Modelling Services for Interoperability Negotiation

Carlos Coutinho

, Adina Cretan

and Ricardo Jardim-Goncalves

Caixa Mágica Software, Rua Soeiro Pereira Gomes, Lote 1-4 B, 1600-196, Lisboa, Portugal

“Nicolae Titulescu” University, 185 Calea Văcăreşti, District 4, Bucharest, 040051, Romania

CTS, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, UNINOVA, Lisboa, Portugal

Keywords: Negotiation, Model-driven Development, Servitisation.

Abstract: The evolution of businesses, driven by the advent of the internet and subsequent globalisation, was very

quick and sudden. Enterprises that had traditional development strategies and that were used to slow

changes are suddenly feeling the urge to evolve or face obsolescence. This fast-paced evolution frequently

leads to mistakes and erroneous decisions, many of which are solely detected after a long period. When this

happens, frequently the only solution is to return to a previous stable business stage before proceeding. The

EU co-funded FP7 TIMBUS project comprises tools and techniques to improve business continuity

featuring an intelligent strategy for digital preservation of business assets and environments based on risk-

management. This paper proposes the modelling of service-based negotiation strategies to help in two

phases: First in the definition of the enterprise business interoperability strategy, and second, in the

reasoning of assets and concepts needed for the capture of digital preservation of business assets.

1 INTRODUCTION

Technology nowadays seems to surpass all concepts

related with information exchange. The market

demand changed a lot from the first information

systems, when interoperation was determined just by

being able to consistently exchange data. Since then,

Information and Communication Technologies

(ICT) have evolved from little more than wired data

exchange to being capable of providing complex

high-performing systems that already handle data

reliability over a large bandwidth.

Thus, market needs are moving towards the next

step and shifting to Enterprise Interoperability.

Major breakthroughs have been achieved in the near

past on this subject: data networking, the Internet

and web-services have contributed to the collapse of

the concept of enterprise isolation and paved the

way for businesses to communicate and interoperate.

This quickly led to the growth of service-

provider-consumer businesses and directed

companies to specialise and improve their services,

moved by the rising competition. The concept of

service has been generalised, as all interactions (e.g.,

complex computer calculations, graphics design, but

also carpet cleaning, wall painting, hardware

assembly) were able to be virtualised as computer-

based services that companies can evoke to perform

their business. The globalised world has entered the

Service era, where an enterprise produces and

consumes services and goods using the internet,

whether these are simple services that provide

specific calculations to complex platforms like

Customer Relationship Management (CRM) systems

with distributed databases, to producing cars,

houses, tools and other products. Virtualisation and

service provisioning are common means of

interoperate; however, this interoperability still

relies on sets of agreed message syntax exchanges

and agreements that are fragile and limited in their

scope.

Emerging paradigms like the Internet of Things

(IoT) and the Internet of Services (IoS) (“Internet of

Services,” 2012), together with the evolving cloud

computing concepts are gradually transforming the

existing reality into a set of available commoditised

virtual objects, services, enterprises and networks.

The globalised market means companies face the

availability of multiple possibilities in terms of

suppliers and service providers, partners and

customers, but also the growth of competition.

Competitiveness is also pressuring enterprises to

build better solutions with fewer resources,

following new trends and supporting new platforms

682

Coutinho C., Cretan A. and Jardim-Gonçalves R..

Modelling Services for Interoperability Negotiation.

DOI: 10.5220/0004883206820688

In Proceedings of the 2nd International Conference on Model-Driven Engineering and Software Development (MDSE-2014), pages 682-688

ISBN: 978-989-758-007-9

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

and methodologies. Business models are recurrently

more complex and detailed. Quality improvements

and standards evolve more frequently, as well as

concerns regarding accessibility, expansion,

continuous improvement and upgrade, compliance

to new technologies and support of new platforms,

frameworks and development paradigms. Local and

global legislation and regulations are updating more

frequently and deeply, and demanding rapid

compliance from enterprises.

