Description Logic Based Earthquake Damage Estimation for Disaster Management

Abstract

Earthquake damage estimation within the first few hours of devastating earthquake is critical from the point of decision-making and post-disaster planning. With recent developments in earthquake engineering, earthquake damage estimation can be effectively determined using strong ground motion network. However, a number of densely populated cities in earthquake risk zones do not have access to such costly and advanced systems. In this paper, we propose a system for estimating earthquake damage in the early post-disaster period, using a limited number of data sources, such as up-to-date earthquake source parameter, population, building quality, roads and bridge locations. This information, however, is distributed, and therefore poses semantic heterogeneity problems in data sharing. For this reason, it is very challenging to derive accurate damage and loss estimation solely based on the source parameter of the earthquake. To solve this problem, in the proposed system, ontologies are used to ensure semantic interoperability, and description logic (DL) rules are defined for damage estimation. Other features of this study include the use of parliament triple store to store and query disaster ontology, and the provision of damage assessment maps to enable disaster managers to reduce their response time.