collaborative negotiation for ontology-driven enterprise businesses

10
Collaborative negotiation for ontology-driven enterprise businesses Ricardo Jardim-Goncalves a, *, Carlos Coutinho b , Adina Cretan c , Catarina Ferreira da Silva d , Parisa Ghodous d a Faculty of Sciences and Technology, UNINOVA, New University of Lisbon, Caparica 2829-516, Portugal b Caixa Ma ´gica Software, Lisbon 1600-196, Portugal c Computer Science Department, ‘‘Nicolae Titulescu’’ University of Bucharest, Bucharest 031046, Romania d Lyon Research Centre for Images and Intelligent Information Systems, Claude Bernard Lyon 1 University, Villeurbanne 69100, France 1. Introduction The advent of the globalisation and the Internet have led to the development of various forms of virtual collaboration in which the organisations are trying to exploit the facilities of the network to achieve higher utilisation of their resources. This means enter- prises (particularly Small and Medium Enterprises (SMEs)) are leaving their traditional vision of having a corporation focused into the development of one or more complete products, to one where the focus is on specialisation and focus on particular business functionalities and services, which are then integrated with other providers to build complete solutions. In this collaborative networked environment, enterprises are developing business areas dedicated to finding and complying with the best set of partners and suppliers for solutions that are aligned with the enterprise’s strategy. Enterprise integration, interoperability and networking are some of the major disciplines that are enabling companies to improve collaboration and communication [1]. To be able to perform, enterprises need to exchange informa- tion, whether this exchange is internal (among departments of the enterprise), external (among the enterprise or part of it and an external party), or both. Enterprise Interoperability (EI) is thus defined as the ability of an enterprise to seamlessly exchange information in all the above cases, ensuring the understanding of the exchanged information in the same way by all the involved parties [2]. EI concerns the alignment or common understanding in various forms and levels, from middleware (e.g., signalling, interfaces, and platforms) to semantics (e.g., concepts, taxonomies, and relationships), workflows and methodologies, solutions and business visions. In an integrated, tight and interoperable environment, EI can be challenged due to various conditions and reasons, e.g., if a new enterprise which joined the ecosystem introduces new conditions, or if one of the existing parties has made a change in the existing interoperable environment. Large corporation enterprises try to contour this problem by setting market standards and influencing the other integrating to comply with these standards. SMEs usually do not have the empowerment to do so, and are therefore more sensible to the oscillations of the environment that involves them, which leads them to the need to constantly change to interoperate with their surrounding ecosystem. As the number of integrating parties grows, and as business interactions and interactions become more complex, the amounts of information that are exchanged become very hard to handle. These assets concern various forms and formats, including e.g., abstract concepts, flows, strategies, hierarchies, culture and tacit knowledge, to more down-to-earth files, records and other business evidences. The path to success on enterprises currently includes storing as much and as detailed information as possible, in order to be able to infer conclusions, tendencies, behaviours and needs from this information. In a recent interview, the IT director from Virgin Atlantic stated their Computers in Industry xxx (2014) xxx–xxx A R T I C L E I N F O Article history: Received 19 November 2013 Accepted 6 January 2014 Available online xxx Keywords: Enterprise Interoperability Ontology Conflict resolution Negotiation Enterprise Information Systems A B S T R A C T The requirements from a globalised world demand that enterprises not only shift their paradigm from product-centrism to component-centrism on integrated products, potentiating the need for tight interoperability dependencies, but also that the product specifications and concepts are fully understood by customers and providers in a transparent manner that surpasses the barriers of language, culture and technology. This paper presents the NEGOSEIO framework, which enables service-based interoperability between parties, closely integrated with semantics and business understanding via the use of reference ontologies in the quest for achieving a stronger interoperability liaison. The paper’s validation and discussion is performed in its application on the ontology negotiation of business environments in the scope of the EU-funded FP7 project TIMBUS for digital preservation of resources and enduring business continuity. ß 2014 Elsevier B.V. All rights reserved. * Corresponding author. E-mail addresses: [email protected] (R. Jardim-Goncalves), [email protected] (C. Coutinho). G Model COMIND-2541; No. of Pages 10 Please cite this article in press as: R. Jardim-Goncalves, et al., Collaborative negotiation for ontology-driven enterprise businesses, Comput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.01.001 Contents lists available at ScienceDirect Computers in Industry jo ur n al ho m epag e: ww w.els evier .c om /lo cat e/co mp in d 0166-3615/$ see front matter ß 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.compind.2014.01.001

Upload: parisa

Post on 24-Dec-2016

229 views

Category:

Documents


4 download

TRANSCRIPT

Page 1: Collaborative negotiation for ontology-driven enterprise businesses

Computers in Industry xxx (2014) xxx–xxx

G Model

COMIND-2541; No. of Pages 10

Collaborative negotiation for ontology-driven enterprise businesses

Ricardo Jardim-Goncalves a,*, Carlos Coutinho b, Adina Cretan c, Catarina Ferreira da Silva d,Parisa Ghodous d

a Faculty of Sciences and Technology, UNINOVA, New University of Lisbon, Caparica 2829-516, Portugalb Caixa Magica Software, Lisbon 1600-196, Portugalc Computer Science Department, ‘‘Nicolae Titulescu’’ University of Bucharest, Bucharest 031046, Romaniad Lyon Research Centre for Images and Intelligent Information Systems, Claude Bernard Lyon 1 University, Villeurbanne 69100, France

A R T I C L E I N F O

Article history:

Received 19 November 2013

Accepted 6 January 2014

Available online xxx

Keywords:

Enterprise Interoperability

Ontology

Conflict resolution

Negotiation

Enterprise Information Systems

A B S T R A C T

The requirements from a globalised world demand that enterprises not only shift their paradigm from

product-centrism to component-centrism on integrated products, potentiating the need for tight

interoperability dependencies, but also that the product specifications and concepts are fully understood

by customers and providers in a transparent manner that surpasses the barriers of language, culture and

technology. This paper presents the NEGOSEIO framework, which enables service-based interoperability

between parties, closely integrated with semantics and business understanding via the use of reference

ontologies in the quest for achieving a stronger interoperability liaison. The paper’s validation and

discussion is performed in its application on the ontology negotiation of business environments in the

scope of the EU-funded FP7 project TIMBUS for digital preservation of resources and enduring business

continuity.

� 2014 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Computers in Industry

jo ur n al ho m epag e: ww w.els evier . c om / lo cat e/co mp in d

1. Introduction

The advent of the globalisation and the Internet have led to thedevelopment of various forms of virtual collaboration in which theorganisations are trying to exploit the facilities of the network toachieve higher utilisation of their resources. This means enter-prises (particularly Small and Medium Enterprises (SMEs)) areleaving their traditional vision of having a corporation focused intothe development of one or more complete products, to one wherethe focus is on specialisation and focus on particular businessfunctionalities and services, which are then integrated with otherproviders to build complete solutions. In this collaborativenetworked environment, enterprises are developing businessareas dedicated to finding and complying with the best set ofpartners and suppliers for solutions that are aligned with theenterprise’s strategy. Enterprise integration, interoperability andnetworking are some of the major disciplines that are enablingcompanies to improve collaboration and communication [1].

