Seminars

On Proactivity And Situation Awareness In The Interactive Computing Research

By: Prof. Merik Meriste

Abstract: Interaction is a pervasive aspect of computing today. However a satisfactory foundational framework for interactive computation, like TM for algorithms, is still lacking. Our group’s interests are in the area of interactive systems for some years. We study foundations and implementation tools of situation aware proactive systems embedded in some environment.
We are convinced that the further research should focus on potential ways of designing, assembling, and analysis of computing systems based on autonomous, proactive components. In many applications it is important that the new theories and methods focus specifically support the verifiable satisfaction of situation awareness requirements imposed upon the computing systems.
This talk presents our approach in studies of proactivity and situation awareness. By proactivity we mean component’s ability to anticipate the evolution of its environment, to choose goal-directed activities for better satisfaction of the component’s goals, and in the case of a well-designed system to better satisfaction of the system’s goal. We will consider aims and directions of our on-going long-term research framework on proactivity and situation awareness of artefacts. Three simultaneously on-going and interacting threads are considered in this framework - proactive modelling, technology platforms and tools, and pilot applications and assessment methods. The emergent behaviour in proactive systems and their relationship to complex systems theory are of special interest.

Bio: A senior researcher at the Research Laboratory for Proactive Technologies, Tallinn University of Technology (TUT) and a professor of proactive system technologies at the University of Tartu (UT). He currently conducts research for modelling and simulation of situation-aware autonomous (proactive) agents and their networks (TUT) and applied research for agent technologies (UT). Focus of the research is on modelling network-enabled capabilities by means of networked situation aware autonomous agents. His group in UT has developed a novel approach of self-organising proactive maps applied in various situation aware decision support systems.
He obtained a M.Sc. degree in Applied Mathematics (1973) and a Doctoral Degree in Software Systems and Engineering (1984) from the University of Tartu. Has almost 40 years experience in software development, computer science and software systems research and teaching in Universities of Tartu, Helsinki, Turku and Kuopio and Tallinn University of Technology. He is member of NATO RTO task groups, Estonian representative in ESF COST ICT Domain Committee, Estonian expert member of the Fr7 ICT Committee.

A Graph Transformation-Based Framework to Detect Interactions Between Cross-Cutting Modules

By: Tim Molderez

Abstract: Several techniques have been developed to enhance support for modularizing crosscutting concerns, such as aspect-oriented, context-oriented and role-oriented programming. While powerful, the language constructs introduced by these techniques also come with their own set of challenges. One of the major challenges are the complex interactions that can arise between so-called crosscutting modules, which is an umbrella term for language constructs that can encapsulate crosscutting concerns (e.g. aspects, layers, roles, ..). Some of the interactions between crosscutting modules may not be created intentionally; they can lead to unexpected program behaviour and are often hard to detect by manually inspecting source code.
This presentation discusses a framework that can be used to expose such interactions in object-oriented languages that support crosscutting modules. This framework transforms the source code to be analysed to a corresponding graph representation in the delMDSOC model, which is suited to reason about crosscutting modules. The execution semantics of the delMDSOC model, specified as a set of graph transformation rules in the GROOVE tool, is then used to exhaustively search the program's state space while looking for interaction patterns.

Bio: Tim Molderez has a MSc Computer Science at the University of Antwerp (Belgium) and is currently working there as a Ph.D. student at the Ansymo software engineering research group.
His research is concerned with programming languages that support modularized crosscutting concerns, analysing the impact these languages' new constructs have on each other and on the base system. Tim's main research interests are: multi-dimensional separation of concerns, model checking and graph rewriting.

Building VCL Models and Automatically Generating Z Specifications from Them

By: Nuno Amalio

Abstract: VCL is a visual and formal language for abstract specification of software systems. Its novelty lies in its capacity to describe predicates visually. This paper presents work-in-progress on a tool for VCL; the tool version presented here supports the VCL notations of structural and assertion diagrams (a subset of the whole VCL suite), enabling the generation of Z specifications from them.

Bio: Nuno Amalio was awarded a PhD in Computer Science with the University of York (UK) in 2007. He also holds an MSC in Software Engineering with the same University, and a BsC in Computer Science with the University of Lisbon (Portugal). Currently, Nuno is a post-doctoral research collaborator at the University of Luxembourg. Previously, he worked as post-doctotoral research fellow at the City University of London, and he spent one year in industry working at Praxis High Integrity Systems Ltd (Bath, UK) in the development of large-scale critical systems. Nuno's research interests are in the areas of: software engineering, security, formal methods, visual languages and model-driven development.

