River administration in Japan requires the ability to predict flooding events and to respond to them with appropriate judgment and action. The accuracy of systems for flood prediction modeling, which have been studied by the authors, has been increased with a new approach developed by the authors. We have developed a system for analyzing flood data from past typhoons and other flood disasters. We utilize this data to identify and prioritize the important duties of river managers during predicted typhoons. These prioritized duties are arranged into timeline-based disaster prevention action plans. This paper describes the design and development of a support system used to analyze flood data for disaster prevention and action plans. We specifically developed a typhoon search and prediction system, which utilized neural network modeling, for flood fighting, based on disaster prevention and action plans.
展开▼
