Cumincad : CUMINCAD Papers : Search Results

Hits 1 to 1 of 1

Reformat results as: short short into frame detailed detailed into frame

_id	caadria2013_061
id	caadria2013_061
authors	Lyu Junchao, Sun Chengyu and Zhao Qi
year	2013
title	Does the Debate on Stress Effect Lead Evacuation Simulation Models to Different Performances?
doi	https://doi.org/10.52842/conf.caadria.2013.179
source	Open Systems: Proceedings of the 18th International Conference on Computer-Aided Architectural Design Research in Asia (CAADRIA 2013) / Singapore 15-18 May 2013, pp. 179-187
summary	There exists a long time debate over the rational state of the evacuees in the evacuation simulation. Sime et al. insisted that the evacuees behave rationally in most cases. Following this argument Arthur and Passini concluded that if a setting works well under normal conditions, it will have a better chance of working well in emergency conditions. Such an argument for rational decision-making makes statistical models, such as Utility Maximizing Model (UMM), acting as proper framework for choice modelling in evacuation. However, on the other side, some people pointed out that there exists panic in the evacuation and bounded rationality theory, represented by pruned Decision Tree Model (DTM), should be adapted in choice modelling instead. Such a debate over stress effect causes hesitations and uncertainties when researchers try to select a proper model framework for their practical simulation. This study compared the performances (prediction accuracies) of the UMM and DTM with the same set of choice data collected in a virtual evacuation experiment. With a similar level of performance on both sides, it is suggested that the on-going debate does NOT bring any difference to the choice modelling in practical evacuation simulation. It is time to improve the performance rather than to continue the debate.
wos	WOS:000351496100018
keywords	vacuation, Architectural cue, Machine learning, Bounded rationality
series	CAADRIA
email
last changed	2022/06/07 07:59

No more hits.