To be able to perform, enterprises need to exchange informa-tion, whether this exchange is internal (among departments of theenterprise), external (among the enterprise or part of it and anexternal party), or both. Enterprise Interoperability (EI) is thusdefined as the ability of an enterprise to seamlessly exchange

* Corresponding author.

E-mail addresses: [email protected] (R. Jardim-Goncalves),

[email protected] (C. Coutinho).

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

0166-3615/$ – see front matter � 2014 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.compind.2014.01.001

information in all the above cases, ensuring the understanding ofthe exchanged information in the same way by all the involvedparties [2]. EI concerns the alignment or common understanding invarious forms and levels, from middleware (e.g., signalling,interfaces, and platforms) to semantics (e.g., concepts, taxonomies,and relationships), workflows and methodologies, solutions andbusiness visions. In an integrated, tight and interoperableenvironment, EI can be challenged due to various conditionsand reasons, e.g., if a new enterprise which joined the ecosystemintroduces new conditions, or if one of the existing parties hasmade a change in the existing interoperable environment.

Large corporation enterprises try to contour this problem bysetting market standards and influencing the other integrating tocomply with these standards. SMEs usually do not have theempowerment to do so, and are therefore more sensible to theoscillations of the environment that involves them, which leadsthem to the need to constantly change to interoperate with theirsurrounding ecosystem. As the number of integrating partiesgrows, and as business interactions and interactions become morecomplex, the amounts of information that are exchanged becomevery hard to handle. These assets concern various forms andformats, including e.g., abstract concepts, flows, strategies,hierarchies, culture and tacit knowledge, to more down-to-earthfiles, records and other business evidences. The path to success onenterprises currently includes storing as much and as detailedinformation as possible, in order to be able to infer conclusions,tendencies, behaviours and needs from this information. In arecent interview, the IT director from Virgin Atlantic stated their

borative negotiation for ontology-driven enterprise businesses,01.001

Page 2: Collaborative negotiation for ontology-driven enterprise businesses

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx2

G Model

COMIND-2541; No. of Pages 10

new 787 fleet is expected to create more than half terabyte of data foreach flight [3]. As businesses grow every day, so does the amount ofinformation that is produced. In a study conducted by IDC [4], digitalarchive growth will reach data amounts of 40 ZB (Zettabytes) by2020. The main issue is that the number of options and decisionstowards interoperability that need to be taken is escalating innumber and importance, and these decisions have and increasinginfluence on the evolution of the enterprise business.

Sustainable EI (SEI) may be defined as the ability of maintainingand enduring the EI along the enterprise’s systems and applications’life cycle. Achieving a SEI in this context requires a continuousmaintenance and iterative effort to adapt to new conditions andpartners, and a constant check of the status and maintaining existinginteroperabilities [5]. This paper recommends negotiations as theproper way to resolve interoperability discrepancies. Negotiationsallow a proper formalisation of the SWOT (strengths, weaknesses,opportunities and threats), while presenting the various alternativesolutions. It also enables a clear definition of criteria (e.g., impact,downtime, cost, alignment with enterprise strategy, new markets)for the selection of a particular solution, providing a strongjustification for reaching mature choices.

Negotiations are sets of complex actions (e.g., creation of newproposals, evaluation of proposals by acceptance or rejection),some of which may occur in parallel. In a negotiation, multipleparticipants exchange and take decisions in multiple phases over aset of multiple attributes of the negotiation objects (e.g., price, size,quality). In order to formalise and model the negotiation processes,the metaphor of Interaction Abstract Machines (IAMs) is adopted[6]. This approach allows the modelling of the evolution of multiplenegotiation phases in parallel and including non-deterministicaspects.

The negotiation process towards interoperability needs to besupported by a flexible and consistent set of collaboration tools. Amajor improvement in the last decade on this field was achieved bythe development of Service-Oriented Architectures (SOA) [7,8].Web Services have reshaped the existing concepts of solutiondeployment and provisioning, and paved the way for otherimportant concepts using the same paradigm, like functionaldiscovery and subscription in common repositories, orchestrationand composition of services into more complex ones [9], thuscreating highly modular and flexible solutions as required.

While SOA is meaningful in terms of solutions flexibility, it haslimitations with respect to deployment. The advent of distributedcomputing complements SOA to deal with scalability issues, andthe emergent cloud-based solutions are the exponent of thisdevelopment [10].

This paper proposes a framework that offers negotiationmechanisms to support the sustainability of interoperability inbusiness-to-business interactions, in networked enterprise envir-onments. Additionally, the presented approach tackles the issue ofsemantic heterogeneity of such an environment by introducingontologies as the main support in the negotiation process. Section2 describes the problem space and the research questions thatrespond to the described problem. Section 3 presents thetheoretical background for this paper (literature review). Section4 proposes the collaborative negotiation framework in the contextof EI, specifying the negotiation coordination model, the proposedmethodology for sustainability of EI and the architecture of theproposed solution. Section 5 analyses the application of theproposed methodology into a real business case scenario. Finally,Section 6 presents the conclusions and final considerations.

2. Challenges and research questions

The idea proposed by this paper is to search for solutions toachieve and maintain the interoperability between enterprises, not

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

by imposing a new model but by negotiating the changes in thecollaborative networked environment towards reaching a bettersolution. This approach leads to improved and mature solutions forinteroperability, standards and best-practices that will benefit allplayers. According to the presented approach regarding thenegotiation, the participants to a negotiation may propose offersand each participant may decide in an autonomous manner to stopa negotiation either by accepting or by rejecting the received offer.Also, depending on its role in a negotiation (e.g., initiator or guest) aparticipant may invite new participants in the negotiation process.

The business-to-business interaction context in which thesenegotiations take place forces us to model the unexpected and thedynamic aspects of this environment. An organisation mayparticipate in several parallel negotiations. Each negotiationmay end with the acceptance of a contract that will automaticallyreduce the available resources and it will modify the context forthe remaining negotiations. This dynamic evolution of the contexthas been modelled using the metaphor of Interaction AbstractMachines (IAMs), which allows us to limit the acceptance of anegotiation to the available set of resources [11].

To respond to the problems identified above, the following keyopen research question and hypothesis for solution are formulat-ed:

� Can negotiation promote the establishment of stronger ontologyspecifications, in order to push businesses towards improvingthe sustainability of the Enterprise Interoperability?

Under this consideration, a set of hypotheses is presented:

� If business parties detect that changes need to be performed toreach interoperability, negotiation provides an appropriatemethod to find suitable solutions;� If negotiation proves to be an effective way to improve the

definition of the knowledge ontologies, this will reduce therecovery the time spent performing the (re-)establishment ofinteroperability.

