Conferences & Seminars

Conferences and Workshops related to CIAD





  • SARL17: International Workshop on Agent Modelling and Applications with SARL.
  • Tutorial SARL 2017: Tutorial on SARL at ANT-2017.
  • IFSMS17: 4th International Workshop on Information Fusion for Smart Mobility Solutions.


  • IFSMS16: 3nd International Workshop on Information Fusion for Smart Mobility Solutions.
  • ABMUS16: the 1st Workshop on Agent Based Modelling of Urban Systems, in conjonction with AAMAS16.
  • ABMTRANS16: the 5th International Workshop on Agent-based Mobility, Traffic and Transportation Models, Methodologies and Applications, Madrid, Spain.
  • FNC 2016: 10th International Conference on Future Networks and Communications, at Montréal, Canada.
  • MobiSPC 2016: 12th International Conference on Mobile Systems and Pervasive Computing, at Montréal, Canada.
  • ANT16: the 7th International Conference on Ambient Systems, Networks and Technologies, Madrid, Spain.


  • IFSMS15: 2nd International Workshop on Information Fusion for Smart Mobility Solutions.
  • FNC 2015: 10th International Conference on Future Networks and Communications, at Belfort.
  • MobiSPC 2015: 12th International Conference on Mobile Systems and Pervasive Computing, at Belfort.
  • AgentCities15: 2nd International Workshop on Agent-based Modeling and Simulation of Cities.




  • IngéDoc-2012: Conference of the Junior Researchers of UTBM, edition 2012.


  • IngéDoc-2011: Conference of the Junior Researchers of UTBM, edition 2011.


  • KARE08: First International Workshop on Knowledge Acquisition, Reuse and Evaluation.


  • ACW07: First International Workshop on Agent supported Cooperative Work.


  • AOSE-TF 06: Agent Oriented Software Engineering Technical Forum 2006.

Seminars Organized by CIAD


Date Hour Room Details
January 9th 14.00 Building Meeting Room, Belfort Hamza Jaffali, UTBM PhD student
Learning Algebraic Models of Quantum Entanglement
Quantum Information and Quantum Computing are two emerging areas of research, and are on their way to revolutionize our conception and implementation of computations. Recently, many efforts were deployed to unite Quantum Information, Quantum Computing and Machine Learning. They have largely been centered on integrating quantum algorithms and quantum information processing into machine learning architectures. Our approach is quite the opposite. We use Machine Learning techniques to study and classify Quantum Entanglement, a key ressource in Quantum Computing. In our work, we train Artificial Neural Networks to learn algebraic varieties, defined by polynomial equations, that characterize and describe different entanglement classes for pure states. Inspired by the work of Breiding et al., we focus on determining the membership of a state to an algebraic variety, instead of determining the defining intrinsic equations. By sampling tensors living inside and outside a given algebraic variety, we are able to train ReLU networks to classify such tensors. In the case of varieties defined by homogeneous polynomials, we also design and train hybrid polynomial networks.

We give examples for detecting separable states, degenerate states, as well as border rank classification for up to 5 qubits and 3 qutrits.

July 11th 14.00 A200, Belfort Dr. Nafaâ Jabeur from GUTech (Oman)
Embedded and Networked Intelligent Things.
Thanks to continuous progress on mobile and ubiquitous computing as well as the considerable advances on miniaturization, a myriad of devices are being deployed for a wide range of indoor and outdoor applications. These devices are, basically, equipped with communication mechanisms allowing them to share data and coordinate their actions with neighbouring peers. They are also being endowed with some intelligent mechanisms to pro-act and react to events of interest on-time. Due to heterogeneity matters, limited onboard capabilities as well as dynamic environments, these intelligent mechanisms remain difficult to implement. Solutions could, however, be inspired from the behaviour of small insects which are capable of acting selfishly or coordinating their efforts to jointly create complex behaviours. Following this trend, we are investigating options of embedding intelligent mechanisms in individual Real World Things (RWTs) to allow them perform the right actions at the right time for the right needs. The RWTs will ultimately interconnect to achieve goals that are beyond their individual limited capabilities. To meet our goals, we are exploring real-world problems falling within the large scope of Smart Cities, Cyber-Physical Systems, and Internet of Things. The presentation will be an opportunity to shed light on some of our past and ongoing related works.
Dr. Nafaâ Jabeur is Associate Professor (Computer Science Department) and Director of Research at the German University of Technology in Oman (GUtech). He received his Master degree and his PhD degree in Computer Science from Laval University (Quebec, Canada) in 2001 and 2006 respectively. He received his degree of Engineer in Computer Science from Morocco in 1998. He has over than 12 years of experience in the industrial and academic sectors. He is active researcher in a variety of fields, including Drones, Internet of Things, Smart Cities, Artificial Intelligence, Wireless Sensor Networks, Mobility and Transportation, Cartography, and Spatial Data Warehouses.

