providing controlled levels of transparency. Mobile computing also suggests new forms of tele-operation and tele-supervision of processes. A Virtual Laboratory (VL) is another form of collaborative network, representing a heterogeneous, distributed problem solving environment that enables a group of researchers located in different geographical places to work together, sharing resources (equipments, tools, experimental data, etc.), i.e. a specialized form of VO (Fig. 9). Virtual Laboratories are attracting growing interest due to their potential applications in areas such as education, research, medicine, etc., in order to operate tools and equipment in remote locations or hazardous environments, among others. A typical VL involves scientific equipments connected to a network, large-scale simulations, visualization, data reduction and data summarization capabilities, application-specific databases, collaboration tools e.g. teleconferencing, federated data exchange, chat, shared electronic-whiteboard, notepad, etc., application-dependent software tools and interfaces, safe communications, and large network bandwidth.
Therefore, the concept of Virtual Laboratory has to be supported by the following main requirement classes: (i) Remote operation: in order to have access and manipulate tools and equipment located at a remote workshop; (ii) Information management: to store information and data generated by the experiments realized; (iii) Simulation: to visualize and reproduce the actions on the remote workshop, and (iv) Collaborative tools: in order to share and coordinate the experiments among different partners around the world. A more recent term – e-science – is likely to replace the term virtual laboratory.
Figure 9 – e-Science environment
Collaborative remote supervision in manufacturing shares many requirements and challenges with the e-science environment, among which the following can be mentioned: Operation of remote instruments / equipment, remote diagnosis, team discussion, virtual meetings, sharing data at highly interpreted level, tele-mentoring (e.g. explaining a new process), what-if analysis / simulation, exclusive and shared accesses (to data, tools, and devices), access rights definition (to data, tools, and devices), visual checks and video link (conferencing), support for un-experienced users (the supervision partner does not necessarily know the details of the remote environment), and the need to automated many adjustments that were previously done manually (if not assisted by local operators in the physical equipment place). Table 5 summarizes the main facets addressed by VOSTER projects and also the current limitations of the proposed solutions in the remote operation and e-science.
Page 18 of 27
D4.4 Table 5.
Remote operation / e-science support infrastructure
Key facets:Various mechanisms to connect equipments to the webApplication of mobile agents to increase autonomy and independenceof network characteristicsFirst attempt to use GRID as a generalinfrastructure for resources managementPreliminary mechanisms for collaborative experiments managementSpecialized tools for data visualizationand data miningFirst attempt for heterogeneous data integration for multi-disciplinary research
Current limitations :No integration of access rights / visibility mechanismsThe “business perspective” including intellectual property rights is not addressed yet in this contextLack of extensive and robust demonstration casesPoor understanding of cooperationprocessesPoor error recovery mechanisms
Example 7.
VL-E
The VL-E (Virtual Laboratory for e-science) project aims at the design and development of an open, flexible, scalable, and configurable framework providing necessary GRID-based hardware and software, enabling scientists and engineers in different areas of research to work on their problems via experimentation (a virtual scientific community), while making optimum use of the Information Technology. The VL-E provides a distributed high performance computing and communication Virtual Laboratory infrastructure with advanced information management functionalities, addressing in specific the experimentation requirements in the scientific domains of biology, physics, and systems engineering. As such, access to physically distributed data and processes among many sites in the virtual laboratory, that is necessary for the realization of complex experimentations, will be totally transparent to the scientists, giving them the image of working in a single physical laboratory (a special type of virtual organization).
Figure E7.1 – Multi-layer VL concept
An architectural overview of the GRID-based VL-E is given in Figure E7.2. The VL-E makes it possible to attach a wide range of software tools to the laboratory; from basic tools such as simulation, visualization, data storage / manipulation, to advanced facilities. The main functionalities of the VL-E middleware include: remote controlling of devices (COMCOL), visualization in a virtual reality environment (VISE), and federated advanced information management (VIMCO). The VL-E solves various technical problems that scientists face, hence enabling them to focus better on their experiments, while using the GRID infrastructure, simultaneously reducing the costs of experimentation by sharing the expensive resources among them.
Page 19 of 28
D4.4
Micro-arrayMicro-array ConfocalmicroscopeConfocal
microscopeCAVE-Virtual
realityCAVE-Virtual
realityHigh-performance
computingHigh-performance
computingHigh-performance
storageHigh-performance
storage
GRID MIDDLEWARE
DNA-ARRAY APPLICATION MACS APPLICATION OTHERS
VL-E MIDDLEWARE
VO FRAMEWORK
BASE FUNCTIONALITY
Expressive MACS EFCExpressive MACS EFC
VIMCO
VISE COMCOL
VL-E PORTAL & WORKBENCH INTERFACE
applicationtoolkit tier
grid resources tier
applicationtier
gridservices tier
EFC
VIMCO
VISE COMCOL
VL-E PORTAL & WORKBENCH INTERFACE
applicationtoolkit tier
grid resources tier
applicationtier
gridservices tier
Figure E7.2 – VL architectural overview
The GRID infrastructure is promising for the Virtual Laboratory environments supporting scientists and engineers with their complex collaborative experimentation. VL nodes in a network register themselves and become GRID nodes. The GRID-based nodes will then share their full computational and storage capacity, properly managed by the GRID resource management facility. As such, collaborating scientific nodes are supported by a close-to-standard communication infrastructure, empowering them with high-performance and/or high-throughput computing machinery, which is extendable with more resources, as other nodes join the GRID. Another advantage of using GRID for the scientific community is the possibility to share the free published data among the nodes.