3. Literature review

Achieving a sustainable interoperability among organisationsin a networked environment is a crucial factor in order tosuccessfully manage collaborations at all levels: abstract (busi-ness); concrete (technology), including:

� informational (information vs. data);� functional (activity vs. service);� behavioural (process vs. workflow).

Outlining the crucial position of information systems (IS) insidean organisation, Benaben et al. [12] state that the main issue is toensure that the IS of the partners involved in the collaboration areable to work together to constitute a coherent and homogeneousset of IS – the IS of the collaborative situation. To address this issue,Benaben and Pingaud [13] propose the Mediation InformationSystem Engineering Project (MISE Project) which aims at providingcollaborating organisations with a mediation information system(MIS) able to support the interoperability of a collaborativenetwork. The project takes a model-driven approach to develop acomplete MIS design method, taking into account the semanticreconciliation between business and technical levels.

In the same area, Coutinho et al. [14] define a framework tosupport Sustainable Interoperability using Model-Driven Archi-tectures (MDA), Model-Driven Interoperability (MDI), Service-Oriented Architectures (SOA) and Ontologies. This framework

borative negotiation for ontology-driven enterprise businesses,01.001

Page 3: Collaborative negotiation for ontology-driven enterprise businesses

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx 3

G Model

COMIND-2541; No. of Pages 10

allows businesses to build higher inter-knowledge to achievestronger interoperability. Agostinho et al. [15] propose a frame-work which applies MDA transformations to data models tomaintain an interoperable peer-to-peer (P2P) connection betweentwo applications. According to Panetto [16], the Model DrivenInteroperability (MDI) is considered a major methodology forachieving Enterprise Interoperability (EI), adopting MDA layers forthe development of a model-morphism that implements thetransformations among different enterprise models in thedeployment of interoperable enterprise systems.

Jardim-Goncalves et al. [17] state that interoperability issueshave arisen when using instances of meta-models from differentsources, and identify semantic annotation, ontology harmoniza-tion, and merging as examples of important methods for theEnterprise Interoperability Science Base (EISB).

An Enterprise Information System is generally composed of amultitude of applications able to answer certain enterprise needs.Izza [18] considers the integration of Enterprise InformationSystems a crucial problem due to the applications composing theinformation systems of the companies that increasingly requireworking together. The author states that the heterogeneity ofenterprise applications is the major challenge of the integrationproblem due to the multiple technical, syntactical and semanticconflicts that concern these applications. This requires a mediationprocess to deal with these differences.

Many research papers [19] take the approach of usingontologies to address the semantic integration and interoperabilityissues, to deal with the semantic heterogeneity of such anenvironment. Zdravkovic et al. [20] takes the approach of semanticenrichment of the Supply Chain Operations Reference (SCOR)model using Web Ontology Language (OWL), enabling effectiveknowledge management in supply chain networks and facilitatingthe semantic interoperability of systems.

To support the continuous evolution of ontologies, Khattak et al.[21] propose to re-establish the mappings among dynamicontologies by using the changing history of ontology. This hasthe benefit of reducing the time required for reconciling mappingsamong ontologies, compared to already existing systems thatcompletely reinitiate the process.

Also, ontologies play an important role in the development ofMulti-Agent Systems (MAS) for the semantic web due to theheterogeneity of agents. Thus, Laclavık et al. [22] state that the lackof the interconnection with semantic web standards such as OWLis the main disadvantage of MAS. In this respect, the authorsdevelop a semantic knowledge agent model that can be used in anagent-based application where such interconnection is needed.

The issue of using a common ontology has been approached inmany works, such as the one of Torres and Wijnands [23]. Althoughbeneficial in many ways, the use of a common ontology becomesmuch more complex when it deals with multiple application fieldsin what regards creation, updates and efficient structure. In this

Fig. 1. The methodology for building the

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

respect, Sarraipa et al. [24] present MENTOR methodology basedon the mediator ontology concept which assists the semantictransformation among each organisation’s ontology and thereferential one. Additionally, the authors propose to use MENTORas the collaborative ontology-building methodology [25], enrichedwith Qualitative Information Collection Methods (QICM), in orderto improve the approach to elicit knowledge from business domainexperts.

The increasing exchange of knowledge, resources and expertiseamong virtual organisations in a collaborative environment has ledto many conflicting situations. For solving the conflicts, differentkinds of research approaches have been applied, from automaticresolution [26] to mediated resolution approach [27]. Later works[28,29] combine automatic resolution with social mediation forresolving conflicts among users. According to the authors, theautomatic resolution approach is used when the decision is simpleor close to what all users expects, while the social mediationinvolves negotiating a resolution, and is performed by recom-mending possible candidates. It is used when the decision iscomplex or different from what at least one of the users expects.

The negotiation approach plays a key role in solving theconflicts that may occur in a collaborative dynamic environment[30]. However, the inadaptability of agents to evolving negotiationprotocols, and the ambiguity of the agents’ negotiation term arethe main issues that can arise during agent interactions [31,32].Thus, semantic interoperability is an important issue in anetworked enterprise [33]. The same idea of using of ontologiestechnology in order to settle the knowledge conflicts and to solvesemantic ambiguity has been extended into the field of automatednegotiation research [34]. In this regard, Wang et al. [35] proposean ontology-based knowledge representation approach to providea semantic interoperable environment to realise automaticnegotiations in a virtual collaborative environment.

4. Collaborative negotiation framework

The proposed Collaborative Framework offers mechanisms tosupport negotiations in a distributed environment. The proposedframework was named NEGOSEIO, which stands for Negotiationsfor Sustainable Enterprise Interoperability with Ontologies; itincludes a methodology for the definition of the different processesfor capturing knowledge and modelling the interoperable envi-ronment, a formal negotiation model to represent the negotiationsteps, strategies, and the knowledge coming from the negotiation[36], and an infrastructure for handling, formalising and persistingthe negotiation activities.

4.1. Framework methodology

The NEGOSEIO methodology, as seen in Fig. 1, consists on a setof formal steps to build the knowledge considered necessary for a

NEGOSEIO Collaborative Framework.

borative negotiation for ontology-driven enterprise businesses,01.001

Page 4: Collaborative negotiation for ontology-driven enterprise businesses

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx4

G Model

COMIND-2541; No. of Pages 10

strong interoperability connection, implementing the foundationsteps for Enterprise Interoperability defined by Jardim-Goncalveset al. [37], relying on principles that allow interoperability tobecome reinforced.

