Title:	REFINEMENT OF THE QUANTITATIVE MODELS TO ESTIMATE USER'S FEAR IN EVACUATION ROUTE PLANNING: INTRODUCTION OF USER ATTRIBUTES AND NONLINEARIZATION OF THE MODEL
Author(s):	Hiroshi Furukawa and Ryota Koshimizu
ISBN:	978-989-8704-38-2
Editors:	Piet Kommers, Inmaculada Arnedillo Sánchez and Pedro Isaías
Year:	2022
Edition:	Single
Keywords:	Disaster Evacuation, Anxiety, Emotion Modeling, Reassured, Disaster Mitigation, Intelligent Transport Systems
First Page:	45
Last Page:	52
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	During disaster evacuation, fear or panic can force evacuees to make irrational decisions. The final goal of our project is the development of a navigation system that provide routes where evacuees go through reassured. The aim of this study is to improve the model created by our previous study. In the first stage of this study, we considered the user's attributes as additional factors in the fear estimation model, and also constructed the fear estimation model by nonlinearizing the model, i.e., Random Forest Regression and Support Vector Regression. By comparing the results with the previous model, we verified whether the improvement in model accuracy could be measured. The results showed that the model created by Random Forest Regression was the most versatile and the most accurate. In the second stage, in order to evaluate whether the proposed revised method improves the accuracy of the model, we conducted cognitive experiments using the route with the revised model and the shortest route. The results show that the mean value of the level of fear is lower for the revised method than for the shortest path. It is expected that the pedestrian navigation system based on the proposed method can provide routes where users can evacuate reassured, avoiding places where they may feel great fears.
	Go Back