Testing Various Riot Control Police Formations through Agent-Based Modeling and Simulation

A crowd may gather for various reasons such as to protest in front of government buildings or to cheer on their teams at sporting events. This crowd, influenced by a number of conditions can become aggravated and then violent, has the potential to change dynamics and create public disorder with possible outcomes of people being hurt or damage to public or private properties. Managing and controlling such a disruptive crowd is an important responsibility for police to maintain and restore the public order and safety. Effective crowd management and control strategies need to be considered and planned in advance to mitigate potential disruption and damages caused by a riotous crowd. One of the basic strategies is to utilise police tactics and to strategically deploy police teams in formations. The correct application of police formations can positively influence and be effective in dispersing an agitated crowd. This paper presents the effectiveness of various police formations in a riotous situation using agent-based modeling and simulation. The crowd management/control system that we have developed previously has the built-in crowd models and basic features to conduct simulation experiments. Using this system, the loose cordon and three-sided box formations are simulated to expulse a riotous crowd from either the four-way interaction or “T” intersection, and their results are compared and analyzed. The simulation system can be used to further research on the effect of other police formations in various situations, which potentially helps police prepare better crowd management/control strategies.