The increasing complexity of processes and

interactions between businesses, companies,

providers and customers means interoperability can

no further be seen as a tacit and empiric science, but

needs to be well established and founded, with

detailed definitions, flows, strategies, procedures

and interactions. Plus, the quality standards for

service provisioning have risen, which means the

price of losing interoperability in the current scope

may mean the loss of business and credibility in an

irreparable way. Strategies and decisions are

increasingly more dependent on interoperability and

so are also the investments and the whole business

kernel.

These frequent business changes shake all the

interoperability links between the enterprises,

leading to periods of adaptation where business

operation is not possible. The urge to rapidly regain

interoperability often leads to unfounded, poorly-

chosen solutions, which lead to inefficiency and

rework.

Often these problems are only detected very late

in the interactions of the business enterprises, due to

many factors, either because there are still attempts

to correct bad decisions or because of the inertia

inherent to most businesses to make profound

changes. The fact is that in many cases, when the

decision comes to overcome such situations, the

only solution is to go back to an earlier stage of the

business evolution and start all over again. The

problem is that this is frequently not possible

anymore, due to the environment changes that in the

meantime have occurred.

The EU co-funded FP7 project TIMBUS

(TIMBUS, 2013) faces these problems and proposes

solutions that include a reasoned Digital

Preservation of business assets, where this reasoning

is performed by risk management. The main

innovation of TIMBUS project is therefore its focus

on risk assessment based digital preservation of

business processes, thus not only bringing together

but also advancing traditional digital preservation,

risk management and business process management

disciplines. Preservation is often considered as a set

of activities carried out in isolation within a single

domain, without of taking into account the

dependencies of third-party services, information

and capabilities that will be necessary to validate

digital information in the future. Existing DP

solutions focus on more simple data objects which

are static in nature. The unique aspect of TIMBUS is

that it is attempting to advance state of the art by

figuring out how more complex digital objects can

be preserved and later restored in the same or

different environments.

This paper proposes a solution for two distinct

problems described previously: by establishing a

formal model, framework and strategy for

negotiation, the authors propose a methodology to

improve the definition of new solutions for

interoperability between enterprises, and also to

improve the reasoning behind the risk-management

analysis to select business assets for digital

preservation. These methods and framework are

being evaluated in the scope of the TIMBUS project.

Section 2 presents the background analysis on

literature over the proposed solution. Section 3

presents the proposed solutions and how they are

being applied to the determined scope. Section 4

presents the validation in the proposed use-case, and

Section 5 presents the conclusions and future work.

2 LITERATURE REVIEW

The proposed methodology is based on the kernel

aspect of negotiations, proposing formal models and

strategies, supported by a framework which includes

several concepts inspired by the work of project

Manufacturing SErvice Ecosystem (MSEE), a

consortium project of the ICT Work Programme, of

the European Community's 7th Framework

Programme (FP7) (“MSEE Project,” 2012),

including Model-Driven engineering, SOA, but

extending it to Cloud-based solutions.

2.1 SSME and MSDE

The term Services Sciences, Management and

Engineering (SSME) was coined by IBM (Maglio et

al., 2006) to deal with an holistic approach stating

that businesses can be the result of a set of services –

the conjunction of people, technology, and

organisations to create value, towards becoming

very adaptive and flexible, reusable and

commoditised. The SSME aims to improve the

sustainability of the development processes,

monitoring and controlling assets e.g., the quality,

ModellingServicesforInteroperabilityNegotiation

683

productivity and innovation of services and the

exchange and widespread of services.

SSME vision states that to define a business,

more than dealing only with its tangible assets

(hardware, software, and related documentation) –

hence Technology, businesses should also be

analysed according to their processes, environment,

procedures, quality standards, towards achieving the

business optimisation that is needed for being

competitive.

