Experimental study on population density of different occupant types in supermarket evacuation

Population density of different occupant types is the most critical factor affecting pedestrian evacuation in supermarket. Unfortunately, the experimental population density, especially on weekdays and holidays, is very scarce for reference. In order to obtain exact experimental occupant density, six investigations and experiments have been conducted in a Carrefour supermarket on workdays and holidays. Four occupant types (male, female, elderly and child), about 154,327 customers, were separately recorded in 6 days. By analyzing the experiment data with black-box model of occupant density built in this article, we found that the proportion of female, male, elderly, and child are 52.3%, 38.4%, 5.1%, and 7.2% respectively. Proportion of child on weekdays is about 3.6% but it rapidly rises to 10.9% on holidays while the number of the elderly is relatively stable, which is about 1,200 each day. Furthermore, the average peak value of occupant density in the market at the rush hours is about 0.15 persons/m2 appearing at 16:00~20:00 on weekdays while it is about 0.32 persons/m2 emerging at 13:00~19:00 on holidays. At last, based on the experimental data, two numerical simulations on supermarket evacuation are conducted. By simulating, we find that the total evacuation time on holidays is about twice as long as that on weekdays. These data and conclusions obtained from this experiment would provide valuable reference to evacuation model, evacuation simulation, and fire protection design of market.