A real-time stochastic evacuation model for road tunnels

Abstract

Traditionally most of egress models have been mainly used for performance-base assessment and forensic analyses. But their use has been extended to other applications such as real-time mode. In this paper we present EvacTunnel® a real-time model integrated in a Decision Support System (DSS) for emergency management in road tunnel. The proposed model is based on the idea that, in some scenarios such as road tunnels, egress calculations can be performed by addressing a small set of random parameters that have a great impact on outcomes. These parameters are identified as pre-movement times (recognition and response time), travel distances and unimpeded walking speeds. Based on Monte Carlo methods, the model has the capability to perform multiple simulations by changing random variables. As a first stage of verification process, the proposed model is compared with other validated evacuation models. The results are essentially coincident in all models. Based on this analysis it can be argued that the model provides consistent and reasonable results. But the main difference is that the proposed model can provide results faster than real-time (less than 5s) while the run time of the other models is really higher.