The first step is to model the basic foundations (services andinfrastructure) of the framework in a Model-Driven Architecture’sComputational-Independent Model (MDA CIM), which defines thenegotiation concepts (e.g., the IAM states). The authors propose theuse of Model-Driven Interoperability (MDI) to implement theinteroperability solutions and transformations throughout all MDAabstraction levels, using negotiation to handle the interoperabilityneeds and information exchanges (see Fig. 4). In this sense, a CIM-levelled abstraction model is created consisting on sharedunderstanding regarding concepts, business models, semanticreasoning, and related aspects of the negotiating parties (e.g.,agreed term meanings, negotiation behaviours). These models arethen transformed into corresponding Platform-Independent Mod-els (PIM) where semantic ontologies are defined and reasoned intoa proper reference ontology, using the MENTOR methodology (seeFig. 2) but improved to include negotiation of the ontologicalrelationships. Furthermore, these models are transformed intoeach negotiation partner’s specific Platform-Specific Models (PSM)such as databases and services [5]. This way, any interoperabilitychange should be reflected in the corresponding model and thus beable to be assimilated by the other parties [38].

One important aspect of the proposed methodology is that ithighlights the fact that in all the interoperability building processnegotiation is always present, firstly to build the foundations forinteroperability: negotiation on the establishment of referenceontologies, negotiation on multiple abstraction layers of theenterprise artefacts, negotiation on the selection of the supplychain, and on the provided services definitions, to the negotiation ofthe platforms to be used for deployment. This forms the basicplatform for interoperability between the various interested parties.

The interoperating parties shall then be able to make use of theNEGOSEIO infrastructure (see Section 4.3), a cloud-based set ofservices, including machine infrastructure for negotiation (Infra-structure as a Service – IaaS), Platforms for development andestablishment of integration activities (Platform as a Service –PaaS), and negotiation services (Software as a Service – SaaS), for

Fig. 2. The MENTOR

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

performing the multiple negotiations and for handling theinteroperability ‘‘node’’ between the involved parties.

After this interoperability baseline is achieved, the NEGOSEIOmethodology specifies that continuous maintenance with the useof negotiation must be applied to it in order to perform theplatform and service adaptations coming from any interoperabilitychanges to implement the principles to improve the sustainabilityof EI. These may encompass changes to the ontologies andontological relationships as new concepts are embraced andevolved, and the reference ontology is updated, as well as thecontinuous negotiation rules that are evolving according to thepresented proposals for interoperability change and are used tohelp on the decision-support activities.

4.2. Negotiation process for improving the sustainability of EI

Different negotiation scenarios that can be found in the lifecycles of the business to business negotiations can be modelled.Some of these scenarios can range from simple cases of selection ofpossible partners and a direct outsourcing of a job, to morecomplex scenarios of concurrent negotiations with multiplepartners to outsource a non-divided job or concurrent negotiationswith the possibility to dynamically split the job during thenegotiation, as shown by the authors [11].

This process can be exemplified on a simple scenario, where aconflict occurring in a network of enterprises threatens tojeopardise the interoperability of the entire system. Hence, thisprocess (as well as the whole methodology of promoting acontinuous maintenance) is expected to detect these interopera-bility problems as early as possible, hopefully together with a riskmanagement strategy. The worst situation will be if theinteroperability loss risk is already occurring or if the occurrencehad not been foreseen by a risk, but these can be seen as specialcases, and the same process can be applied as a contingency plan.

The first step of the process to prevent the loss of interoperabili-ty consists in identifying the Enterprise Interoperability issue. Thefollowing steps refer to analysing the problem, evaluating possiblesolutions and selecting the solution. The proposed solution forconflict resolution regards reaching a mutual agreement throughnegotiation. The benefit of this approach is the possibility to reach

methodology.

borative negotiation for ontology-driven enterprise businesses,01.001

Page 5: Collaborative negotiation for ontology-driven enterprise businesses

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx 5

G Model

COMIND-2541; No. of Pages 10

a much more stable solution, unanimously accepted, and in ashorter period of time.

The design and coordination of the negotiation process musttake into consideration:

� Timing (the time for the negotiation process will be pre-set);� The set of participants to the negotiation process (which can be

involved simultaneous in one or more bilateral negotiations);� The set of simultaneous negotiations on the same negotiation

object, which must follow a set of coordination policies/rules;� The set of coordination policies established by a certain

participant and focused on a series of bilateral negotiations;� The strategy/decision algorithm responsible for proposals

creation;� The common reference ontology, consisting of a set of definitions

of the attributes used in negotiation.

The negotiation process begins when one of the enterprisesinitiates a negotiation proposal towards another enterprise, on achosen negotiation object. This enterprise shall be named theInitiating Enterprise (E1). This enterprise then selects thenegotiation partners and sets the negotiation conditions (for

Algorithm: Pseudo code representation of the negotiation process

Inputs: Enterprises E1, . . ., En

Outputs: The possible state of a negotiation: success, failure

BEGINon receive s tart f rom E1 {

send in itial offer to partner ;}on receive o ffer f rom partne r {

evaluate off er;if (con ditio ns set by the NO/NF ar e not

offer is rej ecte d;if (tim e all ows i t) {

send ne w off er to par tne r;} els e {

exit;} end i f;

} els e {send new off er to ano ther par tner

} end i f;If (rec eive an acc ept ed offe r) {

If (off er is accep ted in all bila success;

} els e {If (tim e all ows i t) {

send ne w off er to pa } els e {

failure;} end i f;

} end i f;If (rec eive a reje cte d offer ) {

If (off er is activ e in othe r bila failure in a ll neg oti ation

} end i f;} end i f;

}END

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

example sets the timing for the negotiation). The negotiationpartners are represented by all enterprises on which the proposedchange has an impact. It is assumed for this example that thisinformation is available to E1 (if not, the first step would consist ina simple negotiation in which all enterprises are invited toparticipate at the negotiation of the identified solution. Hence, it isassumed that all enterprises which are impacted by this changewill accept the negotiation).

After the selection of the set of invited enterprises (E2, . . ., En),E1 starts bilateral negotiations with each guest enterprise bysending of a first proposal. For all these bilateral negotiations, E1sets a series of coordination policies/rules (setting the conditionsfor the mechanism of creation and acceptance of proposals) and anegotiation object/framework (NO/NF) represented by an inter-operability node (see Fig. 3), setting the limits of solutionsacceptable for E1. Similarly, the invited enterprises set their ownseries of coordination policies and a negotiation object/frameworkfor the ongoing negotiation.

After the first offer sent by E1, each invited enterprise has thepossibility to accept, reject or send a counter offer. For each offerthat is sent, the participating enterprises E1, E2, . . ., En, follow thesame algorithm:

met) {

;

teral n ego tiations ) {

rtne r;

teral n ego tiations ) {s;

borative negotiation for ontology-driven enterprise businesses,01.001

Page 6: Collaborative negotiation for ontology-driven enterprise businesses

Fig. 3. NEGOSEIO’s interoperability solutions negotiation node.

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx6

G Model

COMIND-2541; No. of Pages 10

This algorithm can then make use of the negotiationstrategies and of the negotiation persistence model defined inRef. [36], e.g., Subcontracting, Block, Divide, which are thenimplemented in the infrastructure of NEGOSEIO (see Section 4.3)in the shape and scope of a interoperability solutions negotia-tion node.

