Alidmat, Omar Khair Alla Abdel Rahman
(2021)
Extended Cellular Automata Simulation Model For Fire Crowd
Evacuation.
PhD thesis, Universiti Sains Malaysia..
Abstract
In recent years, crowd evacuation in case of fire accidents has attracted
considerate attention. Fire accidents occur in crowded buildings may cause heavy
casualties. The study of fire crowd evacuation has become extremely necessary to
minimize the loss of life and property. Large fires pose dangers; hence, computer
simulations are conducted as alternative tools to the deficiencies in conducting actual
fire evacuation experiments. Researchers have simulated evacuees’ movements in
panic situations, such as fires, using the cellular automata (CA) model to predict and
analyze evacuees’ behaviors during these panic situations. This could help minimize
accidents and save lives. However, those researchers have either investigated fire
accidents in a static scenario or propagate inaccurate fire circular fronts shape, such as
the adoption of a square fire front shape. They have also applied a lot of constraints on
the environmental and accident factors, such as fire location, fire spread speed,
obstacles, which could show evacuees movements appeared unrealistic. In addition,
the models used by those researchers ignored the effects of crowd pressure applied on
evacuees around overcrowded exits during fire evacuation. In this research, the spiral
fire movement technique was adopted using CA model to simulate the fire circular
surface propagation shape, which presents a non-static fire-spreading behavior that
able to estimate the average number of evacuees could be injured or killed by fire. In
addition, the new extended CA parameters (fire spreading, congestion and path), the
set of mathematics formulas, were introduced to simulate the decision-making of
evacuees in terms of movements and judgments and their choices of actions.
Actions (login required)
 |
View Item |
