Research Project: CoSA


   
Component based framework for an effective and efficient simulation of (agent) systems


Abstract:

The goal of this project is to develop a modeling and simulation framework which is sufficiently flexible and modular to support research on modeling and simulation methods, the design of domain dependent simulation tools, applications in diverse areas, and modeling and simulation of complex, heterogenous (agent-based) systems.

Keywords:

Modelling, Model components, Simulation, Framework, Open architecture, Multiple formalisms, Multi-agent, Experiment design, Validation, Testing, workflows, Quality, Plug-ins, Efficient computations, Algorithm selection.

   
Research Project Information


   Runtime:    Apr 01, 2003 until Dec 31, 2013
  
   Project coordination:    Prof. Dr. rer. nat. habil. Adelinde M. Uhrmacher
  
   Scientific staff:    Dipl.-Inf. Alexander Steiniger;
   M.Sc. Tobias Helms
  
   Former staff:    Dr.-Ing. Jan Himmelspach; Dr.-Ing. Roland Ewald;
   Dr.-Ing. Stefan Leye; ;Dipl. Inf. Stefan Rybacki;
   Dipl. Inf. Mario Krahmer; Dipl. Inf. Susanne Biermann;
   Dr.-Ing. Mathias R√∂hl; Dr.-Ing. Matthias Jeschke;
   Dr.-Ing. Mathias John
  
   Former student staff:    Johannes R√∂ssel; Valerius Weigandt; Mathias Rimane;
   Enrico Seib; Dortje L√∂per; Ulrike Borchardt;
   Robert Waltemath; Simon Bartels; Johannes Becherer;
   Thomas Beer; Gabriel Blum; Nico Eggert;
   Roland Ewald; Stefan Friedrichs; Florian Gewandt;
   Lydia Jost; Sven Kluge; Stefan Leye; Sebastian Lieske;
   Frank Manteufel; Steffen Maas; Rene Michalski;
   Thomas N√∂singer; Christian Ober; Bj√∂rn Paul;
   Jan Pommerenke; Kathrin Rohloff; Oliver R√∂wer;
   Hans Schipke; Rene Schulz; Carl Tuemmler;
   Felix Willud; Felix Woitzel; Martina Gierke
  
   Contractor(s):    DFG (DFG GEPRIS Link)

 

Multi agents

Agent based systems are software systems which operate in open and dynamic environments. In regards to the analysis of their functionality and performance, simulations are claiming a more important role. Thereby agent characteristic properties (flexibility, autonomy, resource intensity and mobility) enforces special simulation methods adressing them. Based upon previous research in the field of agent oriented simulation and actual developments in modeling and simulaton a new framework shall be developed being built of reusable components which offers its users the ability for efficient and effective experimentation with multi agents systems in a virtual dynamic environment. For an effective execution of such simulation studies a flexible modeling and simulation software is needed. Thereby flexibility means the potentialities for designing, controlling and interpreting experiments. Efficiency of a simulation framework is multi facetted. It does not only allude the topic of execution mechansims (e.g. parallel - distributed vs. sequential) - it depends also on the ease of use. The area of conflict between effectivity and efficiency shall be explored by using the emerging framework and by selected multi agent simlation studies.

 

Modelling and Simulation in general

Albeit the existence of specialized methods as those mentioned in the section above, many (if not most) modelling and simulation methods are not restricted to a certain modeling paradigm nor application. As the goal of this project was the development of a general framework for modeling and simulation of multi agent systems (MAS) it was a relatively small step to get rid of the hard MAS specialization. The resulting concepts can be used without the notion of agents being around, for all modelling and simulation application areas and fields of research, including multi agent systems. The clue about the concepts developed is that the integration of more specialized methods is the primary idea behind the concept of the general framework developed - and thus the primary project is still "solved".

   
Publications