Nafaâ has participated in several R&D projects, authored/edited 2 books, and authored more than 80 research papers in prestigious conferences and high ranked journals. He is particularly collaborating with educational, industrial, and governmental organizations in several countries to design, implement, and introduce drone-based solutions in a variety of application fields, including surveillance, inspection, and logistics.
A part from his academic and research activities, Nafaâ is also entrepreneur. He is, indeed, the CEO of two startups; one in Tunisia for the development of IT-based solutions and one in Belgium for technology transfer. SlidesPDF

May 23th 14.00 A200, Belfort Dr. Bowei Chen from University of Glasgow (UK)
Probabilistic Modelling and Machine Learning for Online Advertising.
Online advertising has become a significant source of revenue for web-based businesses. In this talk, I will introduce several of our recent studies which develop probabilistic models and machine learning algorithms to improve the current advertising delivery system and sales model. In the first study, we propose an optimal dynamic model for online publishers which can unify programmatic guarantee and real-time bidding in display advertising. The model solves the problem of algorithmic pricing and allocation of the non-guaranteed page views into guaranteed contracts with stochastic demand. In the second study, we develop an average price advertising options which can provide flexible guaranteed delivery to advertisers. Jump-diffusion processes are used to model the evolution of the spot market prices from advertising slots. A general option pricing algorithm is obtained based on Monte Carlo simulation. In addition, an explicit pricing formula is derived for the case when the option pay-off is based on the geometric mean. This pricing formula is also a generalised version of several other option pricing models discussed in related studies. In the third study, we develop a novel computational framework which brings multimedia metrics like contextual relevance, visual saliency and advertisement memorability into real-time bidding. It aims to improve online user’s experience as well as maintain the benefits of online publisher and advertiser in display advertising.
Bowei Chen is an Assistant Professor at the Adam Smith Business School of University of Glasgow. He received a PhD in Computer Science from University College London and works in the cross sections among data science, machine learning and business studies, with special focus on digital marketing and quantitative finance. He was an Assistant Professor in the School of Computer Science at University of Lincoln, and has held visiting scholar and professional positions at Université de Technologie Belfort-Montbéliard, Copenhagen Business School, Aarhus University, University of Bath, and National University of Singapore.
February 14th 14.00 Meeting Room, Building D, Belfort Dr. Jean-Michel Ilié from Sorbonne Université
E-HoA – Vers un agent intentionnel à planification contextuelle pour robot roulant autonome.
Les agents HoA (higher Order Agent sont des agents intentionnels intelligents primitivement destinés à la guidance des usagers dans des espaces d’activité définis, comme des aéroports ou des campus intelligents (smart). Le projet présenté consiste à étendre les capacités de tels agents en proposant l’architecture E-HoA (Embedded HoA). Nous visons à construire des entités hautement concurrentes capables de repousser l’intervention humaine, s’agissant de remplir des missions dans un espace routier. L’entité est ici un robot roulant fonctionnant sous ROS, qui offre à la fois des capacités robotiques et de simulation.
Dans ma présentation, je rappellerai les atouts et mécanismes des agents HoA qui par leurs aspects formels (algébriques) et semi-formels (via l’apprentissage), offrent puissance de description, planification contextuelle, adaptation au contexte, et réactivité aux aléas dans un environnement changeant. Je présenterai aussi les principes de l’architecture E-HoA destinée à maintenir la consistance entre les intentions de l’agent au regard des actions concrètement réalisées. Pour des questions de performances et d’indépendance modulaire, cette architecture est développée suivant des paradigmes de systèmes distribués.
Jean-Michel Ilié est Maître de conférences au grade de la classe exceptionnelle de son Université à Paris et exerce ses activités de recherche au laboratoire LiP6. Spécialiste de l’analyse des comportements dans les systèmes complexes, il a tout d’abord élaborer des mécanismes complémentaires destinés à réduire les effets d’explosion combinatoire inhérents aux techniques de vérification formelle et d’évaluation de performances des systèmes. Depuis 10 ans, il améliore le développement des agents intentionnels évoluant dans des environnements ambiants en intégrant des mécanismes de planifications semi-formels renforcés par des techniques d’apprentissage. Soucieux de faire partager son savoir, il enseigne les innovations technologiques et a monté en 2018 pour l’IUT de PARIS, une des toutes premières licences pro des métiers de l’informatique ayant la spécialisation IoT.
February 1st 9.30 Meeting Room, Building D, Belfort Dr. Serge Iovleff from university of Lille
Co-Clustering of Binary Data with Gaussian Co-variables
The simultaneous grouping of rows and columns is an important technique that is increasingly used in large-scale data analysis. In this talk, I present a novel co-clustering method using co-variables in its construction. It is based on a latent block model taking into account the problem of grouping variables and clustering individuals by integrating information given by sets of co-variables. Numerical experiments on simulated data sets and an application on real genetic data highlight the interest of this approach.
Slides Part 1PDF, Slides Part 2PDF