These interoperability nodes are defined in NEGOSEIO as alogical relationship that interconnects two or more enterpriseswhich intend to become interoperable. Multiple nodes can be usedto represent all the interoperability solutions negotiation occur-ring in the system for each enterprise or with other enterprises, ascan be seen in Fig. 3.

These strategies and algorithms have been implemented andvalidated in the NEGOSEIO development prototypes [39], built ontop of a Multi-Agent System (MAS) developed in the JADE

Fig. 4. Multiple negotiations for

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

framework [40], consisting on a central node (‘‘System Controller’’)and computation agents on each interacting enterprise (‘‘TriggerAgent’’).

The two entities above can be specified as:

� System Controller is the central point of the negotiation nodesystem, responsible to control the interactions between all clientagents (Negotiation Managers) connected through the MAS. Thisapplication is responsible for controlling the negotiations in theenvironment, the negotiation flows and also to preserve thenegotiation knowledge in order to be capable of helping theclients in the negotiations decisions. The negotiation steps arethen stored in the shape of knowledge rules powered by anNEGOSEIO’s inference engine (see Fig. 5), which was imple-mented using the Drools rules engine [41].

Enterprise Interoperability.

borative negotiation for ontology-driven enterprise businesses,01.001

Page 7: Collaborative negotiation for ontology-driven enterprise businesses

Fig. 5. The NEGOSEIO architecture.

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx 7

G Model

COMIND-2541; No. of Pages 10

� Each Trigger Agent it is connected to the System Controller. Thisapplication, besides performing all communications with theSystem Controller, also has the task to fire a trigger when achange occurs in the Trigger Agent system. When this happens,the trigger notifies the System Controller that a negotiationround should start to handle the change. The negotiation moduleis responsible to notify the user that is necessary to make anaction over the current negotiation round.

4.3. NEGOSEIO infrastructure and architecture

As stated above, the NEGOSEIO methodology’s initial stepsperform the capturing of the knowledge about the involvedenterprises, particularly information coming from each businessontology, and also reasoning and negotiating the definitions for areference ontology. This knowledge will be the foundation for thedevelopment of the MDA/MDI models of each enterprise, and thebasis for the development/customisation of the various specificnegotiation services involved between the enterprises.

Fig. 4 describes the set of specialised services which are builtaccording with the influence of this knowledge, and that willdeveloped to improve the sustainability of EI. The resulting multi-layered infrastructure is developed as a set of overlappingservices, starting from the Middleware layer which aims toensure the handling of all heterogeneity issues (e.g., communica-tions, syntax, session, data) of the basic middleware level of thenegotiation. On top of this layer, a richer set of services(Coordination Services layer) shall be built. These servicesperform complex activities like transaction management, thedefinition and management of the negotiation data model, storageand management of the negotiation data and of the businessstates which implement the negotiation model rules. This layer isresponsible for handling semantic discrepancies between thenegotiating parties via the use of one or more ontologies [42]. Italso includes an agent-based architecture [43] to support thecomplexities of the negotiation operations through the Middle-ware layer [44] (as seen in Fig. 3).

The Negotiation Manager, residing on the client enterprise side(see Fig. 3), implements the business decisions that need to betaken for the negotiation, e.g., starting a new negotiation, invitingother parties to join the negotiation, making proposals to thenegotiation, and accepting or rejecting proposals.

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

The overall architecture of NEGOSEIO can then be seen in Fig. 5,which presents the distribution of the proposed modules in thevarious negotiation levels, according to its corresponding func-tionality.

Regarding the information exchange, behaviour and otheraspects of interoperability itself, the proposed framework wasbuilt using the popular, flexible and robust services in SOA. In orderto manage the issues regarding size and scalability, the SOAplatform for the framework is implemented on top of a cloud-based system.

Additionally, data access, its models and data exchange can alsobe a problem for interoperability. Negotiation parameters, ontol-ogies and other entities rely on data modelling, specification andconsistency and therefore the best way to define the data modelsand the data exchange is to use a standard. In this case the selectedsolution is to model the data for databases and data access using theISO10303 STEP and EXPRESS language specification [45,46].

Finally, to allow a sustainable, flexible and generic approachtowards the implementation of the underlying infrastructure, aCloud web-service-based platform is proposed for this framework.Considering the current trend on services provisioning and that thecomplexity of the involved interactions may be from simple to verydemanding, the proposed implementation for it encompasses thesubscription of cloud-based services for this matter (IaaS, PaaS andSaaS). These highly flexible infrastructures and service providerplatforms allow the subscription of simple or more demandingprocessing according to the negotiation environment needs.

5. Business case scenario for framework validation

This paper is being validated by its implementation in the scopeof EU co-funded FP7 project TIMBUS [47], where NEGOSEIO isproposed as a solution for determining the interoperabilitysolutions between the enterprises seeking to harmonise theknowledge ontologies between the various parties for the purposeof capturing and structuring business-related knowledge for long-term digital preservation.

TIMBUS stands for ‘‘Digital Preservation for Timeless BusinessProcesses and Services’’, and focuses on having resilient businessprocesses, based on metadata information stored in businessontologies. In order to maintain the business continuity of anenterprise, TIMBUS has the objective to ensure continued access toservices and business assets, providing processes and methods for

borative negotiation for ontology-driven enterprise businesses,01.001

Page 8: Collaborative negotiation for ontology-driven enterprise businesses

Fig. 6. Acquisition of business knowledge metadata.

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx8

G Model

COMIND-2541; No. of Pages 10

digitally preserving business processes. It includes a flexible andextensible software architecture, intelligent tools and technologiesfor performing the major functionalities of business and knowl-edge capturing, analysis and maintenance, preservation andstorage of the information and meta-information, and futureredeployment of the stored information to knowledge [48].

On its initial step, acquisition of business knowledge, the focusconcerns determining, capturing, modelling and indexing businessknowledge into an information system, which is a very challengingtask. The information model needs to be flexible enough to handlethe heterogeneity in concepts and business information, able tocomprise information from a business environment consisting inmultiple systems, computers and other devices, people and othersources of information.

The proposed approach is to have an open architecture which isable to expand its ability to contact these information sources. Inthis sense it is able to retrieve information with more or lesshuman intervention (i.e., although it is capable to access theinformation sources directly, it must be also able to receiveinformation retrieved by other tools or by human intervention, thisis relevant to handle some tricky security problems and to enhanceconfidence on the target systems) [49]. The proposed method is todevelop generic extraction mechanisms, ones which are able toattach numerous specifically implemented extractors or adapters,which then are able to retrieve correctly the information.

Fig. 6 shows the overall architecture of the Context Acquisitionmodule, which works in a Multi-Agent System (MAS) paradigm,with dedicated and tailored agents extracting business metadatafrom the target environment. A central acquisition module gathersthe inputs from the multiple agents running on heterogeneousoperating systems, environments and platforms, legacy systemsand records and business tacit knowledge captured by ques-tionnaires and interviews, reasons these inputs, and models andstores the data into the information system.