Ressource Allocation in Delayed/Disruption Tolerant Network

By: Gabriel Sandulescu

Abstract: Over recent years, the popularity of mobile networks has grown as a consequence of the ubiquity of communication devices. In many mobile networks, nodes can move freely and connect to their neighbours as they move into each other's radio range. However, due to high node mobility, limited bandwidth or energy, radio obstruction or malicious attacks, the route between source node and destination node may be disrupted frequently and for relatively long periods of time. Traditional adhoc networking models that assume continuous connectivity are largely incapacitated in such challenged environments. In response to that, a new class of networks, known as delay- and disruption-tolerant networks (DTNs), have emerged with applications in vehicular communications, emergency response, the military, and wildlife monitoring, to mention a few. This thesis undertakes a study of the resource-aware routing problem in DTNs along four lines.
First, the thesis presents the design and comparative evaluation of a delay-tolerant routing protocol (ORWAR) which optimises message transmission and bandwidth usage during opportunistic encounters. This is done through the estimation of contact duration between nodes, the selection of the most suitable message to forward at any contact opportunity, and message differentiation.
Second, the thesis proposes a hybrid scheme whereby opportunistic and infrastructure-based communication can be combined to overcome network partitions and packet losses. Collaboration between two networks characterised by different capacities, costs, and performance levels has been shown to be not only cost-effective, but also capable of increasing network survivability in the combined network.
Third, it proposes and evaluates a mathematical model and an algorithm that can be used to compute the optimal level of redundancy and replication of a routing protocol as a function of message characteristics, such as size and time-to-live.
Fourth, a holistic approach to resources is proposed, where variations in the spatial and temporal distribution of various resources can feed strategies to reduce resource consumption. Using estimates of vicinity resources, a routing protocol may not only use up fewer resources overall, but may also consume resources preferentially from nodes with higher resource levels, sparing whenever possible those with limited supplies.

Bio: Gabriel Sandulescu works towards his Ph.D. in Computer Science at the University of Luxembourg under the supervision of Professor Simin Nadjm-Tehrani. He is also employed part-time with the European Fund Administration in Luxembourg, where he has been involved in software engineering since 1999. His current research interests include new service paradigms in conjunction with network convergence, resource allocation in mobile networks, and distributed systems.

Test generation for LTE functionality with Conformiq Designer and SpecExplorer

By: Olga Grinchtein

Abstract: LTE RAN Integration & Verification organization at Ericsson AB is responsible for integration and verification of LTE Radio Access Network. LTE, Long Term Evolution, is the next generation mobile network beyond 3G. The technology allows for speeds more than 300Mbps.
LTE Radio Access Network consists of LTE Radio Base Station (RBS), which supports the LTE air interface and performs radio resource management. 
LTE RBSes are interconnected with each other by means of the X2 interfaces. Testers at the organization are responsible for test design and test execution of different features of LTE RBS. While a lot of work have been done on automation of test execution, test design is still manual process. One way to generate test cases automatically is by using model-based testing. Evaluation of model-based testing approach was done in a project with Conformiq Designer and SpecExplorer, tools for automatically designing and creating test cases. During the project, the tools was used for test generation for security functionality during X2 and intra-cell handover, as well as for input generation to distribution of resources algorithm.

Bio: Olga Grinchtein is a Test Research Engineer at LTE RAN Integration & Verification organization, Ericsson AB in Stockholm.
She received MSc in Computer Science from The Weizmann Institute of Science in Israel in 2002. She received a Ph.D. degree in Computer Science from Uppsala University in Sweden in 2008.
Her research interests include model-based testing, combinatorial testing, mutation testing, model learning and timed systems.

Safe Co-Evolution of Models and Access Control Policies

by: Qin Zhang

Abstract: The access control is becoming a mandatory feature for most systems due to the importance for preserving confidentiality and safety both of clients and internal privacy. A challenging problem is to manage system functionality and access control policy during its evolution.
 
In this paper, we develop a theory of safety co-evolution of models (behavioral description of the system) and access control policies (expressing permissions and prohibitions). We propose a mapping mechanism from access control policies to model properties, safety and reachability, and techniques for property preservation during model evolution process. Thus our proposal allows the incremental modification of the behavioral and access control concerns, with safe mechanism to guarantee that permissions and prohibitions are invariant when the system evolves.