SSME also notes that an important asset of

businesses is the human factor, i.e. the capabilities of

its human resources and their interactions determine

the agility and flexibility of a business. Issues like

motivation, skills, team building and development,

leadership, personal involvement and achievements

are leading the priorities of enterprises.

All these aspects must be developed in the scope

of a business vision and strategy, which itself can be

analysed, studied and optimised by statistical

methods and Ishikawa (cause-and-effect) diagrams

and analysis towards the creation of servitised

strategies that can be reused as business

development frameworks.

The MSEE project targets to pave the way for

service development in Europe, with the creation of

virtual manufacturing factories (Factories of the

Future), which shall make use of extended

servitisation for the shift from product-centrism to

product-based services, distributed in virtual

organisations and ecosystems.

This project proposed a Model-Driven Service

Engineering (MDSE) architecture, largely inspired

in the concepts of SSME, which accounts enterprise

services to be modelled into three major aspects

(views): IT, Machine (and operation) and Human

Resources. The MDSE models are developed using

various specifications, e.g., the EN/ISO 19440

standard, the GRAI modelling language

(Doumeingts et al., 2006), the POP* language

(Athena Consortium, 2011) and the Unified Service

Description Language (USDL).

2.2 Model-Driven Architectures

The term Model-Driven Architectures (MDA) was

coined by the Object Management Group (OMG),

and promotes the evolution of solutions through

successive transformations of higher-level models

into lower-level models, which eventually may

result in going down to the level of code generation

(OMG, 2011). This represented a change of the

undergoing paradigm that professed that system

architectures are built by designing and maintaining

its code. In this case, the changes are performed in

the models, which are then transformed into code.

This means that interoperability may start from

the very enterprise foundations, where it is easier to

discuss business-related concepts and ideas, and then

the progressive steps of transformation into lower-

level models may also be synchronised to refine this

interoperability, so that the overhead of transforming

the concepts into code is performed by automation

tools.

The development paradigm of MDA allows the

definition of multiple levels of abstraction in the

modelling of businesses, using descriptive languages

and schemes e.g., Unified Modelling Language

(UML), Object Constraint Language (OCL), and

Unified Enterprise Modelling Language (UEML) to

define the solution foundations. Applications should

be designed right from a high-level abstract

Computation Independent Model (CIM) where all

business related functionalities, objectives, methods,

context, requirements and definitions are specified

regardless of any implementation (i.e., pure design).

Then, this model shall be subject to

transformations into a more detailed Platform

Independent Model (PIM), where the business

concepts and rules are converted into activities,

tasks, ontologies, structures and algorithms,

although still independently of the underlying

platform.

Finally, other vertical transformations and

conversions shall turn the PIM into a Platform

Specific Model (PSM), which provides the

foundations for the development of the application,

now targeted to a specific platform. Using the

proposed framework, changes to any model (CIM,

PIM) may trigger alterations in the other parties’

models, which then, by transformation towards new

PSMs, swiftly change the application towards

compliance with the new model.

2.3 Model-Driven Interoperability

The Model-Driven Interoperability (MDI) concept

derives from MDA: it comprises the same

abstraction layers, but in this case the target to be

modelled is the interoperation between the involved

parties. The idea behind MDI is to define models for

each MDA level that allow the exchange of

information. If the MDA can be described as a set of

vertical transformations from a conceptual high-

level model to a progressively detailed model, then

MDI may be seen as a set of horizontal

transformations to allow interoperability at each

MDA level, e.g., Process, Product and

MODELSWARD2014-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment

684

Organisational models with the System

Requirements at CIM level and transformations of

these models into interoperability models.

