Evaluating Jason for Distributed Crowd Simulations

Large-scale crowd simulations require distributed computer architectures and efficient parallel techniques to achieve the rendering of visually plausible images while simulating the behaviour of crowds of autonomous agents. The Java-based multiagent platforms, devoted to provide the agents with the required lifecycle, represent a key middleware in crowd systems. However, since they are oriented to maximize portability and to reduce the development cost, they may reduce performance and scalability, two important requirements in large-scale crowd simulation systems. This paper studies the performance and scalability provided by Jason, a well known Java-based BDI-MAS platform, as a plausible framework to be used for large-scale crowd simulations. The performance evaluation results show that some improvements should be performed in order to make Jason a suitable middleware for large-scale crowd simulations.