3.5 VO breeding environments support Early proposals, too much technology-driven, underestimated the difficulties of the VE/VO creation process and suggested very dynamic scenarios but the agility and dynamism required for VOs are limited by the process of trust building. Even if flexible support infrastructures become widely available, the aspects of trust building and the required reorganization at the enterprise level are hard to cope with in cooperative business processes. “Trusting your partner” is a gradual and long process. The definition of “business rules”, contracts for VE/VO or even common ontologies also take time, especially when different business cultures are involved. In this sense, very dynamic organizations formed by enterprises without previous experience of collaborating together might be limited to scenarios of simple commerce transactions (e.g. buy-sell). The creation of long term clusters of industry or service enterprises represent an approach to overcome these obstacles and can support the rapid formation of VE / VO according to the business opportunities (Camarinha-Matos and Afsarmanesh, 2001). The concept of cluster of enterprises, which should not be confused with a VO, represents an association or pool of enterprises and related supporting institutions that have both the potential and the will to cooperate with each other through the establishment of a long-term cooperation agreement. Buyer-supplier relationships, common technologies, common markets or distribution channels, common resources, or even common labour pools are elements that typically bind the cluster together. This is not a new concept as a large number of related initiatives have emerged during the last decades, namely in Europe and USA. But the advances in information and communication technologies now bring new opportunities to leverage the potential of this concept, namely by providing the adequate environment for the rapid formation of agile virtual enterprises.
The more frequent situation is the case in which the cluster is formed by organizations located in a common region, although geography is not a major facet when cooperation is supported by computer networks. Nevertheless, the geographical closeness has some advantages for cooperation as it may facilitate better adaptation to the local (culture) needs and an easier creation
Page 20 of 29
D4.4 of a “sense of community”. But with the development of more effective communication infrastructures such long-term associations are not necessarily motivated by geographical closeness. Cultural ties, even particular human relationships are also motivating factors to form such associations which represent in fact the VO Breeding Environments (VBE) for the dynamic formation of VE/VOs. For each business opportunity found by one of the BE members, acting as a broker, a subset of the VBE enterprises may be chosen to form a VO/VE for that specific business opportunity.
Figure 10 illustrates the process of creating a VO with and without a breeding environment.
“Open universe”“Controlled-border universe”
Breeding Environment
VE
Market trendsCompetitiveness
Business opportunity
• Wide partners search & selection• Common infrastructures definitionand parametrization
• Sharing principles• Contract and cooperation agreement
• Partners search & selection• Cooperation agreement• Common infrastructure• Sharing principles
• Controlled partners selection• Contract & cooperation details• Infrastructure parametrization
Figure 10 – Breeding environment approach for dynamic VO creation
The enterprises involved in a given breeding environment are normally “registered” in a directory (part of a portal or service market), where their core competencies are “declared”. Based on this information, the VO/VE initiator / creator can select partners when a new business opportunity is detected. Clearly, several VOs can co-exist at the same time within a VBE, even with some members in common. A VBE, being a long-term organization, presents an adequate environment for the establishment of cooperation agreements, common infrastructures, common ontologies, and mutual trust, which are the facilitating elements when building a new VE. Although not well understood in earlier projects, it is now clear to many VOSTER projects that the formation of dynamic VE/VOs requires an appropriate “breeding” or “nesting” environment.
3.6 Implementation technologies and standards Each of the analysed projects adopts one or more implementation paradigms in the systems they developed. In this case seven main paradigms were identified in the several projects:
• Multi-Agent System • Portal • Relational DB • Federated DB
Page 21 of 30
D4.4 • Knowledge Management • Object Oriented • Workflow based • Service based
It shall be noted that these paradigms are not mutually exclusive. Different aspects of the infrastructure can be based on different paradigms and naturally each project might use various paradigms. It is also the case that not all the identified terms correspond to real paradigms. For instance, the “concept of portal” (which has various definitions) can be, in fact, based on various other paradigms. It is also arguable whether a “portal” corresponds to a class of infrastructures or simply to an implementation paradigm. The adoption of a multi-agent or service-based paradigm to implement some functions does not necessarily mean that a real agent-based or service federation infrastructure is offered. For instance, the infrastructure provided at each site might organize the locally offered functions as services while the VO as a whole is not organized as a service federation but rather as a transaction-oriented infrastructure. Since the common role of all projects is to provide some collaboration facilities among distributed entities, it is meaningful to identify the main middleware technologies used to provide a basis for the desired distribution and to try to identify tendencies. According to the analysis made, the most commonly used technologies in the VOSTER portfolio are CORBA, RPC or RMI and SOAP. Nevertheless other middleware systems such as DCOM, ASP/JSP, JINI, etc. have also been used. At least in one case, GRID is used. It seems that VO projects are mostly “followers” of the mainstream middleware technologies available at the time the project is launched. Only in a few cases there are some contacts between consortia members and major middleware “standardization bodies”. This is probably due to the following reasons:
- The main focus of attention of the VOSTER projects is on understanding and supporting collaboration rather than on developing basic / generic computational infrastructures.