The proposed organisation of the stored metadata is to storeconcepts and relationships into an information structure thatproperly consolidates this information, supported by a database topersist the specific data. The schema for the central informationsystem (known as Business Information Context) needed to bedefined and tailored in a modelling format, which needs to be the bestto capture all relevant aspects of an organisation and its IT systems.

In TIMBUS, the need is to capture Enterprise Architectureinformation for a very broad purpose, including e.g., businessprocesses, technological architectures and infrastructures, regula-tions and legislation, software applications. More than modelling

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

the entities themselves, the project aims to also model thedependencies between the business assets. It was found verydifficult to come up with a single modelling structure that wouldcover all possible business variations. Hence, Antunes et al.proposed [50] for that purpose the use of ontologies, particularlythe development of an extensible domain knowledge model,comprised by one generic ‘‘common’’ ontology, called the Domain-Independent Ontology (DIO), whose structure is defined by on theOpen Group standard ArchiMate 2.0 modelling language’s meta-model [51]. This base ontology is then complemented by otherDomain-Specific Ontologies (DSOs), which are tailored to handlethe specific needs of the targeted business.

As each business has its specificities, the ontologies are verydiverse and are prone to several misinterpretations. To a particularbusiness partner who is a furniture provider, the concept of a gateis totally different from a software provider, or from an electronicsprovider. Even in the same business, when defining concepts andrequirements it is common to have slight differences inknowledge; this is where the proposed platform presents itsadded-value, enabling negotiation mechanisms and strategies toharmonise the ontologies representing business knowledge [50].

The project is being validated in multiple industrial businesscases, like one for civil engineering dam management for thePortuguese National Laboratory for Civil Engineering [52].

Each of the retrieved concepts shall then need to be validated bythe whole system, and this action frequently generates numerousconflicts, with multiple different semantic interpretations of thesame concept, or different procedures and methods to define it.Negotiation in this scenario is then an essential tool to enable thecentral knowledge ontology to be developed.

The application of the proposed NEGOSEIO interoperabilityframework to TIMBUS used the negotiation strategies specified inthe above proposed interoperability framework to allow eachontology element to be classified and related with the otherontology terms, thus contributing to the proper development andharmonisation of the project’s reference ontology. The develop-ment of TIMBUS is still ongoing as the project is still developing itstools and ontologies (DIO and DSOs), thus the application ofNEGOSEIO has only been possible in small-scaled prototypes and isstill being validated in this real business case.

6. Conclusions and final considerations

Negotiation towards the achievement of interoperabilitysolutions is a subject that although is being addressed tacitly, is

borative negotiation for ontology-driven enterprise businesses,01.001

Page 9: Collaborative negotiation for ontology-driven enterprise businesses

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx 9

G Model

COMIND-2541; No. of Pages 10

regularly performed in an empiric and informal manner. As thesubjects at stake (in this case, the negotiation and harmonisation ofontological concepts) are growing in number and complexity, theamount of time spent doing this negotiation tends to grow rapidly.Maintaining a relaxed negotiation methodology will thus lead totwo possible outcomes:

� either the negotiation will be simple and superficial, conductingto poor and immature solutions which in time will result in morefrequent interoperability problems;� or the negotiation will be thorough and complex, and as there is

no formal approach this will result in frequent rework of thenegotiation subjects without reuse and without learning frompast decisions, which will take longer negotiation time.

This paper proposes an intelligent negotiation framework tomanage parallel and concurrent negotiations in order to achieve andmaintain the interoperability between the organisations’ systemsand applications and its validation in an industrial scenariodeveloped under the Enterprise Information Systems concept.

As a result of the ongoing experimentation, the followingconclusions can already be extracted:

� TIMBUS embeds a mechanism that detects changes in theextracted metadata, thus being able to trigger actions upon thesechanges. By triggering NEGOSEIO to analyse these changes andtry to reason ontology changes, it was found that the negotiationmechanism really improves the decision analysis:� not only because it clearly enumerates the decisions that need

to be taken,� but also it provides support coming from previous decisions’

knowledge.� The use of NEGOSEIO helped reducing the decision time and

providing suggestions for best solutions (based on past historicalrules), thus confirming empirically the hypotheses that negotia-tion is an appropriate method to support decisions oninteroperability solutions, and also that by improving thedefinition of the ontologies with a strong, formal method whichstores the reasoning behind the decision-making, the under-standing of the ontological terms is better supported, thusenforcing this understanding by all involved parties;� As a conclusion, it can be stated that this particular case of having

a better understanding of the terminology and semanticconcepts of a particular business will contribute to ensure thatthe business processes and flows are understood by all theinteroperating enterprises, therefore promoting a strong andsustained Enterprise Interoperability.

Acknowledgments

The authors wish to acknowledge the support of the EuropeanCommission through the funding of the UNITE, MSEE, IMAGINE,and TIMBUS FP7 projects.

References

[1] H. Panetto, J. Cecil, Information systems for enterprise integration, interoperabili-ty and networking: theory and applications, Enterprise Information Systems 7 (1)(2013) 1–6.

[2] M.-S. Li, R. Cabral, G. Doumeingts, K. Popplewell, Enterprise InteroperabilityResearch Roadmap, European Commission – CORDIS, 2006 http://cordis.euro-pa.eu/ist/ict-ent-net/ei-roadmap_en.htm, http://cordis.europa.eu/fp7/ict/enet/ei-rrrev_en.html.

[3] M. Finnegan, Boeing 787s to create half a terabyte of data per flight, says VirginAtlantic, Computer World UK (2013).

[4] J. Gantz, D. Reinsel, The Digital Universe in 2020: Big Data, Bigger Digital Shadows,and Biggest Growth in the Far East Executive Summary: A Universe of Opportu-nities and Challenges, 2012.

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

[5] R. Jardim-Goncalves, C. Agostinho, A. Steiger-Garcao, Sustainable Systems’ Inter-operability: a reference model for seamless networked business, in: 2010 IEEEInternational Conference on Systems, Man, and Cybernetics (SMC 2010), 2010,1785–1792.

[6] J. Andreoli, P. Ciancarini, R. Pareschi, Interaction abstract machines, Communica-tion 280 (1993) 1–18.

[7] S. Dustdar, W. Schreiner, A survey on web services composition, InternationalJournal of Web and Grid Services 1 (1) (2004) 1–30.

[8] M.P. Papazoglou, P. Traverso, S. Dustdar, F. Leymann, Service-oriented computing:a research roadmap, International Journal of Cooperative Information Systems 17(02) (2008) 223.

[9] C. Coutinho, A. Cretan, R. Jardim-Goncalves, Sustainable interoperability frame-work for supporting negotiation processes, in: Proceedings of the 6th Interna-tional Conference on Interoperability for Enterprise Systems and Applications (I-ESA 2012), 2012, pp. 365–372.

