Artificial Neural Networks for Risk Decision Support in Natural Hazards: A Case Study of Assessing the Probability of House Survival from Bushfires

Risk decision-making in natural hazards encompasses a plethora of environmental, socio-economic and management-related factors, and benefits greatly from exploring possible patterns and relations among these multivariate factors. Artificial neural networks, capable of general pattern classifications, are potentially well suited for risk decision support in natural hazards. This paper reports an example that assesses the risk patterns or probabilities of house survival from bushfires using artificial neural networks, with a simulation data set based on the empirical study by Wilson and Ferguson (Predicting the probability of house survival during bushfires, Journal of Environmental Management 23 (1986) 259–270). The aim of this study was to re-model and predict the relationship between risk patterns of house survival and a series of independent variables. Various configurations for input and output variables were tested using neural networks. An approach for converting linguistic terms into crisp numbers was used to incorporate linguistic variables into the quantitative neural network analysis. After a series of tests, results show that neural networks are capable of predicting risk patterns under all tested configurations of input and output variables, with a great deal of flexibility. Risk-based mathematical functions, be they linear or non-linear, can be re-modelled using neural networks. Finally, the paper concludes that the artificial neural networks serve as a promising risk decision support tool in natural hazards.     

Fiji's worst natural disaster: the 1931 hurricane and flood

At least 225 people in the Fiji Islands died as a result of the 1931 hurricane and flood, representing the largest loss of life from a natural disaster in Fiji's recent history. This paper explores the causes of disaster and the potential for recurrence. The disaster occurred because a rare event surprised hundreds of people—especially recently settled Indian farmers—occupying highly exposed floodplains in north‐west Viti Levu island. The likelihood of a flood disaster of such proportions occurring today has been diminished by changed settlement patterns and building materials; however, a trend towards re‐occupancy of floodplains, sometimes in fragile dwellings, is exposing new generations to flood risks. The contribution of this paper to the global hazards literature is set out in three sections: the ethnicity, gender and age of flood fatalities; the naturalness of disasters; and the merit of choice and constraint as explanations for patterns of vulnerability.     

Deaths in natural hazards in the solomon islands

Archival and library search techniques have been used to establish extensive databases on deaths and damage resulting from natural hazards in the Solomon Islands. Although the records of fatalities are certainly incomplete, volcanic eruptions, tropical cyclones, landslides, tsunami and earthquakes appear to have been the most important. Only 22 per cent of the recorded deaths have resulted from meteorological hazards but a single event could change this proportion significantly. Five events in the fatality database account for 88 per cent of the recorded deaths. Future death tolls are also likely to be dominated by a small number of events. While the expected number of deaths in a given period is dependent upon the length of record considered, it is clear that a disaster which kills one hundred or more people in the Solomons can be expected more frequently than once in a hundred years.     

Artificial Neural Networks for Risk Decision Support in Natural Hazards: A Case Study of Assessing the Probability of House Survival from Bushfires

Risk decision-making in natural hazards encompasses a plethora of environmental, socio-economic and management-related factors, and benefits greatly from exploring possible patterns and relations among these multivariate factors. Artificial neural networks, capable of general pattern classifications, are potentially well suited for risk decision support in natural hazards. This paper reports an example that assesses the risk patterns or probabilities of house survival from bushfires using artificial neural networks, with a simulation data set based on the empirical study by Wilson and Ferguson (Predicting the probability of house survival during bushfires, Journal of Environmental Management 23 (1986) 259–270). The aim of this study was to re-model and predict the relationship between risk patterns of house survival and a series of independent variables. Various configurations for input and output variables were tested using neural networks. An approach for converting linguistic terms into crisp numbers was used to incorporate linguistic variables into the quantitative neural network analysis. After a series of tests, results show that neural networks are capable of predicting risk patterns under all tested configurations of input and output variables, with a great deal of flexibility. Risk-based mathematical functions, be they linear or non-linear, can be re-modelled using neural networks. Finally, the paper concludes that the artificial neural networks serve as a promising risk decision support tool in natural hazards.