Projects like the Advanced Technologies for

Interoperability of Heterogeneous Enterprise

Networks and their Application (ATHENA) defined

a framework that supports interoperability

throughout the various abstraction levels and

business aspects of enterprise software engineering

(Athena Consortium, 2007). (Lemrabet et al., 2010)

provide simplified views over the MDI concept and

the ATHENA Interoperability Framework (AIF)

concepts and solutions on actions to develop each

level of interoperability:

 Interviews, workshops and BPMN

choreography diagrams for CIM levels;

 Diagrams, definition of business goals and

BPMN collaboration diagrams for PIM

levels;

 Service Oriented Architectures (SOA) and

BPEL implementations at PSM levels.

(Chen et al., 2008) define a roadmap on the

possible approaches towards the development of

enterprise architectures accounting interoperability.

3 FRAMEWORK FOR

NEGOTIATION

The proposed framework, presented on (Coutinho et

al., 2013), defines three main pillars for its

feasibility: a development methodology, a

negotiation model and an operative framework.

3.1 Methodology

The proposed methodology consists in the steps

described on Figure 1:

Figure 1: Development Methodology.

According to the proposed methodology, the

steps for business enterprise interoperability

development are:

 Knowledge: Acquisition of the information

about the target business and interoperability

requirements;

 Modelling: Through the development of

models using MDA and MDI, the enterprise

interoperability becomes predictable,

transparent and flexible to changes;

 Servitisation: The resulting artefacts from

the models are shaped in the form of

services, whether to define the foundations

of the business itself or to shape the

interoperability between enterprises;

 Flexibility: The deployment of platforms,

infrastructure and services must be highly

flexible to cope with the constant changes

required by the business environment.

The central point of the proposed methodology

however is negotiation. Negotiation will be used

transversely in all the described steps, to reason the

knowledge acquisition into understanding what to

capture and to which extention of detail or

granularity, to be the central element on the MDI

models to perform the enterprise interoperability in

all MDA abstraction levels (CIM, PIM, PSM), and

to determine which are the best solutions for

servitisation and flexibility/deployment, according to

the past experiences.

3.2 Negotiation Model

The proposed negotiation model consists on a set of

entities that will store the granular steps of the

negotiation called negotiation atoms. The

negotiation atoms shall rely on quintuples with the

following information:

M = < T, P, N, R, O >

where:

 T denotes the timestamp of the system,

assumed discrete, linear, and uniform;

 P denotes the set of participants in the

negotiation framework. The participants

may be involved in one or many

negotiations;

 N denotes the set of negotiations that take

place within the negotiation framework;

 R denotes the set of coordination rules

among negotiations that take place within

the negotiation framework;

 O denotes the common ontology for the set

of definitions of the attributes that are used

in a negotiation.

Hence, this is the basic “state” element that

stores the environment state of each negotiation step.

ModellingServicesforInteroperabilityNegotiation

685

Negotiations are therefore cycles of actions which

include several strategies to perform the activities,

each consisting on numerous of these quintuples.

Some of the defined strategies for negotiation are:

 Subcontracting (resp. Contracting) for

subcontracting jobs by exchanging proposals

among participants known from the

beginning;

 Block component for assuring that a task is

entirely subcontracted by the single partner;

 Divide component manages the propagation

of constraints among several slots,

negotiated in parallel and issued from the

split of a single job;

 Broker to automate the process of selection

of possible partners to start the negotiation.

For each strategy, multiple steps and quintuples

are defined and stored in the negotiation

infrastructure, in the shape of business rules that

need to be processed by a rule engine (in the

proposed case, the rule engine that is being used is

Drools).

The use of this formal negotiation model allows

the storage of the historic negotiation data, essential

to infer later and assist in the decisions to be taken in

the future.

3.3 Negotiation Framework

The negotiation framework proposed in (Cretan et

al., 2012) is based on three negotiation levels, as

shown on

Figure 2.

Figure 2: Negotiation Levels.

Firstly, a lower Middleware level implements

communication services and provides support for the

aspects related with the basic infrastructures,

handling the heterogeneity related with multiple

negotiation players, using an Enterprise Service Bus

(ESB) on interactions for dealing with technical

interoperability issues.

