State-space models reveal bursty movement behaviour of dance event visitors

Abstract Pedestrian movements during large crowded events naturally consist of different modes movement behaviour. Despite its importance for understanding crowd dynamics, intermittent behaviour is an aspect missing in the existing literature. Here we analyse data generated from nearly 600 Wi-Fi sensors entertainment Johan Cruijff ArenA football stadium Amsterdam. We use state-space modeling framework to investigate motion patterns. Movement models field ecology are used individual pedestrian movement. Joint estimation multiple tracks allows us statistical properties measured metrics. show that behavioural switching not independent external events, and probability being one states changes over time. In addition, distribution waiting times deviates exponential best fit by a heavy-tailed distribution. The indicative bursty which here first time shown characterise dense crowds. Bursty has important implications various diffusion-related processes, such as spreading infectious diseases.