A state-of-the-art review of multi-agent modelling of crowd dynamic

Abstract

This paper reviews the multi-agent modelling (MAM) which applied as a platform to simulate the interaction of crowd dynamics. The research and development of MAM related to crowd dynamics can generate specific analytical evaluation, improve the safety level of the pedestrians and the quality operation of the urban space. Furthermore, crowd dynamics have become a major concern for urban space and are mainly constrained imposed by the ever-increasing populations. Consequently, these scenarios causing more attention are paid to the crowd safety in capacity assessment, especially during an emergency evacuation. This paper presents a concise review on the development of MAM over the past few years. The discussion on the state of the art of the multi-agent methods includes the history of the multi-agent method and the development of multi-agent modelling. This paper also discussed the state of the art of multi-agent methods in modelling crowd dynamic by focusing only on the microscopic approach. The advantages of multi-agent modelling approach have become one of the best methodological approaches in analysing and predicting crowd performance in the real world as a substitute for physical experimentation in a current research field.