Date Hour Room Details
November 28th 11.00 M208, Montbéliard Pr. H. Aoki from university of Nagoya (Japan)
Autonomous driving technologies and applications
November 13th 16.15 A206, Belfort Dr. Tomas Krajnik
FreMEn: Frequency Map Enhancement for Long-Term Autonomy of Mobile Robots
While robotic mapping of static environments has been widely studied, life-long mapping in non-stationary environments is still an open problem. We present an approach for long-term representation of natural environments, where many of the observed changes are caused by pseudo-periodic factors, such as seasonal variations, or humans performing their daily chores. Rather than using a fixed probability value, our method models the uncertainty of the elementary environment states by their frequency spectra. This allows to integrate sparse and irregular observations obtained during long-term deployments of mobile robots into memory-efficient models that reflect the recurring patterns of activity in the environment. The frequency-enhanced spatio-temporal models allow to predict the future environment states, which improves the efficiency of mobile robot operation in changing environments. In a series of experiments performed over periods of weeks to years, we demonstrate that the proposed approach improves mobile robot localization, path and task planning, activity recognition and allows for life-long spatio-temporal exploration.
September 27th 14.30 Meeting Room, Building D, Belfort Pr. Nidal Kamel
SVD-Based Tensor-Completion Technique for Background Initialization
Extracting the background from a video in the presence of various moving patterns is the focus of several background-initialization approaches. To model the scene background using rank-one matrices, this paper proposes a background-initialization technique that relies on the singular value decomposition (SVD) of spatiotemporally extracted slices from the video tensor. The proposed method is referred to as spatiotemporal slice-based SVD (SS-SVD). To determine the SVD components that best model the background, a depth analysis of the computation of the left/right singular vectors and singular values is performed, and the relationship with tensor-tube fibers is determined. The analysis proves that a rank-1 matrix extracted from the first left and right singular vectors and singular value represents an efficient model of the scene background. The performance of the proposed SS-SVD method is evaluated using 93 complex video sequences of different challenges, and the method is compared with state-of-the- art tensor/matrix completion-based methods, statistical-based methods, search-based methods, and labeling-based methods. The results not only show better performance over most of the tested challenges, but also demonstrate the capability of the proposed technique to solve the background-initialization problem in a less computational time and with fewer frames.
Pr. Nidal Kamel received the M.Sc and PhD degree (Hons) from the Technical University of Gdansk, Poland, in 1993. His PhD work was focused on subspace-based array signal processing for direction of-arrival estimation. Since 1993 he has been involved in research projects related to estimation theory, noise reduction, optimal filtering, and pattern recognition. He developed SNR estimator for antenna diversity combining, single-trial subspace-based technique for EEG extraction form brain background noise, and introduced a subspace-based data glove system for online signature verification. His present research interest is in brain signal processing, image enhancement, and pattern recognition. Currently, he is Associate Professor at the PETRONAS University of Technology. He is IEEE senior member.
June 21th 14.00 Auditorium A200, Belfort Pr. Xiaowei Tu
Capteurs, Systèmes et Equipements intelligents pour véhicules autonomes ou pilotage assistées