- The majority of the consortia members are not software engineers. Although this is probably natural, the situation should raise some concerns regarding whether the current VO community has the right skills to address the open interoperability problems. A stronger interaction with the software engineering and operating systems communities could be advisable. It is also interesting to note that the “GRID community”, which also strated to use the term “virtual organization”, is becoming aware of our activities but the interactions between the two groups are still minimal. In terms of the VOSTER portfolio only the Dutch VL project explicitly deals with GRID. Java is becoming the most common development language. UML is frequently used as a modelling tool. SQL is also very common in terms of information querying language. Also in this area it seems that VOSTER projects are followers of the mainstreams.
Page 22 of 31
A particular note for the multi-agent developments. Although various projects claim they follow a multi-agent approach, there is no clear preference for a specific development platform. This is consistent with the fact that in spite of the fast developments observed in the MAS area and the existence of a large number of open source platforms (most of them Java-based), none of them is robust enough to support the needs of a highly distributed, heterogeneous and autonomous environment as a VO. In fact most of these platforms have been tested in applications running in a single site. It is also important to note that at the FIPA level, the main “de facto” standardization organization for MAS, there is not yet any activity related to the support of VOs. The THINKcreative project has discussed the high potential of MAS approaches, both in terms of development technology and modelling framework, but there is a need for an interaction between the VO community and FIPA.
D4.4
4. Facility / interior systems TUD
4.1 Introduction The flexibility of facilities and interior systems as well as the integration of information and communication systems into a building tremendously influence the performance of Virtual Organisations. Information and communication technology should be integrated into the built environment, instead of being an “add-on”. Within the following sections, the interdependencies between ICT and facilities are illustrated.
4.2 Classification
4.2.1 VO-Life Cycle Phases and VO-Functions Each VO function needs specific support by facility and interior components. The interdependencies between each VO function, its characterization and the required facility & interior support is specified in Table 6.
Table 6: VO-Functions with facilities & interior systems specification VO-Phase creation operation dissolution
VO-function
Characterization and facilities & interior systems specification VO
Pla
nnin
g
Part
ners
sear
ch
sele
ctio
n/re
gist
ratio
n
Ente
rpri
se C
atal
ogue
s
Mar
ketp
lace
/ C
lust
er
Con
trac
t neg
otia
tion
Dis
tribu
ted
b.p.
pla
nnin
g
Dis
tribu
ted
logi
stic
mgm
t.
Perf
orm
ance
Ass
essm
ent
Proc
ess m
odel
ling
/ si
mul
atio
n Sp
ecia
lized
do
mai
n fu
nctio
ns
Liab
ility
Def
initi
on
Prod
uct /
Ser
vice
life
cycl
e su
ppor
t
Perf
orm
ance
his
tory
othe
rs
Non-formalized VO-functions and trust building • Integrated usage of video and audio
communication tools, • Information sharing & documentation
Formalized VO-functions with need for interactive team work. • Integrated usage of tools and devices for
information sharing, presentation, and interactive, team-oriented modification
• Integrated usage of video and audio communication tools
Formalized VO-functions
• Access to shared information • Little or no need for interaction and
communication
Page 23 of 32
D4.4
4.2.2 VO Work Environments ICT (e.g. video-conferencing systems, collaborative systems, CSCW-systems, etc.) shall support VO-scenarios instead of dominating them. One can distinguish three main classes of VO-work environments, single workspace, team areas, and integrated CSCW-Labs. This classification is depicted in Figure 111. Each VO-work environment is supported by different facility and interior systems and components, such as presentation systems (PRS), lighting devices (LIG), special furniture, or moveable wall systems (FUR). Figure 11 summarizes the specification of the VO work environments and its required facilities & interior systems and components.
Enterprisenetworks
Human-orientedCollaborative networks
Basic facilities & interior systems
Collaborative environments
Advanced collaborative
environments
VO/VE
ProfessionalVirtual
Communities
VL /e-ScienceRemote
supervision
Easy Access to different information and communicationtechnologies, less cabling, ergonomic work place
integrated IT-devices
Shared spaces, Access to integrated collaborative systems,Flexible re-configuration of space layout and technical systems
Integrated IT & furniture / cabling
Remote access to equipment,Joint experiments management,
Central control unit, ROOMWARE, interactive, integrated presentation devices,
Combined natural & artificial lighting, solar heat gain, and optimised air conditioning
SINGLE WORKPLACES
TEAM AREA
INTEGRATED CSCW-LAB
Figure 11: Main classes of VOs and corresponding facilities and interiors
4.2.2.1 Single Workplaces: The “traditional” VOs (and Virtual Enterprises) mostly have been focused on the basic office facilities and interiors to support business collaboration among enterprises, including: easy access to different means of information and communication technology, less cabling, and ergonomic furniture. Each VO team member has the equipment available in his office. OR: Each VO team member is able to use these systems wherever she/he is working (e.g. a laptop computer with integrated/additional devices is available).
4.2.2.2 Team areas: The second level of facilities and infrastructure is dedicated to support collaboration among groups / teams and considers primarily the local organisation as part of the professional virtual community. Team areas primarily support the local interaction of VO-team members and secondarily the integration of VO-members working remotely. VO-members of one organization can meet virtually VO-members from other organizations by attending electronically supported meetings and by using embedded ICT-infrastructure in this special team area. Team areas are not single rooms but open “office spaces”.
Page 24 of 33
D4.4 4.2.2.3 Especially equipped CSCW-labs: The third level is focused on a specialized human collaboration. It combines both inter-organizational and human collaboration, but includes the access to remote equipment (e.g. machines, cameras, voice control, blinds, etc.). A central facility is especially equipped for “virtual meetings”. In contrast to team areas CSCW-labs are closed spaces (single rooms).
