Developing insurance solutions software for natural hazard loss estimation

The population explosion in the second half of the twentieth century has caused exponential growth in economic losses associated with the natural and man-made disasters. Efforts to better understand economic consequences of catastrophic events have resulted in a number of computer programs that are best exemplified by the well known HAZUS software from the Federal Emergency Management Agency. HAZUS implements a multi-hazard loss estimation methodology that focuses on big picture solutions to regional problems. It has proven valuable when used within the framework specified by the program. This framework, however, is not ideally suited to predict loss-benefit ratios for individual structures of the types used by insurance industries. There are several commercial insurance programs, each suffering from their own model limitations and with closed source codes hindering industry wide contributions. The focus of this research is on the methodology by which basic insurance solution software may be developed and implemented to meet the societal and commercial needs for regulating and writing the policies issued to individual home owners.; The insurance solution software developed as part of this investigation offers two types of analysis. An event based analysis assumes specific design scenarios while a maximum probable loss analysis considers the likely outcome over a user defined time span. The input data required by the computer program is structured in three levels, depending on the user expertise and the extent by which the input data may be generated. The investigation requirement ranges from a basic inquiry to an engineering review of the characteristics of the building. Using the city of Mayaguez in the western part of Puerto Rico as a test bed, building inventories are developed to analyze a real construction environment. Assimilating the outcome from the analysis tools, retrofitting measures, and uncertainty modeling, seismic and hurricane wind fragilities are quantified. Long term projections based on the maximum probable loss analysis are also reported. The changes in existing vulnerability functions are examined to reflect differences in the building types and construction practice.