Crowd simulation is an important cornerstone for effective urban planning and emergency management. Crowd simulation models vary widely in the complexity that is given to each individual in the crowd. The simplest models treat individuals as the equivalent of physical particles, the most complex ones emulate cognitive processes to make the behaviour realistic. Complex does not always mean better, since simplified models often help to uncover fundamental principles by abstracting away from irrelevant detail. In this project we will explore an intermediate level of modelling the individual’s decision making during evacuation scenarios based on the principles of game theory.