Advanced

Modelling Ascending Stair Evacuation

Eriksson Lantz, Caroline LU (2015) VBRM05 20151
Division of Fire Safety Engineering
Abstract
The thesis presents a basic validation study on the use of evacuation models for the simulation of ascending stair evacuation. The validation of the evacuation models is performed against a benchmark experiment consisting of a 50-floor ascending evacuation. The aim is to find which models – depending on the input calibration effort -can provide conforming results against the benchmark experiment regarding total evacuation time and walking speeds at each floor. This is performed by selecting five evacuation models for validation, representing different types of modelling assumptions. These models are then used to simulate ascending evacuation applying default settings and modified settings. The study indicates that models under... (More)
The thesis presents a basic validation study on the use of evacuation models for the simulation of ascending stair evacuation. The validation of the evacuation models is performed against a benchmark experiment consisting of a 50-floor ascending evacuation. The aim is to find which models – depending on the input calibration effort -can provide conforming results against the benchmark experiment regarding total evacuation time and walking speeds at each floor. This is performed by selecting five evacuation models for validation, representing different types of modelling assumptions. These models are then used to simulate ascending evacuation applying default settings and modified settings. The study indicates that models under consideration are not conservative when applying their default settings but models which have the possibility to alter the reducing speed factors per floor generally give better conforming results to the experimental results. The study also shows that the models estimation of the walked distance is a crucial factor. (Less)
Popular Abstract
Validation study on the use of evacuation models for simulation of long ascending stair evacuation

Properties which slow down walking speed with walked vertical distance and the estimation of the distance walked are the crucial factors when evacuation models simulate evacuation in long ascending stairs.

Increased urbanisation leads to the construction of more and often complex underground facilities to cope with an increasing demand of e.g. transportation facilities. In case of evacuation from such a facility, long ascending stairs may have to be travelled and it can be assumed that physical fatigue can influence the walking speed negatively and thus the total time of evacuation. There is limited information and validation... (More)
Validation study on the use of evacuation models for simulation of long ascending stair evacuation

Properties which slow down walking speed with walked vertical distance and the estimation of the distance walked are the crucial factors when evacuation models simulate evacuation in long ascending stairs.

Increased urbanisation leads to the construction of more and often complex underground facilities to cope with an increasing demand of e.g. transportation facilities. In case of evacuation from such a facility, long ascending stairs may have to be travelled and it can be assumed that physical fatigue can influence the walking speed negatively and thus the total time of evacuation. There is limited information and validation available on the reliability of evacuation models in long ascending stair evacuation cases. Since these evacuation models are used for designing purposes it is a crucial topic to study.
The validation study of the ascending evacuation scenario in long stairs using a full scale experiment as benchmark was performed and focused on the simulated total evacuation time and walking speed. Three different input configurations were applied to the models to find possible changes in the results in relation to the degree of user effort in the input calibration phase. The configurations were; default settings, applied reducing speed factors and a modified distribution of the initial walking speed.
The evacuation models’ (EXIT89, FDS+Evac, Pathfinder, Simulex and STEPS) default settings do in general underestimate the evacuation time while with applied reducing speed factors the evacuation time is better conforming to the 50 floor benchmark experiment study. Modification of the models initial walking speed distribution does not give better corresponding results and should thus be kept default. It is the models possibility to simulate deceleration of walking speed that is an important factor and not the model type (continuous, coarse- or fine network). The validation study also indicates that the vertical distance walked should be determining for decelerating walking speed and thus indirectly fatigue, not the horizontal distance walked. Surprisingly, since it was not a factor thought of at the commencement of the study, the models’ assumptions affecting the travelled distance is one of the most settling factors. Large divergences in walked distance, where the evacuation model STEPS estimate the distance to approximately 30 % less than the benchmark experiment, shows that the calculation of walked distance should be thoroughly considered when using evacuation models for simulations as it will affect walking speed and the total evacuation time.
EXIT89 and Simulex are not applicable for simulation of evacuation in long stairs since deceleration of walking speed not is available. The models, FDS+Evac, Pathfinder and STEPS, are not recommended for simulation of ascending evacuation in long stairs with default settings but with the application of reducing speed factors. They can be used for simulation of evacuation cases similar to the benchmark experiment, approximately 50 floor ascending stair evacuation, with considerable user input regarding deceleration of walking speed. (Less)
Please use this url to cite or link to this publication:
author
Eriksson Lantz, Caroline LU
supervisor
organization
course
VBRM05 20151
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Ascending stair evacuation, stair evacuation modelling, walking speed in stairs.
report number
5495
ISBN
LUTVDG/TVBB—5495--SE
language
English
id
7793450
date added to LUP
2015-08-31 13:16:48
date last changed
2015-08-31 13:16:48
@misc{7793450,
  abstract     = {The thesis presents a basic validation study on the use of evacuation models for the simulation of ascending stair evacuation. The validation of the evacuation models is performed against a benchmark experiment consisting of a 50-floor ascending evacuation. The aim is to find which models – depending on the input calibration effort -can provide conforming results against the benchmark experiment regarding total evacuation time and walking speeds at each floor. This is performed by selecting five evacuation models for validation, representing different types of modelling assumptions. These models are then used to simulate ascending evacuation applying default settings and modified settings. The study indicates that models under consideration are not conservative when applying their default settings but models which have the possibility to alter the reducing speed factors per floor generally give better conforming results to the experimental results. The study also shows that the models estimation of the walked distance is a crucial factor.},
  author       = {Eriksson Lantz, Caroline},
  isbn         = {LUTVDG/TVBB—5495--SE},
  keyword      = {Ascending stair evacuation,stair evacuation modelling,walking speed in stairs.},
  language     = {eng},
  note         = {Student Paper},
  title        = {Modelling Ascending Stair Evacuation},
  year         = {2015},
}