4.2.3 Facilities Classification To support the defined VO-functions specified in section 4.2.1 facilities and interior components are classified into the following system categories: Presentation Systems (PRS): These systems are used to project or to receive/scan content for a specific audience (VO-team members). Videoconferencing Systems (VIC): These systems allow to exchange video / audio streams. In most cases data exchange and application sharing is supported as well. Furniture Systems (FUR): These systems shall support ergonomic work as well as easy and fast reconfiguration of space layout and usage scenarios of rooms. Lighting Systems (LIG): These systems support direct, indirect or task lighting. In this report we only analyse artificial, electrical lighting. Glazing and other types of lighting devices are not explained, even though the usage of natural lighting is more sustainable and healthy. Others (OTH): It should be possible to control all the systems above as well as all HVAC-components remotely. At the same time the number of remote control devices should be minimized in order to reduce the additional “learning” efforts for the usage of the electronic equipment. Additionally accessibility to all media (electrical power, computer networks) shall be supported.
Figure 12: ICT-Systems Integration
Page 25 of 34
D4.4
4.2.4 Interrelations: VO-functions - ICT - Facilities & Interiors Table 7 illustrates the proposed requirements specification for the above defined VO work environments. Each line of the matrix specifies the relationship between one facility & interior component and each of the three VO-work environment classes. Each component belongs to one of the facility & interior categories specified in chapter 4.2.3. Table 7: Classification of “embedded” ICT-facilities and relation to facilities and interior system components Facilities & Interior VO Work Environment Type: System Component Single Room
Single WorkspaceTeam Area CSCW-Lab
PRS: Presentation PRS: Beamer
PRS: Smartboard PRS: Document Camera VIC: Video Conf.
CAM: Camera VIC: Cam. with stored positions VIC: Cam. with remote control VIC: Cam. with auto focus INT: Integrated VIC: Integrated video/audio VIC: Headset (mic./earphones) AUD: Audio VIC: Room microphone VIC: Loudspeaker
LIG: Lighting LIG: Task light LIG: Dimming LIG: Indirect FUR: Furniture FUR: Integrated cabling FUR: Reconfigurable tables FUR: Flexible chairs FUR: Roomware OTH: Others OTH: WLAN OTH: Under floor cabling OTH: Central control unit OTH: Remote HVAC control
LEGEND : Not required Suggested Strongly
recommended The following examples illustrate how facility & interior systems influence the performance of IT-infrastructure and the performance of VOs/VEs in general. Example 8:
Presentation devices
To support co-operation of groups it needs real and/or virtual spaces where team members can informally interact among each other. Beamers and smart boards must be installed in a way, that these presentation devices can be used in combination with several sources and software types:
• for presentations and interactive work for VO-team members in one single room, • for presentation and interaction with “remote” audience, • for presentation, interaction, and VC-supported discussion with “remote” audience,
Page 26 of 35
D4.4 Example 9:
Videoconferencing and supporting facilities
Trust and team spirit are basic principles of Virtual Organisations. In locally distributed team-work scenarios it is necessary to support continuous interaction among both individual team members and among groups and teams in a cost efficient, easy to handle way. Video-conferencing units (VCU) are just a tool that might support informal, less structured interaction. However the pure availability is not sufficient enough to support interaction.
A complete set of facilities influences the performance of video conferencing units:
• Natural and artificial lighting systems as well as shading devices influence the performance of video components,
• Glazing influences not only the natural lighting but also the natural heat gain in wintertime and the “over-heating” in summer time.
• Geometry, material selection, flooring and curtains influence acoustics and thus the performance of audio devices,
The list above shows, that the excellent technical performance of the communication tool itself can be influenced negatively through missing or misfunctioning interior systems.
Example 10:
Flexible Furniture
The structure of Virtual Organisations changes quickly and requires not only flexible software tools but also flexible “physical” team-spaces. The layout of rooms and whole buildings must be designed and technically equipped in a way that it supports necessary modifications easily.
One example for adaptable offices is the “Scene-Lab” at Braunschweig University of Technology that was designed by the author and his research group. The central pieces delivering flexibility are the “moveable” meeting table and the central control unit. The “moveable” tables are special constructions. Computers and monitors can be retracted into the tables so that the different use scenarios can be quickly brought about.
Conferencing Scenario - no computers -
Design Scenario -computers and space for drawings-
Computing Scenario- all computers available -
Figure E10.1: Different use-cases of reconfigurable tables Example 11:
Media Control Unit
The central control unit integrates the control functions for presentation devices, the video conferencing unit (including cameras), lighting-devices, the air-conditioning system, and blinds. In this way all devices are controlled from one single point using one single user interface.
The user interface is installed on a moveable, radio-controlled touch panel. Additionally the main unit can be connected to the Internet and thus all devices can be remotely controlled through a web interface.
Figure E11.1: Central Media Control Unit
Page 27 of 36
D4.4 Example 12:
ROOMWARE
Another example for ICT-facilities-integration is the i-Land project at FhG/IPSI in Darmstadt, Germany. Within this project a complete set of furniture, demonstrating the idea of disappearing computers, is being developed.
The InteracTable is the first in a series of information devices that investigates general shapes and orientations of interaction areas. It is designed for displaying, discussing, and annotating information objects by a group of two to six people.
