{"title":"通过大型语言模型代理模拟评估家庭对飓风食物短缺的低估","authors":"Junkang Xu, Chao Fan","doi":"10.1016/j.ijdrr.2025.105734","DOIUrl":null,"url":null,"abstract":"<div><div>Food shortages are a significant challenge to public health during natural disasters like hurricanes. Household food shortages during disasters often result from limited access to grocery stores, limited supplies of food, and limited preparation of food at home before a disaster. Existing literature has examined food shortages regarding food supply security and access to food providers. Little is known about how households estimate the duration of food shortages and make decisions about food preparation before disasters. This research proposes a method to quantify the causal impacts of socio-demographics, past experiences, economic status, and trust in community infrastructure on the estimation of food preparation, which also informs effective warning information to enhance the food preparation of households. We adopt a post-Hurricane Harvey household survey conducted in Harris County to apply Generalized Structural Equation Models and find that households with past hurricane experience are more prone to underestimating the impacts than those without such experience. Single-gender households underestimate food shortages more than balanced-gender households, while education is not a significant determinant. Furthermore, we incorporate the decision mechanisms into the Large Language Model and simulate household estimation of food shortages to different types of warning information tailored by economic, infrastructure, and social impacts. Warning information emphasizing infrastructure impacts is found to be effective for most households, while social impact-focused warning significantly improves elderly households’ estimations. Our work offers insights into human perceptions of disaster impacts and customized warning interventions in reducing the underestimation of hurricane-induced food shortages and emphasizes the need for context-specific delivery.</div></div>","PeriodicalId":13915,"journal":{"name":"International journal of disaster risk reduction","volume":"128 ","pages":"Article 105734"},"PeriodicalIF":4.5000,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Assessing household underestimation of hurricane food shortages via large language model agent simulation\",\"authors\":\"Junkang Xu, Chao Fan\",\"doi\":\"10.1016/j.ijdrr.2025.105734\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Food shortages are a significant challenge to public health during natural disasters like hurricanes. Household food shortages during disasters often result from limited access to grocery stores, limited supplies of food, and limited preparation of food at home before a disaster. Existing literature has examined food shortages regarding food supply security and access to food providers. Little is known about how households estimate the duration of food shortages and make decisions about food preparation before disasters. This research proposes a method to quantify the causal impacts of socio-demographics, past experiences, economic status, and trust in community infrastructure on the estimation of food preparation, which also informs effective warning information to enhance the food preparation of households. We adopt a post-Hurricane Harvey household survey conducted in Harris County to apply Generalized Structural Equation Models and find that households with past hurricane experience are more prone to underestimating the impacts than those without such experience. Single-gender households underestimate food shortages more than balanced-gender households, while education is not a significant determinant. Furthermore, we incorporate the decision mechanisms into the Large Language Model and simulate household estimation of food shortages to different types of warning information tailored by economic, infrastructure, and social impacts. Warning information emphasizing infrastructure impacts is found to be effective for most households, while social impact-focused warning significantly improves elderly households’ estimations. Our work offers insights into human perceptions of disaster impacts and customized warning interventions in reducing the underestimation of hurricane-induced food shortages and emphasizes the need for context-specific delivery.</div></div>\",\"PeriodicalId\":13915,\"journal\":{\"name\":\"International journal of disaster risk reduction\",\"volume\":\"128 \",\"pages\":\"Article 105734\"},\"PeriodicalIF\":4.5000,\"publicationDate\":\"2025-08-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International journal of disaster risk reduction\",\"FirstCategoryId\":\"89\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2212420925005588\",\"RegionNum\":1,\"RegionCategory\":\"地球科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"GEOSCIENCES, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International journal of disaster risk reduction","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2212420925005588","RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GEOSCIENCES, MULTIDISCIPLINARY","Score":null,"Total":0}
Assessing household underestimation of hurricane food shortages via large language model agent simulation
Food shortages are a significant challenge to public health during natural disasters like hurricanes. Household food shortages during disasters often result from limited access to grocery stores, limited supplies of food, and limited preparation of food at home before a disaster. Existing literature has examined food shortages regarding food supply security and access to food providers. Little is known about how households estimate the duration of food shortages and make decisions about food preparation before disasters. This research proposes a method to quantify the causal impacts of socio-demographics, past experiences, economic status, and trust in community infrastructure on the estimation of food preparation, which also informs effective warning information to enhance the food preparation of households. We adopt a post-Hurricane Harvey household survey conducted in Harris County to apply Generalized Structural Equation Models and find that households with past hurricane experience are more prone to underestimating the impacts than those without such experience. Single-gender households underestimate food shortages more than balanced-gender households, while education is not a significant determinant. Furthermore, we incorporate the decision mechanisms into the Large Language Model and simulate household estimation of food shortages to different types of warning information tailored by economic, infrastructure, and social impacts. Warning information emphasizing infrastructure impacts is found to be effective for most households, while social impact-focused warning significantly improves elderly households’ estimations. Our work offers insights into human perceptions of disaster impacts and customized warning interventions in reducing the underestimation of hurricane-induced food shortages and emphasizes the need for context-specific delivery.
期刊介绍:
The International Journal of Disaster Risk Reduction (IJDRR) is the journal for researchers, policymakers and practitioners across diverse disciplines: earth sciences and their implications; environmental sciences; engineering; urban studies; geography; and the social sciences. IJDRR publishes fundamental and applied research, critical reviews, policy papers and case studies with a particular focus on multi-disciplinary research that aims to reduce the impact of natural, technological, social and intentional disasters. IJDRR stimulates exchange of ideas and knowledge transfer on disaster research, mitigation, adaptation, prevention and risk reduction at all geographical scales: local, national and international.
Key topics:-
-multifaceted disaster and cascading disasters
-the development of disaster risk reduction strategies and techniques
-discussion and development of effective warning and educational systems for risk management at all levels
-disasters associated with climate change
-vulnerability analysis and vulnerability trends
-emerging risks
-resilience against disasters.
The journal particularly encourages papers that approach risk from a multi-disciplinary perspective.