Après une présentation de notre équipe de travail, les travaux de recherche appliquée au sein de notre laboratoire sur les systèmes de contrôle et de mesure de l’Institut de Mécatronique et d’Automatisation de l’Université de Shanghai. Notre recherche concerneles quatre axes principaux suivants : la conception de capteurs intelligents, la conception modulaire de contrôleurs de véhicule, la réalisation des équipements combinant les deux premiers et de la simulation visuelle en réalité virtuelle ou augmentée. Sur chacun de ces thèmes, plusieurs exemples seront présentés afin d’illustrer les applications industrielles visées. L’ensemble de ces travaux a fait l’objet de dépôts de brevets, de publications scientifiques et de contrats industriels. L’un de nos objectifs et de créer un environnement de travail et des plate-formes expérimentales permettant la synergie des travaux avec des professionnels et des chercheurs de part le monde.
May 31th 10.30 Meeting Room, Building D, Belfort Dr. You Li
LiDAR and Its Trend in Automotive Application
In this talk, Dr. Li will present his work at Renault Innovation, including principle of LiDAR, different types of LiDAR, usages for autonomous driving (e.g. object description and localization), and especially, challenges for automotive usage and its trend in the future.
May 3rd 14.00 Meeting Room, Building D, Belfort Hui ZHAO
Agent-based Dynamic Rescheduling of Daily Activities
When simulating individuals’ daily plan, in order to determine the effect on the road network, lots of unexpected events need to be considered, like traffic jam and weather changes. Therefore, there will be a mismatch between the original plan and the executed one. Faced with this situation, individuals need to adjust the rest of the activities to make a new schedule. This paper analyzes the causes of rescheduling, and establishes a new rescheduling model, combining strengths of existing rescheduling models. The model in this paper considers the rescheduling possibilities and choices as much as possible. It takes time pressure and schedule similarity into consideration when updating a schedule. Furthermore, this paper analyzes joint trip/activity execution by studying the cooperation between agents during the rescheduling process.
This talk will be given to the FAMS18 conference.
April 16th 9:30 Meeting Room, Building D, Belfort Dr. Li Sun – Lincoln Centre for Autonomous Systems (L-CAS), University of Lincoln, UK
Computer vision and machine learning for robotics applications
In this talk, Dr. Li will present his work for three European projects. Dr. Li Sun’s PhD was working for EU FP7 CloPeMa project (, aiming to achieve autonomous robot laundering. In this project, he was working on visually-guided dual-arm manipulation of deformable clothes. He was the main contributor of two robot demonstrations, i.e. dual-arm clothes on-table flattening, autonomous clothes categorization, and interactive sorting. In his research, a generic computer vision architecture using stereo robot head is proposed for multiple-laundry tasks including grasping, recognition, flattening, and sorting. Latter, he was working for EU H2020 RoMaNs project ( and then EU H2020 ILIAD project (, focusing on deep learning based table-top object detection and long-term semantic mapping of the indoor and outdoor scene.
March 29th 14.00 Meeting Room, Building D, Belfort Yazan MUALLA
Comparison of Agent-based Simulation Frameworks for Unmanned Aerial Transportation Applications
Recently, the applications of Unmanned Aerial Vehicles (UAVs) in aerial transportation are gaining more interest. Due to operational costs, safety concerns and legal regulations, agent-based simulation frameworks are preferably used to implement models and conduct tests. With the abundance of such frameworks, this paper introduces a methodology to compare the most widely used frameworks. The methodology is based on the ISO software quality model, and uses a weighted sum scoring system to give points to the frameworks under study. The proposed criteria in the methodology consider agent-based simulation features and adapt specific features of unmanned aerial transportation. Preliminary comparison results and recommendations are provided