The current version of the InteracTable® is realized with a touch-sensitive plasma-display (PDP). The IT components are mounted below. The margin of the table surface can be used to lean on it and to place additional material.
Since an oval-type display has no selected orientation (e.g., left or right) one has to provide new means of interaction, such as gesture-based interaction with information objects.
Display space on paper or by an electronic information device and the possibility for displaying, annotating and editing large contents is a crucial point for most visually-oriented tasks. Teams in so-called project rooms often use large areas of assembled sheets of paper (usually covering the walls) to create and organize their information.
Furthermore, there is the requirement to be able to interact with “electronic content” in a very intuitive way. The objective of the DynaWall® is to serve as an "interactive electronic wall" represented by a touch-sensitive information device.
The CommChairs combine the functionality of a modern office desktop computer environment and mobility chairs with the comfort of armchairs. Thus, the Communication Chairs allow working in a standard personal computer environment without being bound to desks. The chairs enable users to connect to shared workspaces and devices like the InteracTable® or the DynaWall® as well as edit and annotate objects on these Roomware® components remotely. In the i-LAND project the position and the orientation of each CommChair is tracked.
This information will be processed in order to initiate a coupled session with shared displays between the CommChairs when they are moved together to form a subgroup. This is based on sensor and localization technology.
4.3 Synthesis Many components of integrated ICT-infrastructure systems are available in the participating organisations. Using a specific template each VOSTER partner was asked to describe its facilities & interiors infrastructure supporting the use of its software developed. Furthermore, the partners had to specify the usage intensity of the various components within several VO-phases. However, the usage of the components is not only restricted to the projects contributing to VOSTER. In order to improve the granularity of the response spectrum, a more general, consolidated, qualitative evaluation pattern was developed as depicted below. By introducing a rating by “weights” it became possible to quantify the response spectrum. Table 8: Evaluation pattern for facilities & interior systems and components in my personal office/
work environment within our
department/company used in combination with
our software product for our clients in general (client:=Voster use case)
Available A-Pe = 4 A-De = 4 A-So = 5 A-Cl = 3 Suggested S-Pe= 1 S-De = 2 S-So = 3 S-Cl = 1
Page 28 of 37
D4.4 The availability of a component is rated higher than the suggestion (awareness). The availability / suggestion to use a component in combination with a software is rated highest within one category than the availability within the organisation.
4.3.1 Single Workspaces
SINGLE WORKSPACES - CHARACTERIZATION Key Facets Current Limitation
Furniture (181) • Flexible (ergonomic) chairs
Videoconferencing (155) • Integrated video / audio
Others (149, 130) • Under floor cabling • Wireless LAN
Lighting (127) • Task lights
There are very limited features to support team-oriented work within one single office. Furthermore, information and communication devices remain an “extensions” or “add-ons” to the office environment. Integration and quick re-configuration is neither supported nor easily achievable.
Table 9: “Weighted” evaluation of facilities / interiors for VOs Type: SINGLE WORKSPACES
VO-Phase creation operation dissolution
VO-function
Component VO
Pla
nnin
g
Partn
ers s
earc
h se
lect
ion/
regi
stra
tion
Ente
rpris
e C
atal
ogue
s
Mar
ketp
lace
/ C
lust
er
Con
tract
neg
otia
tion
Dis
tribu
ted
b.p.
pla
nnin
g
Dis
tribu
ted
logi
stic
mgm
t.
Perf
orm
ance
Ass
essm
ent
Proc
ess m
odel
ling
/ si
mul
atio
n Sp
ecia
lized
do
mai
n fu
nctio
ns
Liab
ility
Def
initi
on
Prod
uct /
Ser
vice
life
cycl
e su
ppor
t Pe
rfor
man
ce h
isto
ry
othe
rs
VIC: Video Conf. VIC: Cam. With auto focus 7 2 8 7 7 4 3 7 7 7 59 VIC: Integrated video/audio 17 5 4 25 21 7 8 21 21 17 9 155 VIC: Headset (mic./earphones) 8 8 8 8 8 8 4 8 12 72 VIC: Loudspeaker 12 12 12 4 12 4 12 4 8 4 84LIG: Lighting LIG: Dimming 1 1 1 1 1 1 1 1 1 1 1 11 LIG: Indirect 6 4 8 10 6 9 4 10 5 1 10 1 1 75 LIG: Task light 17 13 2 10 14 13 9 14 10 10 5 10 127
FUR: Furniture FUR: Flexible chairs 12 16 12 17 13 12 13 12 13 12 12 13 12 12 181
OTH: Others OTH: WLAN 10 6 8 8 12 10 11 6 11 10 6 12 10 10 130 OTH: Under floor cabling 10 10 10 12 12 10 11 10 11 10 10 11 9 13 149 100 55 30 53 111 101 76 74 96 96 37 97 38 79 According to the aggregated ratings of the different interior and facilities components depicted in Table 9 one can see that flexible (ergonomic) chairs, integrated video/audio capabilities of PCs/Laptops, under floor cabling, wireless LAN, and individual task lights are defined by the VOSTER members as the most important key facets for individual VO-work environments. In comparison to the other VO-Work environments the VIC-camera auto focus feature and the capability to dim lighting devices are evaluated as less important to support VO-functions in single workspaces. Users request that audio devices shall support both modes, private work with headsets as well as a “public mode” using loudspeakers in order to allow visitors to participate in audio and/or video conferences.
