The analysis of crowd behaviour is among the best known applications of complexity science to social systems. Individuals in a crowd influence each other through local, often physical (e.g. pushing) interaction. Simple (yet often very effective) approaches model such influence as 'forces' and use physical analogies to analyse global behaviour. More complex research includes differences between individuals and different external influences. Many interesting phenomena such as the emergence of structure and phase transitions can be observed in crowds. Practical relevance in found in civil protection applications such as emergency management or crowd control (e.g. during mass events such as concerts or football games).

We assume that large portions of city infrastructure are interconnected, and equipped with different types of sensors (cameras, motion sensors, internal functionality sensors etc.). We also assume most people in the city to carry mobile phones. In accordance with the smart environments vision, the mobile phones will be more than just phones. They shall also be autonomous, internet connected devices (which many current phones already are) capable of gathering and transmitting sensor data about user context (location, motion, possibly emotional state), accessing remote information on the network, and communicating with each other and with parts of the environment in an ad-hoc peer-2-peer manner. There are a number of relevant variants of this scenario that involve the modelling of complex socio-technical systems. The simplest is a crowd control scenario where the collective crowd behaviour is influenced by the ICT system. Peer-2-peer interactions between the mobile devices can diffuse information about escape routes, behaviour of users and user groups (where is panic starting), recommend to the user’s what should be done and try to calm people down. Information can also be passed to and from infrastructure. The interesting question is how the inclusion of ICT changes system behaviour compared to conventional models and how the ICT system evolves.