MoDELS / UML 2005



Travel information



Doctoral symposium

Educators symposium

Keynote speakers


Technical papers





Conference calls

Paper submission

Important dates

Previous editions



Workshop Title Schedule Summary Website
W1 Tool Support for OCL and Related Formalisms - Needs and Trends Tue, Oct 4
W2 MoDeVA -- Model Design and Validation Mon, Oct 3
W3 MARTES -- Modeling and Analysis of Real-Time and Embedded Systems Tue, Oct 4
W4 Aspect Oriented Modeling Sun, Oct 2
W5 Model Transformations in Practice Mon, Oct 3
W6 WiSME -- Workshop in Software Model Engineering Mon, Oct 3
W7 Model Driven Development of Advanced User Interfaces Sun, Oct 2
W8 NfC -- Models for Non-functional Aspects of Component-Based Software Tue, Oct 4
W9 MDD for Software Product-lines: Fact or Fiction? Sun, Oct 2
W10 Use Cases in Model-Driven Software Engineering Sun, Oct 2

Day Sunday, Oct 2 Monday, Oct 3 Tuesday, Oct 4
Workshop W4, W7, W9, W10 W2, W5, W6 W1, W3, W8

W1: Tool Support for OCL and Related Formalisms - Needs and Trends
Tuesday, October 4, 2005
Thomas Baar, EPFL Lausanne, Switzerland (contact)
Dan Chiorean, University of Cluj-Napoca, Romania
Alexandre Correa, University of Rio de Janeiro, Brazil
Martin Gogolla, University of Bremen, Germany
Heinrich Hußmann, University of Munich, Germany
Octavian Patrascoiu, University of Kent, United Kingdom
Peter H. Schmitt, Universität Karlsruhe, Germany
Jos Warmer, Ordina, The Netherlands

Model-centric methodologies and new technologies such as MDA, MDSE, LDD, or DSL attract now a lot of attention both in academia and in industry. Since they propagate a shift from the implementation code to more abstract but nevertheless detailed and precise models, their successful application in industrial projects heavily depends on matured tools support. The traditional way to make a model more precise is by using a textual constraint language such as OCL. Recently, an increasing amount of work has been spent on OCL tools by various organizations so that software developers has today the choice among more than 10 academic and commercial tools. However, compared to similar tools supporting other textual languages, e.g. integrated development environments (IDEs) for Java, tools for OCL are still rather archaic.

The increasing importance of OCL for model-centric methodologies on one hand and the improving but not perfect tool support for OCL on the other hand naturally raise a lot of questions. Which features should an OCL tool offer to encourage the usage of OCL in practice? Is it feasible to make OCL more executable and to provide an animator for OCL? Which consequences for future tools have the fact that OCL is incorporated in a number of other formalisms? Should we strive for a common architecture of OCL tools which would enable us to reuse standard components? What is the relationship between OCL and similar formalisms such as JML, SQL, or graph-grammar based formalisms? Are there unclear issues in the OCL language descriptions that still prevent a smooth tool support?


W2: MoDeVA -- Model Design and Validation
Monday, October 3, 2005
Benoit Baudry (INRIA, France)
Christophe Gaston (CEA/LIST, France)
Sudipto Ghosh (Colorado State University, USA)

Design and validation methods appear to be more and more necessary in an industrial context. This fact is due to several factors. Software systems are more and more complex and cannot be understood by a stand-alone human without a proper standardisation. MDA(Model Driven Architecture) or more generally object/component oriented design methods have been defined in order to overcome a part of this problem. Moreover, systems large scale and complexity induce important risks of bugs or unpredicted behaviours resulting from interactions between subsystems. Formal methods have been intensively applied to evaluate reliability of systems. These methods generally require adequate specification and structuring languages to describe (a part of) the system under validation. For instance, modular first order languages are suitable for this purpose. One of the main problem encountered when trying to combine design and validation features, is that structuring languages suitable for one of the features are generally not suitable for the other. In this way, object-oriented paradigm is suitable for large scale system design, since it allows anthropomorphic design based on services exchanges of basic entities. However, this paradigm is not suitable (without restriction) for validation activities, since any enrichment of a system is likely to cause loss of global properties. In the opposite way, modular paradigm ensures properties preservation but the price to pay is an amount of design difficulties.

