Evacuation Planning: a Case Study of the Walcheren-Zuid Beveland Area, the Netherlands

Evacuation Planning: a Case Study of the Walcheren-Zuid Beveland Area, the Netherlands


A J Pel, M C J Bliemer, S P Hoogendoorn, Delft University of Technology, NL


An innovative evacuation network model is applied to the Walcheren-Zuid Beveland area, the Netherlands. We elaborate on results gained from various simulations and show how these can be drawn on in real-life evacuation planning.


During the past decades, large storms, mudflows, hurricanes, bush fires, floods, and many other hazards have caused massive economic and social damage as well as loss of life. Explicit cost-benefit analyses show that by accepting evacuation as an optional response to such hazards, peoples? lives can be preserved at much lower costs. This is recognized and adopted by some national governments, among which the Netherlands. Unfortunately, to date, the process of evacuation is still relatively unknown, which impedes founded decisions when the moment is there. Better decisions can be made and society will benefit, if the process of evacuation is better understood. Given the complexity of the problem, a simulation-based model is required to get a better understanding of the process at hand. In this paper, we present a new evacuation network model providing this insight by specifically dealing with the dynamic interaction between hazard, authority and evacuees. More specific, the evacuees decide on their departure time, destination and route in which they act upon the (perceived) hazard conditions and the information and evacuation instructions (departure time window, evacuation route(s) and destination(s)) given by the authority. At the same time, the spatiotemporal effects of the hazard may cause road sections to become limited accessible.

The analytical macroscopic evacuation model enables forecasting traffic flow operations on a road network in case of an evacuation in a dynamic way, incorporating the spatiotemporal effect of the hazard. The newly proposed model is innovative as it combines a voluntary evacuation (with possible pre-trip and en-route decisions) and a mandatory evacuation (with prescribed routes and departure times) within one single generic model framework. The magnitude with which the evacuation instructions influence the evacuee?s decisions depends on the type of instructions (recommendations or (binding) orders) and the evacuee?s willingness to adjust its behavior in case the given instructions conflict with its personal preferences. This way we can model varying levels of compliance, such that evacuation instructions can be accepted fully, accepted partly, or rejected. In doing so, the model allows assessment of a wide range of evacuation schemes and hazard types, increasing the application domain. Moreover, by introducing a dynamic road infrastructure, in which road characteristics such as capacity and flow direction can be time-varying due to the hazard?s progress in space and time (e.g., blocked links due to flooding) and prevailing traffic regulations and control measures (e.g. contra flow to increase outbound capacity), the model can capture important changes in the infrastructure over time, making it very realistic. As a result, the model can pragmatically facilitate evaluation of alternative evacuation schemes in order to minimize casualties, costs of evacuation, time needed for complete evacuation, or any such criteria. In doing so, the evacuation model has the potential to function as a decision support system for regional authorities and emergency services supervising an evacuation or assigned to constructing an evacuation scheme for planning purposes.

The paper describes the successful application of the model to a case describing the evacuation of the Walcheren-Zuid Beveland area, the Netherlands, illustrating the potential of the proposed model. In the case study, a flood is considered forcing approximately 150,000 endangered residents to evacuate within a limited amount of time. We elaborate on the results gained from the model and show how these can be drawn on in evacuation planning. Various simulations are presented varying in evacuation instructions and public behavioral response. Herein, the voluntary evacuation shows the behavioral response of the public towards the hazard scenario in absence of an active evacuation scheme, while the recommended evacuation provides useful insight into the benefits of the specific evacuation scheme. Furthermore, we model different spatiotemporal patterns of the flood incident such that complete evacuation may or may not be feasible. In the former case, we consider the efficiency of an evacuation scheme (performance over time), while the latter case also gives rise to the notion of sensitivity of an evacuation scheme (performance under uncertainty).


Association for European Transport