Bio: Qin ZHANG is a PhD candidate at LASSY group of Faculty of Sciences, Technologies and Communications of the University of Luxembourg and SMV group of Faculty of Information Science and Technology of University of Geneva (involved in a joint research project) since 2010.
 
He has a M.Sc degree in Computer Science and Engineering from U.Aizu (Japan) and a M.Eng degree in Software Engineering from HIT (China). His main research interests are software modeling and verification and resilience engineering.

A Formal Framework for Dependability and Resilience from a Software Engineering Perspective

Abstract: This presentation aims at providing a rigorous conceptual framework for defining the concepts of dependability and resilience. The terms dependability and resilience, since the seventies, have been used in nearly all the computer information systems and computer science fields. The introduction and use of these concepts in all these fields makes it difficult to have a common and precise definition of the concepts of dependability and resilience. Having such definition is nevertheless mandatory for the software and systems engineering research community that create development processes, languages and tools to support the engineering of products that would be required to be dependable or resilient.
For this, we introduce an abstract and generic terminology, defined mathematically, to be used when speaking about dependability and resiliency. We also provide some abstract semantic descriptions to these terminological elements. This formal framework is defined from a software engineering perspective, which means that we define its components such that they are useful for the development or improvement of analysis, architectural design, detailed design, implementation, verification and maintenance phases. To this aim we provide the necessary elements in accordance with a model driven engineering perspective that enable the definition of new modeling language for dependable and resilient systems.

Bio: Nicolas Guelfi is professor at the Faculty of Sciences, Technologies and Communications of the University of Luxembourg since March 1999, where he teaches, directs PhD students and makes research in collaboration with national and international partners.
Currently, he is the head of the Laboratory for Advanced Software Systems. His main research and development activities concern the engineering and the evolution of reliable and secure distributed and mobile systems based on semi-formal or formal methods. Focus is made on model driven engineering and reuse based on product line engineering methods. He is the author of around 80 publications in books, journals, conferences and workshops.
He has been for three years the Luxembourgian ERCIM representative at the executive committee of the ERCIM consortium and he is co-chairman of the ERCIM working groups on Software EngineeRing for rEsilieNt systEms (SERENE - http://serene.uni.lu ). SERENE considers resilient systems as open and distributed systems that can dynamically adapt in a predictable way to unexpected events. The research group of Nicolas Guelfi is made of PhD students, engineers and post-doctoral members that are running national or international research projects with important cooperation with universities or industries.

Model-Driven Mutation Applied to Adaptative Systems Testing

By: Alexandre Bartel

Abstract: Dynamically Adaptive Systems modify their behavior and structure in response to changes in their surrounding environment and according to an adaptation logic. Critical systems increasingly incorporate dynamic adaptation capabilities; examples include disaster relief and space exploration systems. In this paper, we focus on mutation testing of the adaptation logic. We propose a fault model for adaptation logics that classifies faults into environmental completeness and adaptation correctness. Since there are several adaptation logic languages relying on the same underlying concepts, the fault model is expressed independently from specific adaptation languages. Taking benefit from model-driven engineering technology, we express these common concepts in a metamodel and define the operational semantics of mutation operators at this level. Mutation is applied on model elements and model transformations are used to propagate these changes to a given adaptation policy in the chosen formalism. Preliminary results on an adaptive web server highlight the difficulty of killing mutants for adaptive systems, and thus the difficulty of generating efficient tests.

Bio: Alexandre Bartel is a PhD student at the Security, Reliability and Trust (SnT) Center at University of Luxembourg. He has a MS degree in Computer Science from KTH (Sweden) and GrenobleINP-ESISAR (France). His main research interests are in Service Oriented Architecture (SOA) and Smart-Phone security and testing.

Dependability modeling and evaluation based on AADL and GSPNs

By: Karama Kanoun

Abstract: For efficiency reasons, system designers’ will is to use an integrated set of methods and tools to describe specifications, and also to perform dependability analyses. AADL (Architecture Analysis and Design Language) has proved its capacity for architectural modeling.
This presentation defines a modeling framework allowing the generation of dependability-oriented analytical models from AADL models. The aim is to facilitate the evaluation of dependability measures, such as reliability or availability. The modeling approach is iterative. The AADL dependability model is transformed into a GSPN (Generalized Stochastic Petri Net) by applying model transformation rules. An automatic transformation tool has been developed. The dependability measures are obtained from the processing of the resulting GSPN by existing tools. The approach is illustrated on the modeling fault tolerant software and hardware components and on a subsystem of the French air traffic control system.

