Conférences invitées

Modular Design of Hybrid Simulation Languages by Hans Vangheluwe

The engineering of a complex Cyber-Physical System (CPS) commonly involves the creation and subsequent simulation of hybrid models. Such models are not expressible in a single existing formalism but rather require a carefully crafted hybrid modelling language, based on a combination/coordination of the consituent languages.
Modular language engineering is thus essential for effective and efficient development of new formalisms, appropriate for a task.

In our approach, each modelling language and its experimentation environment is described using a modular representation of its syntax and simulation semantics.
A white-box technique is presented for explicitly modelling the definition and composition of these specifications.
Once the semantics of the hybrid language is well-understood, the composition specification can be used to synthesize a co-simulation "master" coordination algorithm.

When the combination/coordination/orchestration of the invididual languages' semantics is explicitly modelled, in an appropriate formalism (Timed Automata in our case), analysis of properties (e.g., through model checking) of the constructed hybrid language becomes possible.

The approach is demonstrated by creating a few hybrid languages through composition of simple building block languages such as Timed Finite State Automata (TFSA).
The automated analysis of language properties such as legitimacy and determinism using the tool UPPAAL will be demonstrated.

Hans Vangheluwe is a Professor in the Antwerp Systems and software Modelling (AnSyMo) group within the department of Mathematics and Computer Science at the University of Antwerp in Belgium, and an Adjunct Professor in the School of Computer Science at McGill University, Montreal, Canada.  AnSyMo is an Associated Lab of Flanders Make, the strategic research centre for the Flemish manufacturing industry.
In a variety of projects, often with industrial partners, he develops and applies the model-based theory and techniques of Multi-Paradigm Modelling (MPM) in application domains as diverse as waste water treatment and automotive software.
He is the chair of the EU COST Action IC1404 Multi-Paradigm Modelling for Cyber-Physical Systems (MPM4CPS).


The Discipline of Modeling and Simulation by Dr. Diallo

Modeling and Simulation (M&S) is emerging as a discipline that is increasingly being applied beyond the field of engineering. In fact, a variety of domains ranging from healthcare to the social sciences and the humanities are now adapting M&S as a viable method of scientific inquiry.  In this presentation, we make the case that M&S is in fact its own discipline and present fundamental problems that belong specifically to the M&S domain. We present an overview of M&S thinking along with practical real word examples and discuss how M&S approaches can be applied in different context. Our main objective is to demonstrate that theories, methods and techniques used in M&S are generally applicable to other domains.

Dr. Diallo has studied the concepts of interoperability of simulations and composability of models for the last ten years. He is VMASC's lead researcher in Modeling and Simulation Science where he focuses on applying Modeling and Simulation as part of multidisciplinary teams to study social phenomena, religion and culture. Dr. Diallo is also involved in developing cloud based simulation engines and User Interfaces to promote the use of simulation outside of the traditional engineering fields.

Dr. Diallo graduated with a M.S. in Engineering in 2006 and a Ph.D. in Modeling and Simulation in 2010 both from Old Dominion University. He is the Vice President in charge of conferences and a member of the Board of Directors for the Society for Modeling and Simulation International (SCS). Dr. Diallo has over one hundred publications in peer-reviewed conferences, journals and books chapters.


A Framework to Holistic Modeling & Simulation of Healthcare Systems by Mamadou K Traoré

(Work done in collaboration with Bernard Zeigler, Raphael Duboz, Gregory Zacharewicz)

In this presentation, we introduce a modeling and simulation methodology to support a holistic analysis of healthcare systems through a stratification of the levels of abstraction into multiple perspectives and their integration in a common simulation framework. In each of the perspectives, models of different components of healthcare system can be developed and coupled together. Concerns from other perspectives are abstracted as parameters (i.e., translation of assumptions and simplifications) in such models. Consequently, the resulting top model within each perspective can be coupled with its experimental frame to run simulations and derive results. Components of the various perspectives are integrated to provide a holistic view of the healthcare problem and system under study. The resulting global model can be coupled with a holistic experimental frame to derive results that couldn’t be accurately addressed in any of the perspective taken alone.

Mamadou K Traoré received his BSc in Mathematics (1987), MSc in Computer Science (1989) and PhD in Computer Science (1992) from Blaise Pascal University (Clermont-Ferrand – France). His current research, within the LIMOS CNRS UMR 6158 (Laboratoire d’Informatique, de Modélisation et d’Optimisation des Systèmes), is on formal specifications, symbolic manipulation and automated code synthesis of simulation models, with an emphasis on DEVS. He is also adjunct Professor at the African University of Science and Technology (AUST, Abuja, Nigeria).



Spatial statistics for monitoring land use changes: towards optimal use of big data by Alfred Stein

Changes in land use are investigated widely with remote sensing images. The quality of the images, in terms of their resolution, frequency and spectral characteristics has been improved and is now better suited than ever to reveal important characteristics. Land use, in contrast to land cover, is defined as the use of the land. Much can be detected from the cover, which can readily be observed, but a proper analysis and interpretation, hence the actual relevant information on land use, is more difficult to obtain. It requires more information and good and solid statistical analysis methods. Most importantly, big data and proper ways for analysis now play a role, as satellite information is of an increasing volume. Land use change monitoring plays a prominent role not only in the developed world, but in particular also in the rapidly developing countries in the global south. Land use changes concern both natural, agricultural and urban environments.