Page 29 of 38
D4.4 The need for reduced cable work to connect hardware devices is expressed through the high ranking for the features of under floor cabling and wireless LAN. However, compared to the team oriented work environments the availability of W-LAN is evaluated as less important. Integrated ICT-support in single workspaces is requested for the VO-functions contract negotiation, VO-planning, and distributed business process planning – informal VO functions or formalized functions with need for interactive (remote) team work.
VIC: Cameraw. auto focus
VIC: Audioloudspeaker
LIG:Indirect
LIG:Dimming
OTH:W-LAN
OTH: Under FloorCabling
VIC: IntegratedVideo / Audio
LIG:Task light
VIC:Audio headset
FUR:Flexible Chairs
VO-planning Contract negotiation
Distributedb.p. planning Process
modeling/Simulation
SpecializedDomain Functions Product / Service
Life Cycle Support
CREATION OPERATION DISSOLUTION
Figure 13: Graphical evaluation of facilities / interiors for VOs Type: SINGLE WORKSPACES
4.3.2 Team Area
TEAM AREAS - CHARACTERIZATION Key Facets Current Limitation
Others (192) • Wireless LAN
Presentation (184, 110) • Beamer, • Smartboard
Others (96) • Under floor cabling
Video Conferencing (80, 64) • Room microphone, • Camera with stored positions
Lighting (70, 64) • Indirect light • Dimming
Furniture (63) • Flexible chairs
Team areas support basic interaction techniques. In most environments presentation devices (beamer) are available.
The need for extended interaction is discovered by most of the projects, even if the facilities are not available today (smartboards, room microphone).
Basic ergonomic standards are fulfilled (flexible chairs). Most projects are aware for improved lighting in order to support better usage of VIC and computer devices.
However, sophisticated infrastructure supporting easy and quick re-configuration of space layouts is neither available nor requested.
Page 30 of 39
D4.4 Table 10: “Weighted” evaluation of facilities / interiors for VOs - Type: TEAM AREA
VO-Phase creation operation dissolution
VO-function
Component VO
Pla
nnin
g
Partn
ers s
earc
h se
lect
ion/
regi
stra
tion
Ente
rpris
e C
atal
ogue
s
Mar
ketp
lace
/ C
lust
er
Con
tract
neg
otia
tion
Dis
tribu
ted
b.p.
pla
nnin
g D
istri
bute
d lo
gist
ic m
gmt.
Perf
orm
ance
Ass
essm
ent
Proc
ess m
odel
ling
/ si
mul
atio
n Sp
ecia
lized
do
mai
n fu
nctio
ns
Liab
ility
Def
initi
on
Prod
uct /
Ser
vice
life
cycl
e su
ppor
t Pe
rfor
man
ce h
isto
ry
othe
rs
PRS: Presentation PRS: Beamer 16 12 4 20 20 16 12 16 16 16 4 12 12 8 184 PRS: Smartboard 8 4 4 6 8 8 6 10 8 12 6 8 6 16 110VIC: Video Conf. VIC: Cam. with stored positions 8 2 8 8 4 6 4 8 4 4 4 4 64 VIC: Cam. with remote control 4 4 6 4 6 4 4 4 8 44 VIC: Cam. with auto focus 4 4 VIC: Integrated video/audio 4 8 8 4 8 4 4 4 16 60 VIC: Room microphone 8 4 12 8 8 8 8 8 4 4 4 4 80 VIC: Loudspeaker 4 4 8 4 8 4 4 4 4 4 48LIG: Lighting LIG: Dimming 8 12 8 4 8 8 4 4 4 4 64 LIG: Indirect 8 10 8 8 4 8 8 4 4 4 4 70FUR: Furniture FUR: Integrated Cabling 4 6 4 4 4 4 4 4 4 38 FUR: “Reconfigurable tables” 4 6 6 4 4 4 4 32 FUR: Flexible chairs 8 1 9 8 5 4 5 7 4 4 4 4 63OTH: Others OTH: WLAN 15 15 15 17 17 15 9 7 17 15 9 15 15 11 192 OTH: Central control unit 4 2 4 4 4 4 4 4 4 34 OTH: Under floor cabling 10 2 2 4 12 10 4 2 12 10 2 10 10 6 96 77 17 17 40 104 73 58 41 72 76 31 53 53 65 The aggregated ratings of the different interior and facilities components depicted in Table 10 clearly illustrates the request for supporting interactive team work functionalities, by evaluating wireless LAN, beamer, and smartboard as the most important ICT-components. Under floor cabling, room microphone, camera with stored positions, indirect lighting, and dimming support easy and comfortable video-conferencing and/or CSCW-scenarios. The importance of flexible (ergonomic) chairs decreases, because it is estimated that team areas are used for only a limited time. ICT-systems integration therefore is more important than ergonomic features. In comparison to especially equipped CSCW-Labs the VIC-camera features are discovered as less important. There is a clear request for separated high quality audio functionality expressed through the high ranking of room microphones and the decreased ranking of integrated audio/video capabilities. The need for reducing cable work to connect hardware devices is not expressed in such a clear way as in the case of single workspaces. However, flexible accessibility of computer networks through W-LAN in team areas is ranked highest. Integrated ICT-support in team areas is requested for the VO-functions contract negotiation, VO-planning, and specialised domain functionalities. These are again informal VO functions or formalized functions with need for interactive team work. However the focus of the secondary VO-functions changes slightly to formalized VO functions with need for interactive local team work.
