Vol. 7

This work presents a multi-agent simulation which models the activities of tourists. The goal is to have these agents populate a virtual world, where they are able to evaluate different development scenarios. At the same time, the project is used to explore general computational implementations of mobility simulations.
Our approach is to model each tourist individually as an agent. A synthetic population of tourists is created that reflects current visitor demographics. These tourists are given individual goals and expectations, and are introduced into the simulation with no knowledge of the the environment. They execute their plans, receiving feedback from the environment as they move throughout the landscape. At the end of each run, the agents' actions are compared to their expectations.
This work has a strong computational component. Besides the presentation of the framework and its modules, computational aspects of the pedestrian simulation are discussed. Using sophisticated techniques which adapt the model to the rather special circumstances of hikers in the alps, it is now possible to simulate an area of 15km^2 and more on a single CPU.

Christian Gloor studied Computer Science at the Swiss Federal Institute of Technology Zurich. Within the project "Planning with Virtual Alpine Landscapes and Autonomous Agents", he was responsible for the technical implementation.

Keywords: Simulation, Multi-Agent Systems, Distributed Computing