Pedestrian behaviour and its relation to doorway capacity

According to the Dutch building regulations, called Building Decree, 135 persons per minute should be able to safely escape via a doorway with a width of 1 meter, which is the same as 2.25 persons per second per meter. Since this value has been based on experiments conducted with healthy students, doubts have been raised on the validity of this value. Besides that, the fire department has also raised questions about this value. For these reasons the department of Transport & Planning has conducted a number of experiments for the Ministry of Housing, Spatial Planning, and the Environment to find out if the requirements in the Building Decree are realistic or should be adjusted. Insight into the individual behaviour in relation to the capacity is useful to gain a better understanding of the differences in capacities. These insights can then be used for predicting the capacities for other conditions [Hoogendoorn and Daamen, 2009] and they may provide an explanation for these differences in capacity. They therefore form a basis for recommendations for the Building Decree. This is used as a starting point for this project. For this project, a selection is made from these experiments. The experiments considered are the experiments with varying doorway widths (50 cm, 85 cm, 100 cm, 110 cm, 165 cm, 220 cm, and 275 cm) and the experiment where a door was connected to the wall fixed at an angle of 90°. For all experiments three stress levels (no stress, slow-whoop alarm, and slow-whoop alarm together with a stroboscope light) have been considered. The doorway width, an open door, and the stress levels are called the experimental variables. The experiments have been conducted with a population that was considered to represent the average population in the Netherlands, which consists for about 50% of adults, 25% of children, and 25% of elderly people. The research objective is to gain insight into pedestrian behaviour in relation to the differences in capacities for different evacuation doorway configurations and for different external conditions that occur during an emergency. These insights will form the basis for recommendations to improve the safety of existing and new buildings during evacuation conditions. Therefore, the following research questions have been formulated: 1. Which aspects concerning pedestrian walking behaviour are expected to influence doorway capacities? 2. What are the effects of these aspects on the capacity? 3. How can the results from the research questions above be used to improve the safety of existing and new buildings during evacuation conditions? In order to answer the first research question, a literature review has been performed. From this, three groups of aspects resulted: microscopic and macroscopic characteristics and self-organisation phenomena. The microscopic characteristics considered are the distance headways and the occupied area. The macroscopic characteristics are the density, flow, and speed. In order to analyse self-organisation, the composition, so which positions do pedestrians occupy with respect to each other, of the population is analysed. To answer the second research question, the actual analysis has been performed. It became clear that especially the time of day at which the experiments have been conducted influences the results. In the morning the participants were much more motivated and still physically fit. The capacity was on average 20% higher for the first three experiments than for the other experiments. During the analysis it became clear that the capacity is also higher when the percentage of children is higher. This is probably because children are much more enthusiastic, resulting in higher speeds. Comparing the different doorway widths, the capacity estimations were between 2.20 and 2.40 P/s/m for all doorway widths, taking 20% lower values for the first three experiments into account. The only experiment with a significantly lower capacity was the experiment with an open door (2.10 P/s/m). The stress level ‘slow-whoop alarm together with a stroboscope light’ is the only stress level which seems to have an effect on the capacity. Only a ‘slow-whoop alarm’ does not have a significant impact on the capacity. Since stress only had an effect during the first three runs of each stress level, which were all runs of the first experiment and the first run of each stress level of the second experiment, pedestrians seem to get used to stress and are not affected by it anymore when they have become used to it. After analysing the microscopic characteristics it became clear that for higher capacities, the distance headways are smaller and the occupied angles between the congested region and the wall where the doorway is located in are also smaller. For considering macroscopic variables, density, flow, and speed just upstream of the doorway have been analysed. The capacity increases when the density, flow, and speed just upstream of the doorway all increase. During the analysis of self-organisation it became clear that the higher the mean number of side-by-sides ‘inside’ the doorway, the higher the capacity. Besides that, the lower the mean time headways. To answer the third research question, the differences in capacity have been analysed in relation to the experimental variables. The conclusion was that most of the experiments resulted in higher capacities than the value in the Building Decree (2.25 P/s/m). However, the experiments have been conducted under ideal circumstances, which means lower values are to be expected in reality. The advice is to find out what the effect of ideal circumstances is on the capacity. Since the effect is unknown the advice for now is to adjust the value for the capacity in the Building Decree to the lowest average value that occurred during the experiments, which is 2.10 P/s/m. Concluding, the research questions have been answered and more insight is gained into pedestrian behaviour during evacuation in relation to doorway capacity. More is known about the relation between capacity and experimental variables and between capacity and microscopic and macroscopic aspects but also self-organisation. With these insights it is possible to predict the capacity of not observed situations, such as other doorway widths that have not been analysed.Tranport, Infrastructure and LogisticsTransport & PlanningCivil Engineering and Geoscience