Predicting and Assessing Wildfire Evacuation Decision-Making Using Machine Learning: Findings from the 2019 Kincade Fire

Xu, Ningzhe; Lovreglio, Ruggiero; Kuligowski, Erica D.; Cova, Thomas J.; Nilsson, Daniel; Zhao, Xingping

doi:10.1007/s10694-023-01363-1

Cited by 24 publications

(4 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous research has explored the performance of ML techniques compared to other statistical predictive methods in the field of human behaviour and decision-making. The results of these studies have been mixed, with some reporting better performance for ML techniques (Lindner et al, 2017;Xu et al, 2023), others for mixed logit models (Zhu et al, 2023), and others finding no significant difference between methods (Wang and Ross, 2018). In this study, both the XGBoost ML model and traditional LR model produced similar results in terms of variable importance and their influence on evacuation decisions.…”

Section: Discussionmentioning

confidence: 58%

“…The results indicated that demographics, destination, transport or household characteristics influenced decision-making. Xu et al, 2023 processed data from evacuation in wildfires using seven machine learning models compared to logistic regression. They used survey data from the Kincade Fire and founded that previous safety perception was a key factor in the binary decision of evacuate or stay.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Logistic Regression vs Machine Learning to Predict Evacuation Decisions in Fire Alarm Situations

Balboa

Cuesta

González‐Villa

et al. 2023

Preprint

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 58%

Section: Introductionmentioning

confidence: 99%

Logistic Regression vs Machine Learning to Predict Evacuation Decisions in Fire Alarm Situations

Balboa

Cuesta

González‐Villa

et al. 2023

Preprint

View full text Add to dashboard Cite

“…These models usually have a predetermined (linear/log-linear) model structure. Most of these models can only identify linear trends between each factor and the target variable and are often less accurate and less flexible 7 , 8 . To better capture people’s dynamic behavioral responses during earthquake emergencies, we require more advanced models that can account for more complex, nonlinear relationships 9 .…”

Section: Introductionmentioning

confidence: 99%

Modeling protective action decision-making in earthquakes by using explainable machine learning and video data

Zhang,

Zhao,

Baldwin

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Earthquakes pose substantial threats to communities worldwide. Understanding how people respond to the fast-changing environment during earthquakes is crucial for reducing risks and saving lives. This study aims to study people’s protective action decision-making in earthquakes by leveraging explainable machine learning and video data. Specifically, this study first collected real-world CCTV footage and video postings from social media platforms, and then identified and annotated changes in the environment and people’s behavioral responses during the M7.1 2018 Anchorage earthquake. By using the fully annotated video data, we applied XGBoost, a widely-used machine learning method, to model and forecast people’s protective actions (e.g., drop and cover, hold on, and evacuate) during the earthquake. Then, explainable machine learning techniques were used to reveal the complex, nonlinear relationships between different factors and people’s choices of protective actions. Modeling results confirm that social and environmental cues played critical roles in affecting the probability of different protective actions. Certain factors, such as the earthquake shaking intensity and number of people shown in the environment, displayed evident nonlinear relationships with the probability of choosing to evacuate. These findings can help emergency managers and policymakers design more effective protective action recommendations during earthquakes.

show abstract

“…An important research topic linking extreme fires and tourism development is risk perception for tourist destinations. Risk perception encompasses several areas in the context of fires, for example, fire evacuation decision making [5] and fire safety for heritage villages [6], among others. In the case of tourists, their perceptions of risk are the result of each person's feelings, so it is not possible to state that risk perception is something entirely concrete [7].…”

Section: Introductionmentioning

confidence: 99%

Assessing Fire Risk Perception in the Vale do Guadiana Natural Park, Portugal

Andrade

Couto

Serra

2023

Fire

View full text Add to dashboard Cite

This is an exploratory study aiming to assess the fire risk perception by operators of the Vale do Guadiana Natural Park (PNVG), southern Portugal. To maximize the sample size, a questionnaire survey was distributed among 35 entities with activities in tourism, hunting, and agriculture, as well as among members of PNVG’s co-management commission. For data analysis and interpretation, quantitative and qualitative analyses were used. Survey responses revealed that the entities expressed concern about and made efforts toward the search for and improvement in mitigation strategies in the occurrence of fires. A total of 69.6% of the respondents have knowledge of the occurrence of fires in the region. The qualitative analysis highlights the concern with biodiversity, as well as with the maintenance and cleaning of the PNVG. This study verifies the degree of importance that the tourism sector should give to the impacts caused by fires. The impact of climate change favoring fires was recognized by the entities, as well as the fact that the loss of biodiversity due to fires may have a direct impact on the attractiveness of this tourist destination, indicating the importance of environmental conservation strategies for the region.

show abstract

Predicting and Assessing Wildfire Evacuation Decision-Making Using Machine Learning: Findings from the 2019 Kincade Fire

Cited by 24 publications

References 60 publications

Logistic Regression vs Machine Learning to Predict Evacuation Decisions in Fire Alarm Situations

Logistic Regression vs Machine Learning to Predict Evacuation Decisions in Fire Alarm Situations

Modeling protective action decision-making in earthquakes by using explainable machine learning and video data

Assessing Fire Risk Perception in the Vale do Guadiana Natural Park, Portugal

Contact Info

Product

Resources

About