Bio: Karama Kanoun is Directeur de Recherche at LAAS-CNRS, heading the Dependable Computing and Fault Tolerance Research Group (http://www.laas.fr/~kanoun/). Her research interests include modeling and evaluation of computer system dependability considering hardware as well as software, and dependability benchmarking.
She has authored or co-authored more than 150 publications. She has co-directed the production of the first book on Dependability Benchmarking. She is Chairperson of the Special Interest Group on Dependability Benchmarking of the International Federation for Information Processing (IFIP), and vice-chairperson of the IFIP working group 10.4 on Dependable Computing and Fault Tolerance. She was the principal investigator of the DBench European project (Dependability Benchmarking), and managed the European Network of Excellence ReSIST, Resilience Survivability in IST.

i-CareFlow: A prescription supporting environment using semantics, web  services and adaptive workflow

By: Cédric Pruski

Abstract: Computer-Interpretable Guidelines (CIG) are Clinical Guidelines described in a language that can be interpreted by computers. They are often used to support physicians in a single point in time to design and test guidelines. The next steps are the application of CIGs to determine careflows, the personalization of careflows and their execution.
In this talk I present the iCareflow ontological framework which intends to support the adaptation of guidelines during the design phase in order to obtain personalized careflows. The proposed framework explores how guidelines, policies of the care institution, patient preferences, terminologies and other knowledge sources interact with each other to determine a patient-centric careflow using Semantic Web technologies. Scenarios and further applications of the framework as well as a technical description are also provided.

Specification and Testing of Model Transformations with Object-Z

By: Iram RUBAB

Abstract: Model Transformations are a key artefact in Model Driven Architecture (MDA). MDA raises the level of abstraction for software development. Model Transformations are the set of rules for transforming a platform independent model into a platform specific model. The specification and testing of model transformations is an important task as they are the major artefacts in MDA based development.  In this work, we propose the use of Object-Z, a formal language to be used as a model transformation language. We evaluate Object-Z on the basis of properties for model transformations as collected from literature. We prove that Object-Z can be used effectively as a model transformation specification language. We present a case study in support of the idea and represent Class Diagram to ER Model transformations using Object-Z.
The second part of the work includes a specification based testing approach for model transformations. The approach uses formal specifications and generates test data based on the specifications. The generated test data is converted back to model to be effectively used for testing of model transformations. We can apply our approach on any model transformation written in a formal language. For the proof of concept we take an example transformation written in Object-Z i.e. from Class Diagram to ER. We apply our approach on this case study and generate test data to be used for testing purpose. A partial automation of the proposed approach is also part of this work.

Bio: Iram Rubab did her MSc in Computer Science in M.A. Jinnah University (Islamabad, Pakistan). She just started her PhD under the supervision of Nicolas Guelfi (LASSY, CSC) in the SETER Project.

Exploiting Resource Heterogeneity in Delay-Tolerant Networks using a Holistic Approach

By: Gabriel SANDULESCU

Abstract: Routing in delay and disruption-tolerant networks (DTNs) relies on intermediary nodes, generally referred to as custodians, in order to deliver messages to destination. However, these nodes usually differ significantly in terms of their levels of available resources, such as energy, buffer space and bandwidth. Consequently, routing algorithms need to efficiently exploit custodians holding abundant resources, while also making sure that custodians whose resources are in limited supply are not exhausted. In this presentation we propose a distributed scheme for calculating the level of resources available in the vicinity of a node, as a tool supporting beneficial routing decisions. Knowledge about resources available in the vicinity allows nodes to implement meaningful custodian election and queue management strategies. These strategies can be approached from a holistic perspective based on the availability of the three resources under consideration in node proximity. We demonstrate that, by using this information, a routing protocol may not only use up fewer resources overall, but may also consume resources preferentially from nodes with higher resource levels. As a result, disparities in available resources across the node population are significantly reduced, and nodes are less likely to leave the network as a consequence of resource depletion.