In this presentation, attention will be given to several of these environments. The focus in this presentation is on the methodology, i.e. the role of spatial statistics and image analysis for land use changes.

We distinguish the following:

  • Land use in tropical areas as affected by forestry, agriculture and deforestation. Attention is given to spatial statistical modeling of deforestation in the Amazonian area, where typically land use changes gradually. Spatial point patterns and their usability is showing patterns of deforestation.
  • Hazards, in particular landslides, often occur in areas that have been influenced by human activities. A Bayesian logistic regression model is devoted to landslides occurring along a road section in the Indian Himalayas.
  • Urban land use is investigated within large cities, attention is given to the use of high resolution satellite images that are providing more and better information than ever. Here, modelling of roads and urban city blocks is done for the city of Wuhan in China, using methods based on fuzzy logic.
  • In developing countries like in African countries, the identification and delineation of slum areas for cities is a major issue: they develop rapidly and their managing requires a careful monitoring. In a recent study, expert knowledge and random sets were used to obtain relevant information on where to draw the boundary between a slum area and the surrounding city.

The presentation concludes that spatial statistical methods and remote sensing images provide a good match to address important aspects related to sustainable development goals.

Keywords: Land use, spatial statistics, image analysis.

Alfred Stein is professor in Spatial Statistics and Image Analysis at the department of Earth Observation Science at ITC (University of Twente, Netherlands). His research interests focus on statistical aspects of spatial and spatio-temporal data, like monitoring data, in the widest sense. Optimal sampling, image analysis, spatial statistics, use of prior information, but also issues of data quality, fuzzy techniques, random sets, all in a Bayesian setting. From 1998 onwards he has been working with more than 30 PhD students on a range of spatial (and temporal) statistical topics. Alfred Stein is a member of the CT de Wit Research School for Production Ecology and Resource Conservation and of the SENSE research School. Since 2011 he is the editor-in-chief of the Spatial Statistics journal, the new leading platform in the field of spatial statistics. It publishes articles at the highest scientific level concerning important and timely developments in the theory and applications of spatial and spatio-temporal statistics. He is associate editor of the International Journal of Applied Geoinformation and Earth Observation, Statistica Neerlandica and Environmental and Ecological Statistics. At present, 11 PhD students are working under his supervision.He is editor of several books and of various special issues of journals. (Cf.



The importance of ontological structure: why validation by ‘fit-to-data’ is insufficient by Gary Polhill and Doug Salt

This chapter will briefly describe some common methods by which people make quantitative estimates of how well they expect empirical models to make predictions. However, the chapter’s main argument is that fit-to-data, the traditional yardstick for establishing confidence in models, is not quite the solid ground on which to build such belief some people think it is, especially for the kind of system agent-based modelling is usually applied to. Further, the chapter will show that the amount of data required to establish confidence in an arbitrary model by fit-to-data is (big data aside) often infeasible. This arbitrariness can be reduced by constraining the choice of model, and in agentbased models, these constraints are introduced by their descriptiveness rather than by removing variables from consideration or making assumptions for the sake of simplicity. By comparing with neural networks, we show that agentbased models have a richer ontological structure. For agent-based models in particular, this richness means that the ontological structure has a greater significance, and yet is all-too-commonly taken for granted or assumed to be ‘common sense’. The chapter therefore also discusses some approaches to validating ontologies, though the state-of-the-art is far from a situation in which there are established standards.

(Abstract from the Preprint of book chapter: Polhill, G. and Salt, D. (2017) The importance of ontological structure: why validation by ‘fit-to-data’ is insufficient. In Edmonds, B. and Meyer, R. Simulating Social Complexity: A Handbook, 2nd Edition. Cham: Springer. pp. 141-172 (doi: 10.1007/978-3-319-66948-9_8)

 Keywords: Agent-Based Model, fit-to-data, ontological structure.


Combining survey data and spatial econometrics to estimate willingness to pay for residential housing attributes by Dawn Parker

Dawn Parker is Professor in the School of Planning, Faculty of Environment, University of Waterloo, Canada. Her research focuses on the development of fine-scale models that link the drivers of land-use change and their socioeconomic and ecological impacts, with completed and ongoing projects on organic agriculture in California’s Central Valley, timber harvest and carbon sequestration in eastern deciduous forests in West Virginia, U.S.A., the effects of HIV/AIDS on smallholder agricultural households in Uganda, interactions between land markets, landscaping, and carbon sequestration in ex-urban landscapes, and the co-evolution of urban transit networks and residential neighbourhoods. Her areas of technical expertise include agent-based modelling, land-use modelling, and environmental and resource economics.

She received her BA in Economics from Lewis and Clark College and her PhD in Agricultural and Resource Economics from U.C. Davis. She then completed a post-doctoral fellowship in modelling with Elinor Ostrom at Indiana University. Previously to joining UW, she was a founding member of the Center for Social Complexity and Department of Computational Social Science at George Mason University, USA, where she served as director for the PhD program in computational social science. She has served as Associate Director and Director of the Waterloo Institute for Complexity and Innovation and will return for a second term as Director in 2018.  She formerly served on the steering committee of the Global Land Project and on the editorial board of the Journal of Land Use Science.  She currently serves on the editorial boards of Computers, Environment, and Urban Systems; Change and Adaptation in Socio-Ecological Systems; and Socio-Environmental Systems Modeling. (Cf.