Crowd Entropy-Based Prediction Model: Unidirectional Flow

The prediction of critical conditions during emergency evacuation is a key safety factor in crowd movement and urban transport management. Previous entropy-based models developed for evaluating crowd risk primarily relied on either density or velocity attributes, which alone are insufficient for danger prediction. Therefore, this paper proposes a crowd Boltzmann entropy-based prediction model that integrates the local density with average local speed to identify the critical situations that may result in crowd disasters, empowering sufficient prediction of crowd critical conditions and accurate description of the nature of a crowd motion, therefore enabling early preventative intervention. This proposed method is embedded into a neuro-symbolic evacuation model that has human-level capabilities of reasoning and performance to ensure realistic prediction on the nature of crowd motion. The results reveal that the longest critical conditions last in the nearest area to exit with the highest average entropy of 0.44, where the shortest is recorded in the farthest area to exit with the lowest average entropy of 0.29. This is consistent with previous literature on the crowd dynamics. Finally, the work demonstrates more universal and consistent descriptions of crowd psychology and motions and outperforms well-established prediction approaches such as crowd pressure and flow.