Bio: Gabriel Sandulescu works towards his Ph.D. in Computer Science at the University of Luxembourg under the supervision of Professor Simin Nadjm-Tehrani. He is also employed part-time with the European Fund Administration in Luxembourg, where he has been involved in software engineering since 1999. He obtained a Dipl.-Ing. degree (MSc equivalent) in Electrotechnical Engineering from the Politehnica University in Bucharest in 1992. His current research interests include new service paradigms in conjunction with network convergence, resource allocation in mobile networks, and distributed systems

A Modular Model Composition Technique

By: Qin Ma

Abstract: Model composition is a technique for building bigger models from smaller models, thus allowing system designers to control the complexity of a model-driven design process. However many current model composition techniques are themselves complex in the sense that they merge the internal elements of the participating models in non-trivial ways.
In this paper we apply some of the ideas from  modular programming to reduce the complexity of model compositions. Indeed we propose a model composition technique with a modular flavor that treats the participating models as black boxes. Our technique has several desirable features: it is simple, it does not require a separate language for expressing the composition, and the understanding of the resulting composed model is made easier by the modular nature of the model composition.

Regression Testing for Aspect-Oriented Programming

By: Yves Le Traon

Abstract: The addition of a cross-cutting concern in a program through aspect weaving has an impact on its existing behaviors. If test cases exist for the program, it is necessary to identify the subset of test cases that trigger the behavior impacted by the aspect. This subset serves to check that interactions between aspects and the base program do not introduce unexpected behavior.Vidock performs a static analysis when aspects are compiled with a program to select the test cases impacted by the aspects. It leverages the pointcut descriptor to locate the set of methods impacted by aspects and the selects the test cases that can reach an impacted method. This static analysis has to perform over approximations when the actual point where the aspect is executed can be computed only at runtime and when test cases call polymorphic objects. We measure the occurence of these assumptions in 4986 projects containing 498 aspects to show the have a limited impact. Then we run experiments with Vidock on five case studies and analyzes the impacts that different types of aspects can have on test cases.

Automatic Product Derivation with Pure::Variants and FeatureMapper

By: Marc Leclerq

Abstract: Software product line engineering is an emerging paradigm that allows organizations developing products using reusable assets from a product line, instead of having to develop the products one by one. The Laboratory for Advanced Software System of the University of Luxembourg has recently developed a platform named REACT that is used to support the development of products in a crisis management product line. In this talk, we will introduce two tools, named Pure::Variants and FeatureMapper, that handle parts of the software product line engineering process such as modeling the domain of the product line and the automatic transformation of product variants, and the results of their application to the REACT platform.

Flexible Model Element Introduction Policies for Aspect-Oriented Modeling

by:  Jacques Klein

Abstract:  Aspect-Oriented Modeling techniques make it possible to use model transformation to achieve advanced separation of concerns within models. Applying aspects that introduce model elements into a base model in the context of large, potentially composite models is nevertheless tricky: when a pointcut model matches several join points within the base model, it is not clear whether the introduced element should be instantiated once for each match, once within each composite, once for the whole model, or based on a more elaborate criteria.
This paper argues that in order to enable a modeler to write semantically correct aspects for large, composite models, an aspect weaver must support a exible instantiation policy for model element introduction. Example models highlighting the need for such a mechanism are shown, and details of how such policies can be implemented are presented.
(Presented at ACM/IEEE MODELS 2010)

Bio: In 2003, Dr Jacques Klein received an engineering degree in Computer Science from the ESSAIM (ENSISA) and a Master in Computer Science from the University of Haute-Alsace, France. He received a Ph.D. degree in Computer Science from the University of Rennes, France in 2006 for a dissertation on the weaving of behavioral models (UML) in a Model-Driven Engineering and Product Line contexts. Part of his doctorate work has been to propose new software engineering tools to reduce the cost and the risk of software development by adapting software systems to wide ranges of new contexts. In 2007/2008, he worked for two years at the University of Luxembourg as a scientific collaborator. He participated in the SPLIT project to propose new transformation languages to support software product lines. He is also participating in the VERITY project to ease the design of reliable software systems. Finally, he successfully proposed two research projects, one on the security testing of resilient systems and one on the mix of SPL and AOSD.
In 2009, he moved to a full time position at the CRP-Gabriel Lippmann, Belvaux, Luxembourg, to work on several IT projects as project manager. He worked in the ISC department of the CRP-Gabriel Lippmann. One of the main research topics of the research centre is to foster the development of high-value added business services by using Service-Oriented Architecture (SOA).
In March 2010, he came back to the University of Luxembourg to work in the team of the Prof. Yves Le Traon.