[10] K. Jeffery, B. Neidecker-Lutz, The future of cloud computing: opportunities forEuropean cloud computing beyond 2010, Analysis (2010).

[11] A. Cretan, C. Coutinho, B. Bratu, R. Jardim-Goncalves, A framework for sustainableinteroperability of negotiation processes, in: Proceedings of the 14th IFAC Sym-posium on Information Control Problems in Manufacturing (INCOM’12), 2012, pp.1258–1263.

[12] F. Benaben, N. Boissel-Dallier, H. Pingaud, J.-P. Lorre, Semantic issues in model-driven management of information system interoperability, International Journalof Computer Integrated Manufacturing (May) (2012) 1042–1053, http://www.tandfonline.com/doi/abs/10.1080/0951192X.2012.684712#.UvJyxfl_s75.

[13] F. Benaben, H. Pingaud, The MISE project: a first experience in mediationinformation system engineering, Information Systems: People, Organizations,Institutions, and Technologies (2010) 399–406, http://dx.doi.org/10.1007/978-3-7908-2148-2_46, http://www.springer.com/business+%26+management/business+information+systems/book/978-3-7908-2147-5.

[14] C. Coutinho, A. Cretan, R. Jardim-Goncalves, Cloud-based negotiation for sustain-able Enterprise Interoperability, in: Proceedings of the 18th International Con-ference on Engineering, Technology and Innovation (ICE2012), 2012, pp. 1–10.

[15] C. Agostinho, F. Correia, R. Jardim-Goncalves, Interoperability of complex businessnetworks by language independent information models, in: New World Situa-tion: New Directions in Concurrent Engineering – Proceedings of the 17th ISPEInternational Conference on Concurrent Engineering (CE 2010), 2011, pp. 111–124 (Part 2).

[16] H. Panetto, Towards a classification framework for interoperability of enterpriseapplications Herve Panetto, International Journal of Computer IntegratedManufacturing 20 (8) (2007) 727–740.

[17] R. Jardim-Goncalves, A. Grilo, C. Agostinho, F. Lampathaki, Y. Charalabidis, Sys-tematisation of interoperability body of knowledge: the foundation for EnterpriseInteroperability as a science, Enterprise Information Systems 6 (3) (2012) 1–26.

[18] S. Izza, Integration of industrial information systems: from syntactic to semanticintegration approaches, Enterprise Information Systems 3 (1) (2009) 1–57.

[19] M. Dutra, P. Ghodous, O. Kuhn, A generic and synchronous ontology-basedarchitecture for collaborative design, Concurrent Engineering 18 (February (1))(2010) 65–74.

[20] M. Zdravkovic, H. Panetto, M. Trajanovic, A. Aubry, An approach for formalizingthe supply chain operations, Enterprise Information Systems 5 (4) (2011) 401–421.

[21] A.M. Khattak, Z. Pervez, K. Latif, A.M.J. Sarkar, S. Lee, Y.-K. Lee, Reconciliation ofontology mappings to support robust service interoperability, in: 2011 IEEEInternational Conference on Services Computing, 2011, 298–305.

[22] M. Laclavık, Z. Balogh, M. Babık, L. Hluchy, AgentOWL: semantic knowledgemodel and agent architecture, Computing and Informatics 25 (2006) 421–439.

[23] J.M. Torres, Q. Wijnands, Distributed agents for multi-rover autonomy, in: 20thInternational Workshops on Enabling Technologies: Infrastructure for Collabora-tive Enterprises, June, (2011), pp. 53–58.

[24] J. Sarraipa, R. Jardim-Goncalves, A. Steiger-Garcao, MENTOR: an enabler forinteroperable intelligent systems, International Journal of General Systems 39(July (5)) (2010) 557–573.

[25] J. Sarraipa, R. Jardim-Goncalves, T. Gaspar, A. Steiger-Garcao, Collaborative on-tology building using qualitative information collection methods, in: 5th IEEEInternational Conference on Intelligent Systems (IS 2010), 2010, 61–66.

[26] P.A. Haya, G. Montoro, A. Esquivel, M. Garcıa-Herranz, X. Alaman, A mechanismfor solving conflicts in ambient intelligent environments, Journal of UniversalComputer Science 12 (3) (2006) 284–296.

[27] C. Shin, H. Yoon, W. Woo, User-centric conflict management for media servicesusing personal companions, ETRI Journal 29 (3) (2007) 311–321.

[28] C. Shin, A.K. Dey, W. Woo, Mixed-initiative conflict resolution for context-awareapplications, in: Proceedings of the 10th international conference on Ubiquitouscomputing (UbiComp’08), 2008, pp. 262–271.

[29] C. Shin, A.K. Dey, W. Woo, Toward combining automatic resolution with socialmediation for resolving multiuser conflicts, Cybernetics and Systems 41 (2) (Apr.2010) 146–166.

[30] A.I. Oliveira, L.M. Camarinha-Matos, Electronic negotiation support environmentin collaborative networks, Advances in Information and Communication Tech-nology (2012) 21–32.

[31] H. Dong, F.K. Hussain, E. Chang, State of the art in negotiation ontologies for multi-agent systems, International Journal of Web Services Practices 4 (2008) 157–163.

[32] L. Mazuel, N. Sabouret, A communication protocol for semantic heterogeneitywith incomplete ontology alignment, in: Proceedings of the 8th InternationalConference on Autonomous Agents and Multiagent Systems (AAMAS’09), 2009,pp. 1187–1188.

borative negotiation for ontology-driven enterprise businesses,01.001

Page 10: Collaborative negotiation for ontology-driven enterprise businesses

R. Jardim-Goncalves et al. / Computers in Industry xxx (2014) xxx–xxx10

G Model

COMIND-2541; No. of Pages 10

[33] P.B. Jeon, J. Kim, S. Lee, C. Lee, D.-K. Baik, Semantic negotiation-based serviceframework in an M2M environment, in: 2011 IEEE/WIC/ACM International Con-ferences on Web Intelligence and Intelligent Agent Technology, 2011, 337–340.

[34] T.-Y. Chen, Y.-M. Chen, H.-Y. Cai, Design of knowledge trading negotiation modelfor accelerating acquirement of knowledge, International Journal of ComputerIntegrated Manufacturing 25 (December (12)) (2012) 1085–1101.

[35] X.H. Wang, T.N. Wong, G. Wang, Knowledge representation for multi-agentnegotiations in virtual enterprises, International Journal of Production Research49 (July (14)) (2011) 4275–4297.

[36] C. Coutinho, A. Cretan, R. Jardim-Goncalves, Sustainable interoperability on spacemission feasibility studies, Computers in Industry Journal – Special Issue onInteroperable Enterprises 64 (8) (2013) 925–937, http://dx.doi.org/10.1016/j.compind.2013.06.016.

