Figure 1.
Room evacuation simulation in a panic situation with only one exit.
a, b, c and d is the escape snapshot after 4, 14, 22 and 28 sec, respectively.
Figure 2.
Optimized obstacle setting for 1 to 3 obstacles and 80 participants.
R is the obstacle radius. The blue circle dots represent the obstacles. The GA-optimized obstacle setting results indicated that the obstacles should be near the two sides of the gate.
Table 1.
GA-Optimized obstacle settings for 80 participants.
Figure 3.
The escape speed of simulations for various obstacle configurations for 80 participants.
As a baseline, when there is no obstacle in the room, the average escape speed of the crowd determined using over 100 simulations is 2.64 persons/sec, i.e., 2.64 individuals can leave the room every sec. When there are one, two or three obstacle(s) and the radius of the obstacles are 0.5 m, 0.75 m or 1 m, the average escape speed is listed on the top of the corresponding bar diagrams.
Figure 4.
Schematic defining the tangential force.
is the component force perpendicular to the wall of the gate.
is the component force parallel to the wall. The integral of
over time t is the tangential momentum.
here in the figure is the same as
in equation (1).
Figure 5.
The tangential momentum vs. the escape speed for the simulation results, which comprise nine combinations of parameters of one, two or three obstacles and obstacle radius values of 0.5 m, 0.75 m or 1 m.
The straight line is a linear fit to the data. We analyzed the points with an escape speed greater than 2.64 persons/sec. The reason is that the escape speed for the zero obstacle case is 2.64 persons/sec and the obstacle position configurations that reduce this escape speed are useless and not worth being analyzed. A decreasing trend for the tangential momentum, especially for high velocity values, is easily observed.
Figure 6.
The escape speed and tangential momentum of different obstacle settings.
The most efficient settings are at the side of the door but not in front of the door and the tangential momentum and the escape speed are negatively correlated.
Figure 7.
The escape time vs. the number of escaped participants for human experiments.
Panels a, b and c are for zero obstacle, one obstacle and two obstacles, respectively. The data of the first run (black dots) is above the other two runs in panel a and c, which means the former participants escape slower than the latter ones. In panel b, however, the data of the first run lies between those of the second run and the third run, indicating that the escape speed of the first run is faster than that of the second run and slower than that of the third run. Therefore, the escape time has no correlation with experiments order.
Table 2.
Average escape speed of human experiments and relevant statistical analysis.
Figure 8.
The average escape time vs. the different numbers of obstacles.
The average escape time is the average time of three runs in each of the three obstacle configurations.
Figure 9.
The graph showing the simulation and experimental results.
The average escape speed for zero, one and two obstacles with the optimized positions is shown for the experiment (left) and the simulation (right). The chosen simulation data are the corresponding results when the obstacle radius is 0.5 m. The simulation corresponds well with the experiment in the trend that two obstacles is better than one obstacle and even better than zero obstacle.
Figure 10.
Snapshots of the experiment (blurred for privacy protection).
The left panel shows the venue with the participants in the moving circling when there is an obstacle (lower right corner). The trash bin with three participants pushing outside the gate was used to ensure the gate did not collapse. The right panel shows the escape process, when the participants formed a pushing arc at the gate.