and discussed.
This talk will be given to the ABMTRANS18 conference.


Date Hour Room Details
September 28th 14.00 Meeting Room, Building D, Belfort Fabrice LAURI
Deep Reinforcement Learning
In 2013, a small company in London called DeepMind uploaded their pioneering paper “Playing Atari with Deep Reinforcement Learning” to Arxiv. In this paper they demonstrated how a computer learned to play Atari 2600 video games by observing just the screen pixels and receiving a reward when the game score increased. The result was remarkable, because the same model architecture, without any change, was used to learn how to win in seven different games, and in three of them the algorithm performed even better than a human!

No wonder DeepMind was immediately bought by Google and has been on the forefront of deep learning research ever since.
It has been hailed since then as the first step towards general artificial intelligence – an AI that can survive in a variety of environments.
The roadmap of this seminar is:
– What are the main challenges in reinforcement learning?
– How to formalize reinforcement learning in mathematical terms?
– How do we form long-term strategies?
– How can we estimate or approximate the future reward?
– What if our state space is too big? (Here the answer is simple: deep learning!)
– What are the main deep RL algorithms?
– What performances can these algorithms obtain on classical problems?
– What are the main issues of applying such algorithms?
– How to program such algorithms and validate them on common problems with Ipseity?

September 26th 10.00 Meeting Room, Building D, Belfort Ezzine Missaoui
Intégration des Normes dans les Systèmes Multi-agents Holoniques
Les systèmes multi-agents sont composés d’agents logiciels autonomes et hétérogènes qui agissent selon leurs propres intérêts. Certain mécanisme de coordination doit être adapté pour assurer le bon fonctionnement de l’ensemble du système. Les normes peuvent être considérées comme un moyen puissant de réguler et d’influencer le comportement des agents en spécifiant par exemple des obligations, des autorisations et des interdictions dans un contexte donné. Plusieurs propositions sur les modèles normatifs pour les systèmes multi-agents ont été réalisées afin de concevoir des sociétés d’agents dans des environnements régis par des normes. Cependant, ils ne sont pas adaptés pour supporter les systèmes holoniques, c’est-à-dire ceux qui permettent de modéliser des organisations complexes impliquant plusieurs niveaux simultanément. L’un des principaux défis actuellement rencontrés dans la recherche de SMAH est celui du contrôle social. Cependant, un SMAH ne couvre pas certains aspects tels que ceux liés aux normes à respecter pour assurer la rentabilité des objectifs du système.
September 21th 14.00 Meeting Room, Building D, Belfort Dr. Benjamin Camus
DEVS wrapping of the FMI standard for the co-simulation of Cyber-Physical Systems in MECSYCO
Most modeling and simulation (M&S) questions about cyber-physical systems (CPS) require expert skills belonging to different scientific fields. The challenges are then to integrate each domain tool (formalism and simulation software) within the rigorous framework of M&S process. To answer this issue, we proposed the specifications of the MECSYCO co-simulation middleware. MECSYCO relies on the universality of the DEVS formalism to integrate models written in different formalisms. This integration is based on a wrapping strategy in order to make models implemented with different simulation software interoperable. So far, we successfully defined DEVS wrappers for discrete modeling tools like the MAS simulator NetLogo, and the IP network simulators NS-3 and OMNeT++/INET. Aside from several difficulties met at the software level, making these discrete modeling tools compliant with the DEVS simulation protocol was a straightforward process. This is due to the fact that these platforms have a discrete modeling paradigm very close to DEVS. However, things getting more complex with equation-based tools as their continuous modeling paradigm is very different from the discrete DEVS one. Thus, we need to bridge the gap between the discrete and the continuous paradigms. A more complex wrapping strategy based on the hybrid capacity of DEVS is required. Regarding this issue, wrapping each of these equation-based tools (e.g. OpenModelica, Dymola, Matlab/Simulink) separately would be very inefficient. In this talk, I will detail how we tackle this issue by defining DEVS wrappers for the FMI standard which brings a generic API to manipulate equation-based models and their solvers. We perform this wrapping using the DEV&DESS hybrid formalism and the QSS numerical method. The DEVS wrapping of FMI we propose is not restricted to MECSYCO but can be performed in any DEVS-based platform.
June 29th 14.00 Belfort Pr. Ansar Yasar
Empowering Citizens with Sustainable Transportation in the Cities of Today & Tomorrow
While some may argue that the added value of one research domain is more limited in terms of added economic value than the other, the contribution of transportation research towards the society as a whole is significant. According to several predictions, the transport sector will overtake industry as the largest energy user by 2020. Unfortunately, the sector has major negative economic, social and environmental side effects. The complexity of today’s policy decision making has motivated several international research teams to develop policy frameworks which are finally aimed at mitigating these negative externalities of transport.