On top of it, another negotiation level called

Coordination Services (CS), has the purpose of

assisting the negotiations at a global level (with

different participants on different jobs) and at a

specific level (on the same job with different

participants). It handles the issues regarding

communication at this level (synchronisation

between the CSs of the parties that are taking place

in the negotiation), and manages the on-going

transactions and negotiation data persistence,

controlling the semantic discrepancies between the

negotiating parties.

Finally, at a client-side scope, Finally, the top

negotiation level performs the task of Negotiation

Manager, implementing the business decisions that

need to be taken for the negotiation, e.g., starting a

new negotiation, inviting other parties to join the

negotiation, making proposals to the negotiation,

and accepting or rejecting proposals.

The remaining framework, besides the above

modules that implement the negotiation levels

consists in modules for managing the rule-based

negotiation (rule engine), a reference ontology to

harmonise the terms and concepts existing

throughout the various interoperability steps, and

other support modules, e.g., cloud management

modules, communications, storage using standard

ISO 10303 STEP, and other utility services.

4 VALIDATION ON A REAL

BUSINESS CASE – TIMBUS

This project is being implemented in the scope of the

TIMBUS project. This project has now finished the

definition of basic tools and is starting to implement

its applicability on a set of use-cases.

As stated on section 1, TIMBUS aims to provide

mechanisms to perform business continuity through

risk analysis and digital preservation of business

environment assets. The part of the project which

serves as use-case for this paper is the one that

relates to the capturing of business interoperation

with external parties. Namingly, the information that

is captured from each business can have multiple

interpretations and meanings. Each asset’s

characteristics is much more than only a perspective

of the target asset; it is related to a set of semantic

interpretations which are distinct, depending on who

MODELSWARD2014-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment

686

performs them. The same objective measurement of

a sensor or the output result of a particular

experiment can have completely semantic meanings,

depending on the sensitivity, scope or even business

strategy or objectives of each analyser.

This is a problem which commonly affects the

proper interoperability between systems. The

negotiation model and framework presented in this

paper have the objective to make these objective

characteristics more transparent, by modelling them

via MDA and MDI, allowing the various

stakeholders to be able to understand not only what

is visible but also the motivation, strategy and

semantics behind each measurement and value.

Additionally, the negotiation mechanisms promote

that these semantics and values do not remain static,

but instead are able to evolve and improve through

time, allowing transformations, enhancements and

the introduction of new ideas, trends, platforms and

semantics.

One of these use-cases regards the analysis for

business continuity and risk management towards

digital preservation of the network of dams in

Portugal, performed by the National Civil

Engineering Laboratory (“LNEC,” 2013).

A practical example of this framework relates to

the measurements taken on a given set of dam

sensors, related to temperature and pressure. The

interpretation of these measurements varies,

depending on the purpose of the selected set of

sensors and their location. It also varies according to

the objectives and strategies of the teams which are

analysing these results.

The proposed negotiation framework and models

permit the establishment of a common view, an

agreed and compromising interpretation and

semantic view of the results, and the negotiation

model permits the analysis of the outcome of this

negotiation. Whether this outcome is good or bad

will influence the decisions to be taken on future,

similar negotiations, hence promoting best-practices

and reuse of knowledge.

The applicability of this paper may then be

validated using a set of indicators and validation

rules, which include the amount of different

terminologies and processes that need to be

harmonised throughout the different dams or the

different sensor suppliers, the amount, effort and

cost of the rework happening due to semantic

misalignment before and after the application of the

framework, the amount of time spent on

harmonising these semantic issues with and without

formal negotiation, the advantages in amount of time

and cost of having a rich historic record of previous

negotiations and negotiation steps and resulting

outcomes.