The MoDeVa (Model Design and Validation) workshop aims at being a forum for researchers and practitioners with varying backgrounds to discuss new ideas concerning links between model-based design and model-based validation. More precisely, topics of interest include design processes that support complex system modelling and formal or semi-formal refinement mechanisms. In the frame of validation methodology, model-based testing will be considered as "first-class-citizen" since testing is the primary used technique in the industrial context. Design methodologies including considerations on properties preservation (non-regression testing for example) will be appreciated. Languages to describe or validate models include UML and its MDE (Model Driven Engineering) and MDA aspects, algebraic languages, automata-based language, first order language, propositional languages... The considered design paradigm may be inherited from programming language field, like object oriented design, or more abstract, like component or feature based modelling.


W3: MARTES -- Modeling and Analysis of Real-Time and Embedded Systems
Tuesday, October 4, 2005
Organizers: Sebastien Gerard (CEA, France)
Susanne Graf (Verimag, Grenoble, France)
Øystein Haugen (Univ. of Oslo, Norway)
Iulian Ober (Verimag, Grenoble, France)
Bran Selic (IBM, Canada)

This workshop is a merge of the former workshop series SIVOES and SVERTS. The concern of this workshop is the use of MDA in the context of Real-time, distributed and embedded systems, where a particular emphasis is put on modeling, semantic issues and methods and tools for analysis.


W4: Aspect Oriented Modeling
Sunday, October 2, 2005
Omar Aldawud (Lucent Technologies, USA)
Tzilla Elrad (Illinois Institute of Technology, USA)
Jeff Gray (University of Alabama at Birmingham, USA)
Mohamed Kandé (Condris Technologies, Switzerland)
Jörg Kienzle (McGill University, Canada)
Dominik Stein (University of Duisburg-Essen, Germany)

Aspect-orientation is a rapidly advancing technology. New and powerful aspect-oriented programming techniques are presented at the International Conference on Aspect-Oriented Software Development every year. However, it is not clear what features of such techniques are "common aspect-oriented concepts" and what features are rather language-specific specialties. Research in Aspect-Oriented Modeling (AOM) has the potential to help find such common characteristics from a perspective that is at a more abstract level (i.e., programming language-independent). The ultimate goal of research in AOM is to provide aspect-oriented software developers with general means to express aspects and their crosscutting relationships onto other software artifacts.

This workshop aims to identify and discuss the impacts of aspect-oriented technologies on software modeling, and to set up a shared agenda for future research in aspect-oriented modeling of software systems. To achieve these goals, we invite the participants to present new ideas and discuss the state of research and practice in modeling different kinds of crosscutting concerns at multiple levels: software architecture, detailed design, testing, and mapping models onto aspect-oriented programs. The results of the workshop are expected to contribute towards answering the following key questions:

  • How do aspects emerge and appear in models?
  • In what respect do they help to understand the problem domain? And how do they help to find "better" software solutions?
  • In what regards are current modeling techniques suitable to design aspects?
  • In what respect do they fail to do so?
  • How could those deficiencies be resolved?


W5: Model Transformations in Practice
Monday, October 3, 2005
Jean Bezivin (University of Nantes)
Bernhard Rumpe (TU Braunschweig)
Andy Schuerr (TU Darmstadt)
Laurence Tratt (King's College London)

Model Transformations in Practice is a workshop to be held at the upcoming MoDELS 2005 conference. It aims to provide a forum for the model transformation community to discuss practical model transformation issues. Currently, many different model transformation approaches have been proposed and explored, but there has been too little work on comparing and contrasting various approaches. Without such comparisons, it is hard to assess new model transformation approaches, or to discern sensible future paths and upcoming standards such as the upcoming OMG MOF/QVT recommendation.

The aim of this workshop is to lead to an increased understanding of the relative merits of different model transformation techniques and approaches. A more advanced understanding of such merits is of considerable benefit to both the model transformation and wider modelling communities.


W6: WiSME -- Workshop in Software Model Engineering
Monday, October 3, 2005
Krzysztof Czarnecki, University of Waterloo, Canada
Jean-Marie Favre, University of Grenoble, France
Martin Gogolla, University of Bremen, Germany
Tom Mens, University of Mons-Hainaut, Belgium

Model-Driven Engineering is a form of generative engineering, by which all or at least central parts of a software application are generated from models. Model Driven Engineering should be seen as an integrative approach combining existing software engineering techniques (e.g., testing and refinement) and technical spaces (e.g., 'ModelWare' and 'XmlWare') that have usually been studied in separation. The goal of the workshop is to improve common understanding of these techniques across technical spaces and create bridges and increase the synergies among the spaces. This year's WiSME workshop will concentrate on two complementing themes: Bridging Technical Spaces and Model-Driven Evolution.