In several international policy frameworks, conventional transport models have been used for the quantification of these externalities. When an operational model is required to provide quantitative predictions about human behaviour, some kind of mathematical apparatus is adopted in models. In this talk, we will cover the research domain of activity-based models. In these models, using micro-simulation, full activity-travel patterns of people are predicted in a high resolution of time and space, offering a wealth of information for policy making. The models give us a behavioural insight at an unprecedented level and allow for many interesting interdisciplinary applications. In this talk, I will give a brief overview of the state-of-the-art in activity-based modelling and discuss the interesting developments in this field of research. In addition to applications in the domain of transportation research, I will focus on scientific interdisciplinary applications with several other scientific fields, such as emission and health impact calculations, traffic safety and future electric vehicle (market) projections. Also the talk will cover novel interesting trends in the research field such as the increasing availability of big data and the development of modern survey technology, which offers several opportunities for policy makers but also provides researchers with novel challenges and problems.

June 1th 14.00 Meeting Room, Building D, Belfort Dr. Vukosi Marivate
Machine Learning, Reinforcement Learning
June 1th 10.30 Meeting Room, Building D, Belfort Dr. Nidal Kamel
Subspace-Based Estimators for Image Denoising
Digital images are susceptible to various types of noise that may which affects their quality. In the field of image enhancement, different approaches for noise reduction have been proposed. In general, there are two basic approaches to image denoising, spatial filtering methods and transform domain filtering methods. Spatial filtering methods include linear methods like the mean and Winer and nonlinear methods, like the median and the weighted median. The performance of these filters is highly dependent on the choice of size and orientation of the moving window. Transform domain filtering methods are mostly dominated by the Wavelet, where the image is first transformed into the wavelet domain then a thresholding scheme is applied. The major drawback of the wavelet-based technique is the ringing impairments due to the thresholding process. Recently, the area of the subspace based filters, has gained widespread attention and successfully implemented in various areas of image densoing. In this lecture, two subspace-based techniques to reduce the noise in images are outlined. These techniques are the Least Squares Estimator, and the Time Domain Constraints Estimator (TDC).
May 4th 14.00 Meeting Room, Building D, Belfort Pr. Vincent Chevrier
Mecsyco: Multi-agent Environment for the Co-simulation of COmplex systems
Most modeling and simulation (M&S) questions about complex systems require to take simultaneously account of several points of view. Phenomena evolving at different scales and at different levels of resolution have to be considered. Moreover, expert skills belonging to different scientific fields are needed. The challenges are then to reconcile these heterogeneous points of view, and to integrate each domain tools (formalisms and simulation software) within the rigorous framework of the M&S process.

This talk will present the mecsyco co-simulation middleware ( Mecsyco r elies on the universality of the DEVS formalism to integrate models written in different formalism. This integration is based on a wrapping strategy in order to make models implemented in different simulation software inter-operable. The middleware performs the co-simulation in a parallel, decentralized and distributable fashion thanks to its modular multi-agent architecture.

March 16th 14.00 Meeting Room, Building D, Belfort Stéphane GALLAND
SARL – Agent-oriented Programming Language

SARL is a general-purpose agent-oriented language. It aims at providing the fundamental abstractions for dealing with concurrency, distribution, interaction, decentralization, reactivity, autonomy and dynamic reconfiguration. These high-level features are now considered as the major requirements for an easy and practical implementation of modern complex software applications, and specifically agent-oriented programming.
This talk will introduce you to the advances of the SARL agent programming languages, and provides several examples of usages.



Date Hour Details
February 6th 9.30 Sebastian Rodriguez, Nicolas GAUD
SARL – Agent-oriented Programming Language


Date Hour Details
November 14th 14.00 Vincent HILAIRE
OpenModelica, un langage ad-hoc pour la simulation
September 27th 09.00-18.00 Agent Group Workshop
June 14th 10.00 Luk Knapen (IMOB, Hasselt, Belgium)
Carpooling Model.
March 10th 10.00 Olivier Boissier (EMSE, Saint-Etienne, France)
Multiagent-oriented programming


Date Hour Details
January 31th 14.30 Michael Schumacher (Applied Intelligent Systems Laboratory, Sierre, Switzerland)
Medical and Smart-grid applications at AIS Lab.


Date Hour Details
February 15th 14.00 Gildas Morvan (LGI2A Laboratory, Béthune, France)
Multilevel simulation and its applications.