[37] R. Jardim-Goncalves, C. Agostinho, A. Steiger-Garcao, A reference model forsustainable interoperability in networked enterprises: towards the foundationof EI science base, International Journal of Computer Integrated Manufacturing 25(10) (2012) 855–873.

[38] A. Cretan, C.F. da Silva, C. Coutinho, R. Jardim-Goncalves, P. Ghodous, Frameworkfor ontology-based negotiation to support enterprise interoperability in cloud-based environments, International Journal of Electronic Business Management 11(3) (2013) 168–177.

[39] T. Santos, C. Coutinho, A. Cretan, R. Jardim-Goncalves, Agents and rules for thenegotiation of interoperability solutions, in: Proceedings of the IEEE InternationalTechnology Management Conference & 19th ICE2013, 2012.

[40] JADE Agent development framework, 2012.[41] Drools Rule Engine, 2012 Available: http://www.jboss.org/drools (accessed

14.08.12).[42] J. Sarraipa, J.P.M.A. Silva, R. Jardim-Goncalves, A.A.C. Monteiro, MENTOR – a

methodology for enterprise reference ontology development, in: 4th Internation-al IEEE Conference on Intelligent Systems IS’08, vol. 1, 2008, 32–40.

[43] X. Luo, C. Miao, N.R. Jennings, M. He, Z. Shen, M. Zhang, KEMNAD: a knowledgeengineering methodology for negotiating agent development, ComputationalIntelligence 28 (1) (2012) 51–105.

[44] N. Georgantas, M.A. Ahaman, H. Ameziani, A. Pathak, V. Issarny, A coordinationmiddleware for orchestrating heterogeneous distributed systems, in: LectureNotes in Computer Science Volume 6646: Advances in Grid and PervasiveComputing, 2011, 221–232.

[45] R. Jardim-Goncalves, N. Figay, A. Steiger-Garcao, Enabling interoperability of STEPapplication protocols at metadata and knowledge level, International Journal ofTechnology Management 36 (4) (2006) 402–421.

[46] R. Jardim-Goncalves, C. Agostinho, P. Malo, A. Steiger-Garcao, Harmonisingtechnologies in conceptual models representation, International Journal of Prod-uct Lifecycle Management 2 (2) (2007) 187–205.

[47] TIMBUS, TIMBUS Project page, 2013 Available: http://timbusproject.net(accessed 15.05.13).

[48] M.A. Neumann, H. Miri, J. Thomson, G. Antunes, R. Mayer, M. Beigl, Towards adecision support architecture for digital preservation of business processes, in:Proceedings of the International Conference on Preservation of Digital Objects(iPRES 2012), 2012.

[49] C. Coutinho, Digital preservation of business assets as a risk management strate-gy, in: Proceedings of the 2013 Workshop Open Source and Design Of Communi-cation (OSDOC2013), 2013, http://dx.doi.org/10.1145/2503848.2503858pp. 46-48.

[50] G. Antunes, A. Caetano, M. Bakhshandeh, R. Mayer, J. Borbinha, Using ontologies tointegrate multiple enterprise architecture domains, in: 4th Workshop on Businessand IT Alignment (BITA 2013), 2013.

[51] The Open Group, The ArchiMate Modelling Language, 2012, Available: http://www.opengroup.org/subjectareas/enterprise/archimate (accessed 14.12.12).

[52] LNEC. Available: http://www.lnec.pt/organizacao/dbb/nmmf/estudos_id(accessed 10.06.13).

Ricardo Jardim-Goncalves holds a PhD degree andreceived his habilitation (‘‘Agregacao’’) in IndustrialInformation Systems by the New University of Lisbon(UNL). He is an Associate Professor at the NewUniversity of Lisbon, Faculty of Sciences and Technolo-gy, and a Senior Researcher at UNINOVA (Instituto deDesenvolvimento de Novas Tecnologias). He hasgraduated in Computer Science, with MSc in Opera-tional Research and Systems Engineering. His researchactivities have been focused on Interoperability ofComplex Systems. He has been researching in EuropeanCommission funded projects during the last 20 years,with more than 100 papers published in conferences,

Please cite this article in press as: R. Jardim-Goncalves, et al., CollaComput. Industry (2014), http://dx.doi.org/10.1016/j.compind.2014.

journals and books. He directs GRIS (Group for Researchin Interoperability of Systems) at UNINOVA. He isproject leader in ISO TC184/SC4.

Carlos Coutinho is a Senior Research Engineer and R&DProject Manager at Caixa Magica Software in Lisbon,Portugal. He holds a PhD degree in Electrical andComputer Engineering, awarded by the New Universityof Lisbon (UNL), Portugal, where he also does research,with interests in Enterprise Interoperability, AdaptablePlatforms and Systems, SOA, and Model-Driven Com-puting. He also holds a PMI-PMP title and a post-graduation in Project Management by the universityInstituto Superior Bissaya-Barreto (ISBB) in Coimbra,Portugal. He has more than 15 years of experience in theenterprise IT area, working in several fields from ICT,Service development, Public Administration and theAerospace industry.

Adina Cretan is an academic with wide interest in theresearch area covering Multi-Agent Systems, Distribut-ed Systems, Collaborative modelling, Ontologies, Inter-operability. She holds a PhD in Economic Informatics,awarded by the Academy of Economic Studies inBucharest, Romania, and completed Post-Doctoralresearch in Service Oriented Computing field at LIRISResearch Center University Lyon I, France. She alsoholds an MBA from the joint program of the Academy ofEconomic Studies in Bucharest, Romania and CNAMParis, France. At present, Adina teaches at ‘‘NicolaeTitulescu’’ University of Bucharest, as Associate Profes-sor in the Computer Science Department.

Catarina Ferreira da Silva is currently AssociateProfessor in computer science department of theInstitute of Technology of the University of Lyon I.She is also member of LIRIS UMR 5205 CNRS(Laboratory of Computer Graphics, Images and Infor-mation Systems). Between 2009 and 2012 she wasresearcher at the Centre for Informatics and Systems ofthe University of Coimbra (CISUC), Portugal, within theInformation Systems Group. From 2006 to 2008, shewas teacher at the computer science department of theUniversity of Lyon I and researcher at the LIRIS. Sheobtained a Ph.D. in computer science (2007) from theUniversity of Lyon I. Her main research interests areSemantic Interoperability of Information Systems,Ontologies, Services Science and Cloud Computing.

Parisa Ghodous is currently full Professor in computerscience department of University of Lyon I. She ismember of LIRIS UMR 5205 (Laboratory of ComputerGraphics, Images and Information Systems). Herresearch expertise is in the following areas: Interoper-ability, Semantic Web, Web services, Collaborativemodeling, Product data exchange and modeling andCloud Computing. She is in editorial boards of CERA,ICAE and IJAM journals and in the committees of manyrelevant international associations such as concurrentengineering, ISPE and Interoperability.

borative negotiation for ontology-driven enterprise businesses,01.001