Content based user centric model of a network search system

by: Peter Vojtas

Abstract: Our goal is a progress in automated processing of web data and user preference search (of course Google does it with its indexes, keyword queries and (anthropological) user profile modeling). Our added value is an improvement of web information extraction (alas semantification) and methods for learning user objectives and utility for a multicriterial top-k user preferences querying.
We present models, methods and prototype experiments, mention connected problems of ontology engineering and uncertainty modeling. At the end I briefly present a profile of our Department of Software Engineering.

Bio: Peter Vojtas is a full professor of Computer Science at the Department of Sofware Engineering (School of Computer Science, Faculty of Mathematics and Physics), Charles University in Prague, Czech Republic.
He also keeps a part time position at the Institute of Computer Science, Czech Academy of Science, Prague. He has held the Alexander von Humboldt Foundation Research Fellow, Bonn, Germany in 90's. He has founded Institute of Computer Science at the P. J. Safarik University, Kosice, Slovak republic and acted as founding director several years. In 2004 he permanently moved to Praue. In spring 1994 he was visiting professor at Ohio University, Athens, Ohio, USA.
Peter Vojtas has MSc Suma Cum Laude in Theoretical Cybernetics and CSc(PhD) in Mathematical Logic from Charles University, Prague. He got Professor Titulaire in Mathematical Informatics from the President of Slovak Republic and DrSc.- Doctor of Science from Czech Academy of Science. His research interests include:

• computational intelligence, connecting web to the user, process of the Web semantization
• user preferences - modelling, learning, querying/searching with, user objectives and utility
• uncertainty and preferences in databases and indexes, web reasoning, information systems
• many valued/fuzzy logic programming/Datalog - deduction, induction, abduction
• formerly - set theoretic aspects of asymptotics and Galois-Tukey connections, etc.

He is a member of ACM, IEEE Computer Society and Computational Intelligence Society, Geselschaft fuer Informatik, Czech Society for Cybernetics and Informatics and DVMLG. He acts in editorial board of Computing and Informatics, Neural Network World, several organizing and program committees of conferences, several bodies of Czech Science Foundation and Ministry of Education.

T-Core: Transformation Languages Taylored to Your Needs!

by:  Eugene Syriani

Abstract:  Nowadays, model-driven engineering (MDE) is an emerging technique in industry. MDE and, in particular, domain-specific modelling are techniques that consist of representing a system as a well-defined model at a level of abstraction familiar to the end user. To manipulate these models, one needs to define a model transformation. Being at the heart of MDE, it is important to properly understand model transformation languages and properly use them.
 
 In this talk, we introduce model transformation by outlining common terminlogy and uses. Then we analyse how the plethora of existing transformation languages solve the problems addressed in MDE. This leads to T-Core, a common basis for engineering model transformation languages. Relaizing that there is a need for different transformation languages tailored specific problems rather than having one general-purpose transformation language, I show how to build a complete transformation language in the presented framework.

Model-Driven Derivation of Crisis Management Applications in a Software Product Line Approach

By:  Yiqing Li

Abstract:  The aim of REACT is to develop a software product line for crisis management systems. A software product line is composed of a software platform, providing reusable components, and of applications derived from the platform to provide functionalities of the crisis management systems. This master thesis uses model-driven engineering techniques (like model transformations) to derive crisis management system applications from the REACT platform.  A car crash management system is used to illustrate the derivation process.

Bio:  Yiqing Li is a master student at the Université Henri Poincaré (Nancy, France)

Using VCL as an Aspect-Oriented Approach to Requirements Modelling

by Nuno Amálio

Abstract:

Software systems are becoming larger and more complex. By tackling the modularisation of crosscutting concerns, aspect-orientation draws attention to modularity as a means to address the problems of scalability, complexity and evolution in software systems development. Aspect-oriented modelling (AOM) applies aspect-orientation to the construction of models. Most existing AOM approaches are designed without a formal semantics, and use multi-view partial descriptions of behaviour.

This talk presents an AOM approach based on the Visual Contract Language (VCL): a visual language for abstract and precise modelling, designed with a formal semantics, and comprising a novel approach to visual behavioural modelling based on design by contract where behavioural descriptions are total. By applying VCL to a large case study of a car-crash crisis management system, the talk shows how modularity of VCL's constructs, at different levels of granularity, help to tackle complexity. In particular, it shows how VCL's package construct and its associated composition mechanisms are key in supporting separation of concerns, coarse-grained problem decomposition and aspect-orientation. The case study's modelling solution has a clear and well-defined modular structure; the backbone of this structure is a collection of packages encapsulating local solutions to concerns.

