|
|
|
Keynote Speakers
Evolutionary Simulation Optimisation
|
Professor Juergen Branke |
Tel: 
Email:
Abstract
Simulation is a powerful tool to understand and analyze complex systems. With the increase in available computational power, it now becomes possible to move a step further and consider simulation-based optimization. In simulation-based optimization, the goal is to identify the best out of a potentially large space of alternatives, where the quality of a solution is determined by running a simulation model.
However, there are a number of challenges that differentiate simulation-based optimization from most other application areas. First of all, it restricts the choice of tools to black box optimizers, as additional information such as gradients is not available from a simulation model. Second, running a simulation model is usually computationally expensive, and severely limits the number of solutions that can be evaluated during optimization. Finally, simulations are often stochastic, thus returning a stochastic quality evaluation while traditional optimisation usually assumes a deterministic evaluation.
Metaheuristics are perhaps the most promising approach to simulation-based optimization, and many simulation tools nowadays come with an optimization tool based on metaheuristics such as evolutionary algorithms or simulated annealing. This talk discusses various ways to adapt metaheuristics to address the particular challenges of simulation-based optimisation.
Curriculum Vitae
Juergen Branke is Professor of Operational Research and Systems and Head of Operational Research at Warwick Business School, University of Warwick, UK. He received his PhD from the University of Karlsruhe, Germany, in 2000 and has been an active researcher in the area of nature-inspired and heuristic optimization since 1994. He has published over 130 articles in international journals and conferences on various topics including multiobjective optimization, handling of uncertainty in optimization, dynamically changing optimization problems, and the design of complex systems. He worked on a wide range of applications ranging from scheduling, transportation or traffic light control to engineering design. Professor Branke is Associate Editor of the Evolutionary Computation Journal and Area Editor of the Journal of Heuristics.
Virtual Geometry Assurance for High Quality and Reduced Lead Time
 |
Professor Rikard Söderberg |
Abstract
All manufacturing and assembly processes are afflicted with variation which causes problems during production and in the final product. If assembly requirements are not met (parts do not fit together) this causes scrap or rework. If function requirements are not met (the product does not function properly) customers complain and request new products. If aesthetical requirements are not met (the product does not look good) the product is perceived as a low quality product. All these problems generate costs, delays and loss of customers.
With modern simulation technology these problems can be detected and avoided already in very early design stages, before any physical parts or products have been manufactured. By combining 3D visualization, Monte Carlo simulation, FEA analysis with sensitivity analysis and optimization, visually and geometrically robust solutions can be found and proper tolerances can be allocated. Thereby, costly late changes and fixes are avoided.
RD&T (Robust Design & Tolerancing) is a simulation tool for Virtual Geometry Assurance that was developed in a research project at Chalmers, got commercialized in 1998 and is today used in the daily operation by a number of global companies like Volvo Cars, Volvo Trucks, Jaguar/LandRover, Ford, GKN and Scania.
Presentation on the the 12th of June 2014
Curriculum Vitae
to be added
Modelica - A Cyber-Physical Modeling Language for Systems Engineering and the OpenModelica Environment
|
Prof. Peter Fritzson |
Fax: +46 13 285899
Abstract
Modelica is a modern, strongly typed, declarative, equation-based, and object-oriented (EOO) language for modeling and simulation of complex cyber-physical systems. Major features are: ease of use, visual design of models with combination of lego-like predefined model building blocks, ability to define model libraries with reusable components, support for modeling and simulation of complex applications involving parts from several application domains, and many more useful facilities. This talk gives an overview of some aspects of the Modelica language, and the OpenModelica environment including the ModelicaML UML/Modelica profile for systems engineering - the most complete Modelica open-source tool for modeling, engineering, simulation, and development of system applications (www.openmodelica.org).
Special features are MetaModeling for efficient model transformations, the ModelicaML profile for UML-Modelica cyber-physical hardware-software modeling, the SysMLModelica OMG standard proposal, the Functional Mockup Interface for general tool integration, model-based optimization, as well as generation of parallel code for multi-core architectures
Curriculum Vitae
Peter Fritzson is Professor and research director of the Programming Environment Laboratory, at Linköping University. He is also director of the Open Source Modelica Consortium, director of the MODPROD center for model-based product development, and vice chairman of the Modelica Association, organizations he helped to establish. During 1999-2007 he served as chairman of the Scandinavian Simulation Society, and secretary of the European simulation organization, EuroSim.
Prof. Fritzson's current research interests are in software technology, especially programming languages, tools and environments; parallel and multi-core computing; compilers and compiler generators, high level specification and modeling languages with special emphasis on tools for object-oriented modeling and simulation where he is one of the main contributors and founders of the Modelica language. Professor Fritzson has authored or co-authored more than 210 technical publications, including 16 books/proceedings.