Page 31 of 40
D4.4
PRS: Beamer PRS: Smartboard
VIC: Cameraw. stored positions
VIC: Cameraw. remote control
VIC: Cameraw. auto focus
VIC: Audioroom microphone
VIC: Audioloudspeaker
LIG:Indirect
LIG:Dimming
FUR: Integrated Cabling
FUR: Recon-figurable Tables
OTH:W-LAN
OTH: Under FloorCabling
OTH: Centralcontrol unit
VIC: IntegratedVideo / Audio
FUR:Flexible Chairs
VO-planning Contract negotiation
Distributedb.p. planning Process
modelling/Simulation
SpecializedDomain Functions Product / Service
Life Cycle Support
PerformanceHistory .
Others
CREATION OPERATION DISSOLUTION
Figure 14: Graphical evaluation of facilities / interiors for VOs Type: TEAM AREA
4.3.3 Integrated CSCW-Lab
INTEGRATED CSCW-LAB - CHARACTERIZATION Key Facets Current Limitation
Presentation (178, 106) • Beamer, • Document camera
Others (147, 83) • Wireless LAN • Under floor cabling
Video Conferencing (83) • Room microphone, • Loudspeaker • Camera with stored positions
& auto focus Furniture (83)
• Flexible chairs
Most of the Voster partners have no access to integrated information and communication facilities. However, most of them require integrated, interactive presentation systems (beamer, document camera and smartboard). Remote control for video conferencing equipment is sometimes available and heavily requested. The need for an integrated environment is underestimated. Not even all partners are aware of the advantages and thus the necessity of systems integration.
Page 32 of 41
D4.4 Table 11: “Weighted” evaluation of facilities / interiors for VOs - Type: INTEGRATED CSCW-LAB
VO-Phase creation operation dissolution
VO-function
Component VO
Pla
nnin
g Pa
rtner
s sea
rch
sele
ctio
n/re
gist
ratio
n En
terp
rise
Cat
alog
ues
Mar
ketp
lace
/ C
lust
er
Con
tract
neg
otia
tion
Dis
tribu
ted
b.p.
pla
nnin
g
Dis
tribu
ted
logi
stic
mgm
t.
Perf
orm
ance
Ass
essm
ent
Proc
ess m
odel
ling
/ si
mul
atio
n Sp
ecia
lized
do
mai
n fu
nctio
ns
Liab
ility
Def
initi
on
Prod
uct /
Ser
vice
life
cycl
e su
ppor
t Pe
rfor
man
ce h
isto
ry
othe
rs
PRS: Presentation
PRS: Beamer 17 13 4 9 14 17 9 17 17 17 9 12 9 14 178
PRS: Smartboard 4 4 4 4 4 3 4 3 4 34 PRS: Document camera 13 2 2 2 13 8 2 5 8 13 8 8 5 17 106VIC: Video Conf. VIC: Cam. with stored positions 8 2 2 2 5 8 2 8 8 8 5 8 5 12 83 VIC: Cam. with remote control 13 13 8 13 8 8 17 80 VIC: Cam. with auto focus 8 2 2 2 5 8 2 8 8 8 5 8 5 12 83 VIC: Room microphone 8 2 2 2 5 8 2 8 8 8 5 8 5 12 83 VIC: Loudspeaker 8 2 2 2 5 8 2 8 8 8 5 8 5 12 83LIG: Lighting LIG: Dimming 2 2 2 2 2 2 2 2 2 2 2 2 2 6 32 LIG: Indirect 2 2 2 2 2 2 2 2 2 2 2 2 2 6 32FUR: Furniture FUR: Integrated Cabling 2 2 2 2 2 2 2 2 2 2 2 2 2 6 32 FUR: “Reconfigurable tables” 2 2 2 2 2 2 2 2 2 2 2 2 2 6 32 FUR: Flexible chairs 8 2 2 2 5 8 2 8 8 8 5 8 5 12 83 FUR: Roomware 2 2 2 2 2 2 2 2 2 2 2 2 2 6 32OTH: Others OTH: Remote HVAC control 2 2 2 2 2 2 2 2 2 2 2 2 2 6 32 OTH: WLAN 13 7 7 7 10 13 7 7 13 13 10 13 10 17 147 OTH: Central control unit 2 2 2 2 2 2 2 2 2 2 2 2 2 6 32 OTH: Under floor cabling 8 2 2 2 5 8 2 8 8 8 5 8 5 12 83 122 48 39 44 98 104 44 91 112 122 74 107 79 183 The aggregated ratings of the different interior and facilities components of CSCW-Labs illustrate either the insufficient evaluation capabilities of such complex systems or the request of users for flexible and modular facilities and interior systems. As shown in Table 11 many of the components are equally evaluated. Only beamer, document camera, and wireless LAN achieve higher individual ratings. Furthermore, users can not clearly specify the VO-functions that need to be supported by especially equipped CSCW-Labs. Both facts might lead to the conclusion that users voted for the usage of standardised, modular facilities and interior systems with moderate complexity. However, the availability of control units, supporting integrated control of multiple ICT functionalities will definitely help to decrease the obstacles for using sophisticated CSCW-systems and contribute to improved, ubiquitous computing.
