Phased evacuation: An optimisation model which takes into account the capacity drop phenomenon in pedestrian flows

Abstract

Experiments have been performed to study pedestrian flow through bottlenecks under oversaturated conditions. The data underline a new phenomenon in pedestrian flows: when high-density conditions occur upstream of a bottleneck, the maximum capacity of the bottleneck (i.e. the maximum number of pedestrians that can flow through the bottleneck in a given time interval) can drop. This is referred to as “capacity drop” and should be carefully taken into account when dealing with building evacuation philosophy. Controlling the population density upstream of a bottleneck can maximise the flow through it (i.e. by avoiding capacity drop) and shorten the building evacuation time significantly. This paper focuses on alarm times and egress routes (in terms of their geometrical characteristics like length, width) as means for controlling population densities and therefore reducing building evacuation time. Specifically, the proposed methodology includes an optimisation algorithm that explores the research space (defined by all the possible Route-Alarm Time schedule plans where a Route-Alarm Time schedule plan is in turn defined by a set of alarm times and a set of egress routes) and aims to find the Route-Alarm Time schedule plan that minimises the building evacuation time. We call building evacuation time the temporal gap between the time the first person starts to egress the building until all people have reached an area of safety outside the building. A newly proposed movement model, embedded into the optimisation algorithm, allows the building evacuation time for a given Route-Alarm Time schedule plan to be assessed by simulating the egress dynamics. This movement model is able to reproduce the capacity drop phenomenon observed in the experimental work and can therefore assess the building evacuation time taking into account the capacities of the various bottlenecks adjusted to take into account the actual upstream densities. The optimisation algorithm, with the movement model embedded in it, has been implemented in an object-oriented simulator. The Route-Alarm Time schedule planning refers to a phased evacuation in a static environment where all the egress paths are viable and the pedestrian population is well-trained.