W7: Model Driven Development of Advanced User Interfaces
Sunday, October 2, 2005
Jan Van den Bergh, Hasselt University, Belgium
Heinrich Hussmann, University of Munich, Germany
Andreas Pleuss, University of Munich, Germany
Stefan Sauer, University of Paderborn, Germany

The user interface of an application is often one of the core factors determining its success. While model-driven development is gaining popularity in the software engineering community, model-based user interface development is an important line of research in the human-computer interaction community. Both approaches make extensive use of models to develop software, but currently they are still vastly independent. This workshop aims at integrating the knowledge from both domains, leading to a model-driven development of user interfaces. In particular the focus lies on advanced user interfaces corresponding to the current state-of-the-art in human-computer interaction, such as interfaces supporting complex interactions, visualizations, multimedia representations, multimodality, adaptability, or customization.


W8: NfC -- Models for Non-functional Aspects of Component-Based Software
Tuesday, October 4, 2005
Jan Øyvind Aagedal (SINTEF ICT, Norway)
Geri Georg (Colorado State University, USA)
Raffaela Mirandola (University of Roma "TorVergata", Italy)
Ileana Ober (IRIT, France)
Dorina Petriu (Carleton University, Canada)
Wolfgang Theilmann (SAP Research Belfast, Ireland)
Jon Whittle (George Mason University, USA)
Steffen Zschaler (Technische Universität Dresden, Germany)

Developing reliable software is a complex, daunting, and error-prone task. Therefore, many researchers are interested in improving the support for developers creating such software. Component-based software engineering has emerged as an important paradigm for handling complexity. In parallel, raising the level of abstraction when reasoning about systems, thus using models, is another technique for lowering the complexity. The goal of this workshop is to look at issues related to the integration of non-functional property expression, evaluation, and prediction in the context of component- based software engineering and finding the best techniques to deal-with non- functional aspects in a model based approach, such as, but not limited to, UML-based approaches. This includes semantic issues, questions of modelling language definition, but also support for automation, such as analysis algorithms, MDA-based approaches, or tool-support for refinement steps. As models are only really meaningful if used in the context of a software development process, we also welcome work in this area.

We expect the workshop to foster cooperation between the various research groups in the field. One important expected outcome is a joint workshop report as well as ongoing discussions, e.g., on a workshop mailing list. The aim of this workshop is to bring together practitioners and academics that are currently working around these topics to highlight the ongoing solutions and the problems encountered. The workshop is organized on two half-day sessions: The morning session is dedicated to invited talks and presentations; followed in the afternoon by working sessions. The number and subject of these sessions will be decided by the organizers depending on the position papers.


W9: MDD for Software Product-lines: Fact or Fiction?
Sunday, October 2, 2005
Dr. Douglas C. Schmidt (
Andrey Nechypurenko (
Egon Wuchner (

Scope: key advantages of using model-driven development in conjunction with commonalityvariability analysis (CVA) are (1) rigorously capturing the key roles and responsibilities in a CVA and (2) helping automate repetitive tasks that must be accomplished for each product instance. Often, however, new customer requirements invalidate the results of earlier CVAs, such that a CVA and its derived meta-models, DSMLs, and generators must be modified invasively and intrusively to reflect these new requirements. The primary scope of this workshop will be on theory and methods to reduce the impact of the new unanticipated requiremetns on the (meta)models and model interpretes in order to improve the usability of model-based technologies in real-life large scale applications.


W10: Use Cases in Model-Driven Software Engineering
Sunday, October 2, 2005
Hernán Astudillo (Universidad Técnica Federico Santa María, Valparaíso, Chile)
Gonzalo Génova (Universidad Carlos III de Madrid, Spain)
Michal Smialek (Warsaw University of Technology, Poland)
Juan Llorens (Universidad Carlos III de Madrid, Spain)
Pierre Metz (Darmstadt University of Applied Sciences, Germany)
Rubén Prieto-Díaz (James Madison University, USA)

The integration of use cases within Model Driven Software Engineering requires a better definition of use case contents, in particular use case description of behavior through sequences of action steps, use case pre- and post- conditions, and relationship between use case model and conceptual model. The UML2 specification allows for several textual and graphical representations of use case behavior, but does not provide any rules for transformations between different representations at the same level of abstraction. It does not provide either any rules for transformations of these representations to other artifacts at levels closer to implementation. With this workshop we hope to show how the resourceful application of use case models help to fill the "requirements gap" in the current resarch and practice of model-driven methodologies.


Last updated: July 6, 2005