An Evacuation Simulation Based on Boltzmann Distribution - Advanced Research on Computer Education, Simulation and Modeling - page 331

The initial
distribution of the
people

The choosing
circumstances of
the people

Fig. 2. The circumstances of people's choosing different exits

kk in the exit
attraction function and these parameters directly influence the results of the simula-
tion. However, there isn't a standard in the real circumstances. What we can do is to
summarize an experiential standard by ways of watching some evacuation videos and
so on. Therefore, we do many times of simulation and make the parameter values of
the simulation which we consider is the most close to the real case.

We need to determine some parameters in our model, such as the

12

3 Compare to the Classical Models

In order to test the rationality of our model, we apply the classical Social Force
Model, Cellular Automata Model and our model to the same initial evacuation condi-
tions and then compare the results of different models.
In a square room with the edge length 20m, we assume there is a huge obstacle at
the center of the room which has an area of about 28 square meters. At the very be-
ginning, the people are in the left side of the obstacle and the exit is in the right side
of the obstacle. The results of the three models are shown in Figure 3:

Cellular Automata Model Social Force Model Our Model

Initial People Distribution

The people are able to
avoid the obstacle in the
tangential direction

The path of avoiding the
obstacle is not ideal

The people are able to
avoid the obstacle

Fig. 3. The comparison of the results of the three models