Research On The Early Warning Of Crowded Under Large-area Flight Delays

The statistical results of Civil Aviation Authority show that the number of large-area flight delays in China increased year by year. The terminal will gather a large number of passengers under the condition of large-area flight delays. Once the crowded events happen, it will threaten the safety of passengers. Through the research of early warning, the occurrence of crowed events can be prevented previously.On the base of accident causal chain theory, the generating mechanism of crowded events in terminal is concluded through the analysis of mismanagement, passenger psychological factors, passenger unsafe behaviors and incentive events. According to the flow and density, the state of crowd is divided into free flow, maximum flow, intermittent flow,crowded flow. Through the analysis of the self-organization phenomenon and the dissipative structure characteristics of crowd, it proves the existence of entropy in crowd. Using the assumption of local equilibrium, the local pressure and the flow are considered as generalized forces and generalized stream respectively, the model of entropy production and state entropy are established. Finally, the system entropy that contains of different states of crowd is created.According to the area of different states of crowd in theory and practice, the system entropy of theory(S-theoretical) and practice(S-practical) are concluded. Based on the monitoring data, the system entropy production is calculated. According to the value of S-theoretical and p, the crowed warning is divided into three levels, including red warning, orange warning and yellow warning. The warning level is determined by the comparison results of S-theoretical and S-practical.The system entropy model not only predicts the location and time of crowed events, but also estimates the change direction and speed of the system state.