Using opportunities in big data analytics to more accurately predict societal consequences of natural disasters

Abstract

ABSTRACT The availability of data sources has greatly increased due to advances in technology and data sharing. With these new data sources and significantly larger volume of data, engineers have been presented with a unique opportunity to create more realistic and informative models that can be used in real world applications. This paper presents a probabilistic framework for using big data to assess and predict the well-being of individuals before and in the aftermath of a hazard. Data are used to inform a Capability Approach (CA) where capabilities are defined as important dimensions of well-being reflecting what individuals have a genuine opportunity to do or become. The paper also addresses three of the grand challenges presented by big data: privacy, source validity, and accuracy. As an example, the probabilistic framework is used to study the ability of households in a coastal community to be sheltered in the aftermath of a hypothetical earthquake.