5 CONCLUSIONS

AND FUTURE WORK

Business complexity is rapidly increasing due to

globalisation and, well, evolution. In this fast-pace,

there are options and business decisions that need to

be taken rapidly as well. The lack of maturity of

numerous enterprises leads them to early and poorly

designed solutions for enterprise interoperability,

leading to some obvious mistakes that can be

corrected immediately, and others that are not so

obvious or detectable. When these are finally

detected, some may require a reinstate of some of

the business premises and environment.

While the TIMBUS project is aiming to support

the development of this by performing risk

management and selective digital preservation of

assets, it is also based on the traditional risk

management empirical analysis. This paper

proposed a methodology, a negotiation model and a

framework to support a mature, decision-support

analysis of the business continuity, based on the

modelling of the various entities and aspects related

to enterprise interoperability, supported by a

servitised set of supporting activities which are

defined to perform the interoperability and to

support it. The proposed framework’s future work is

to be validated in the scope of the project TIMBUS’s

use-cases. Additionally, the authors are seeking to

enrich this framework with mechanisms which

enforce contracting and the improvement of formal

negotiation strategies.

ACKNOWLEDGEMENTS

The authors wish to acknowledge the support of the

European Commission through the funding of the

MSEE, UNITE, and ENSEMBLE FP7 projects.

This work is being funded by the TIMBUS project,

co-funded by the European Union under the 7th

Framework Programme for research and

technological development and demonstration

activities (FP7/2007-2013) under grant agreement

nr. 269940.

ModellingServicesforInteroperabilityNegotiation

687

REFERENCES

Athena Consortium, 2007. ATHENA Project:

Specification of interoperability framework and

profiles, guidelines and best practices.

Athena Consortium, 2011. Athena Interoperability

Framework [WWW Document]. URL

http://www.modelbased.net/aif (accessed 12.20.11).

Chen, D., Doumeingts, G., Vernadat, F., 2008.

Architectures for enterprise integration and

interoperability: Past, present and future, in:

Computers in Industry, vol. 59, issue 7, pp. 647–659.

Coutinho, C., Cretan, A., Jardim-Goncalves, R., 2013.

Sustainable Interoperability on Space Mission

Feasibility Studies, in: Computers in Industry - Special

Issue on Interoperable Enterprises, vol. 64, issue 8,

pp. 925–937.

Cretan, A., Coutinho, C., Bratu, B., Jardim-Goncalves, R.,

2012. NEGOSEIO: A Framework for Negotiations

toward Sustainable Enterprise Interoperability, in:

IFAC Journal Annual Reviews In Control, vol. 36,

issue 2, pp. 291–299.

Doumeingts, G., Vallespir, B., Chen, D., 2006. GRAI

GridDecisional Modelling, in: Handbook on

Architectures of Information Systems. Springer, pp.

321–346.

Internet of Services [WWW Document], 2012. URL

http://www.internet-of-services.com

Lemrabet, Y., Bigand, M., Clin, D., Benkeltoum, N.,

Bourey, J.-P., 2010. Model Driven Interoperability in

practice : preliminary evidences and issues from an

industrial project, in: First International Workshop on

Model-Driven Interoperability (MDI’10). ACM, pp.

3–9.

LNEC [WWW Document], 2013. URL

http://www.lnec.pt/organizacao/dbb/nmmf/estudos_id

(accessed 6.10.13).

Maglio, P.P., Srinivasan, S., Kreulen, J.T., Spohrer, J.,

2006. Service Systems, Service Scientists, SSME, and

Innovation, in: Communications of the ACM - Services

Science, vol 49, issue 7, pp. 81–85.

MSEE Project [WWW Document], 2012. URL

http://www.msee-ip.eu/project-overview (accessed

4.10.12).

OMG, 2011. Model Driven Architecture [WWW

Document]. URL http://www.omg.org/mda (accessed

12.20.11).

TIMBUS, 2013. TIMBUS Project page [WWW

Document]. URL http://timbusproject.net (accessed

5.15.13).

MODELSWARD2014-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment

688