Short Bio:

Nuno Amálio was awarded a PhD in Computer Science with the University of York (UK) in 2007. He also holds an MSC in Software Engineering with the same University, and a BsC in Computer Science with the University of Lisbon (Portugal). Currently, Nuno is a post-doctoral research collaborator at the University of Luxembourg.

Previously, he worked as post-doctotoral research fellow at the City University of London, and he spent one year in industry working at Praxis High Integrity Systems Ltd (Bath, UK), where he worked in the development of large-scale critical systems. Nuno's research interests are in the areas of: software engineering, security, formal methods, visual languages and model-driven development.

Goal-Oriented Requirement Engineering for Business Processes

by Ken Decreus

Abstract:

The field of requirements engineering (RE) for business processes has grown during the last several years. As business processes are needed to fulfil organizational goals, the information captured in goal models provides a basis for designing business processes. Although research has started to explore how to transform goal models into business process models, current transformation methods need further research.

This paper proposes a tool-supported method to model goals as part of the business requirements for business processes and to automatically generate business process design skeletons that respond to these business requirements.

Short Bio:

Ken Decreus has a master in Computer Science (University of Ghent) and Business Administration (Vlerick Leuven Gent Management School). He worked for Deloitte Consulting from september 2006 till september 2007, and took a leave of abscence to start a PhD track in the field of Requirements Engineering and Business Process Management.

His research has been presented at international conferences (such as International Conference on Requirements Engineering 2009, and ACM Symposium of Applied Computing 2009), and worked together with Marwane El Kharbili and Elke Pulvermueller on Policy-Enabled Goal-Oriented Requirements Engineering for Semantic Business Process Management (accepted for publication in Internation Journal of Intelligent Systems)

Reusability in modeling language specification

by Frédéric Fondement

Abstract:

Model Driven Engineering (MDE) promotes the usage of models as first class artifacts in system development as a mean to cope with complexity. Moreover, to design models, many advocate the use of domain-specific modeling languages by setting the principle “use the right language for the right purpose”. This situation triggers a proliferation of modeling language.

Thus, the easier it is to define a modeling language, the more scalable the entire approach will be. After introducing the concept of language components, will be proposed a language tuning approach, i.e. a mean to create a dialect from an existing language, even though the original language is not prepared to such usage. The approach was applied in the context of the VETESS project, on a model-based test case generator, where the generator originally handling UML models is adapted to handle SysML models.

Short Bio:

Frédéric Fondement is a research and teaching assistant in the computer science and control department of the ENSISA engineering school. His area of interest includes model- and language-driven software engineering. He received in 2007 his PhD. from the Swiss Federal Institute of Technology in Lausanne (EPFL) for his work on concrete syntaxes for modeling languages, and in 2000 his engineering degree from the University of Mulhouse.

In 2002 he was a research engineer at INRIA Rennes where he developed a model transformation language. In 2000-2001 he was part of the research and development team of ObjeXion Software where he developed a web application modeler.

Reusability in modeling language specification

by Frédéric Fondement

Abstract:

Model Driven Engineering (MDE) promotes the usage of models as first class artifacts in system development as a mean to cope with complexity. Moreover, to design models, many advocate the use of domain-specific modeling languages by setting the principle “use the right language for the right purpose”. This situation triggers a proliferation of modeling language.

Thus, the easier it is to define a modeling language, the more scalable the entire approach will be. After introducing the concept of language components, will be proposed a language tuning approach, i.e. a mean to create a dialect from an existing language, even though the original language is not prepared to such usage. The approach was applied in the context of the VETESS project, on a model-based test case generator, where the generator originally handling UML models is adapted to handle SysML models.

Short Bio:

Frédéric Fondement is a research and teaching assistant in the computer science and control department of the ENSISA engineering school. His area of interest includes model- and language-driven software engineering. He received in 2007 his PhD. from the Swiss Federal Institute of Technology in Lausanne (EPFL) for his work on concrete syntaxes for modeling languages, and in 2000 his engineering degree from the University of Mulhouse.

In 2002 he was a research engineer at INRIA Rennes where he developed a model transformation language. In 2000-2001 he was part of the research and development team of ObjeXion Software where he developed a web application modeler.