Page 33 of 42
D4.4
PRS: Beamer PRS: Smartboard
VIC: Cameraw. stored positions
VIC: Cameraw. remote control
VIC: Cameraw. auto focus
VIC: Audioroom microphone
VIC: Audioloudspeaker
LIG:Indirect
LIG:Dimming
FUR: Integrated Cabling
FUR: Recon-figurable Tables
OTH:W-LAN
OTH: Under FloorCabling
OTH: Centralcontrol unit
FUR:Flexible Chairs
VO-planning Contract negotiation
Distributedb.p. planning Process
modelling/Simulation
SpecializedDomain Functions Product / Service
Life Cycle Support
PerformanceHistory
.Others
CREATION OPERATION DISSOLUTION
PRS: Document Cam
OTH: remoteHVAC control
FUR:Roomware
Figure 15: Graphical evaluation of facilities / interiors for VOs - Type: INTEGRATED CSCW LAB
4.4 Summary on facilities and interiors
Considerable progress has been made in terms of design and development of single hardware and interior components to support VO-functionalities but there are still a large number of system integration issues requiring further interdisciplinary research.
In parallel with technological ICT-developments it is necessary to reach some level of integration of ICT-hardware components, building control elements (sensors and actors) and building elements (glazing, façade, shading devices) in order to achieve cross-disciplinary interoperability and to reduce the supporting and engineering efforts still required to launch and operate a VO.
The fact that most VO-teams involved in VO projects lack integrated design expertise (e.g. various projects are dominated by management or IT-experts but with limited background in architecture and systems integration) justifies the fact that many VO projects are simply extend existing facilities, rather than create new “VO-facility-systems”. A better balance in terms of multidisciplinary teams needs to be addressed when creating (designing, building, re-constructing) facilities to be used by VOs in the future.
As a first result of this report one can conclude that ICT-specialists often underestimate the potential contribution of facilities and interior systems to suitable configuration of ICT-infrastructure
Page 34 of 43
D4.4 supporting Virtual Organisations. Consultants and research departments working in multi-disciplinary environments are more aware of these inter-relationships. Many hardware facilities are available on the market, such as smartboards, videoconferencing units, and sophisticated lighting systems. However, most of them are discovered as “single pieces” instead of modules of an integrated system. Only in a few cases these facilities are part of the development of VO-supporting systems from the early beginnings. Process re-organisation and new forms of business organisation need flexible infrastructure SYSTEMS. Virtual organisations need other configurations of the built-environment and its facilities than traditional organisations, such as “open office spaces”, team areas, office-hotels, etc. Quick, easy, and inexpensive reconfiguration of VO-environments consisting of ICT-infrastructure, furniture, lighting systems, moveable wall systems, etc. should be supported in a much better way. The lack of guidelines for total systems integration and appropriate, flexible building design is widely accepted. VO-projects struggle with the availability of flexible, sustainable, easily and efficiently to modify integrated IT- and facility systems. Currently there are only single components available which need to be combined by each VO-partner itself.
The dramatic changes in the organisational patterns as well as the tremendous improvements in the area of software functionalities and software system complexity requires new approaches in integrated facilities design on a solid theoretical and interdisciplinary foundation.
Especially the integration of ICT-components into the built environment such as furniture, flooring, ceiling, windows, HVAC-systems needs to be supported in order to achieve ubiquitous computing. The easy, efficient usage of infrastructure systems will allow the majority of VO-members to concentrate on the “core issues” of VOs instead of being “dominated” by technical issues. However, it is absolutely necessary to support further joint research efforts to achieve this goal.
Page 35 of 44
D4.4
Page 36 of 45
5. Conclusions UNINOVA
Considerable progress has been made in terms of design and development of infrastructures and support services for VOs but there are still a large number of challenging issues requiring further research.
In parallel with technological developments it is mandatory to reach some level of harmonization of models and approaches in order to reach interoperability and to reduce the engineering efforts still required to launch an operative VO. The lack of common reference models and widely accepted (open source?) reference infrastructures still constitutes a major obstacle for the practical implantation of the VO paradigm. Today any VO project still needs to struggle with the development of its own basic infrastructure.
The interoperability problem, although and old issue in systems integration, remains in the agenda. The fact that most teams involved in VO projects lack strong software engineering expertise (e.g. various projects are dominated by experts in the application domain but with limited background in computer science) justifies the fact that almost all VO projects are mainly “followers” of the mainstream (fashion!) in ICT, rather than breakthrough contributors. A better balance in terms of multidisciplinary teams needs to be adopted when establishing VO projects consortia.
The fast evolution of ICT technology also calls for the need of a strong theoretical technology-independent foundation.
References: 1. Virtual Enterprise Modeling and Support Infrastructures: Applying Multi-Agent Systems Approaches,
L.M. Camarinha-Matos, H. Afsarmanesh, (invited paper) in Lecture Notes in Artificial Intelligence LNAI Nº 2086, pp.335-364, Springer, ISSN: 0302-9743, ISBN 3-540-42312-5, July 2001, issue on Multi-Agent Systems and Applications, M. Luck, V. Marik, O. Stpankova, R. Trappl (eds.).
2. Trends in Virtual Organizations and Support Infrastructures, L.M. Camarinha-Matos, H. Afsarmanesh, (invited paper), Journal Systems, Control and Information, Journal of the Institute of Systems, Control and Information Engineers, Japan, Vol. 46, Nº 10, Oct 2002.
3. Dynamic Virtual Organizations, or not so dynamic?, L.M. Camarinha-Matos, H. Afsarmanesh, in Proceedings of BASYS’02 - Knowledge and technology integration in production and services, Kluwer Academic Publishers, ISBN 1-4020-7211-2, pp. 111-124, Sept 2002.