A Study of Isolated Area Identification Model for Natural Disaster Prevention based on Prim Algorithm

• Main Authors: Huang, Chun-Hsiang, 黃俊翔
• Other Authors: Wang, Wen-Ching
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/5ce9d8

碩士 === 國立臺東大學 === 資訊管理學系碩士班 === 102 === In recent years, because of climate changes, temperatures and rainfall index were often high all over Taiwan. Due to Taiwan's manufacturing industry and over-exploitation of natural resources on the hillside, natural disaster was not only man-made but also happened by the nature itself. Misfortune does not come singly, a disaster always comes with a secondary disaster, which will expand the scale of the main disaster increasingly and cause unprecedented large-scale disaster. Although the disaster preventing technology has been continuing to be improved to reduce the disaster impact, due to the complexity and uncertainty of disaster, it’s still difficult to achieve early warning and to control the hazards effectively. In 2009, Morakot Typhoon achieved a record-breaking average rainfall that had rewritten the largest single-day rainfall record in the past decade. In many places, the rain came pouring down on one day more than the total yearly average. Due to the instant storm, Taiwan's south and east region had flood, debris flow and other disaster, and even destroyed the Xiaolin village. This study based on the concept of pre-disaster prevention is more important than disaster rescue. The research uses Taitung region as a research area. The research takes Taitung region as a research area by means of the methodology of design science research, developing the isolated island area identification by Prim algorithm in accordance with the risk value measured by the factors including rainfall and potential of debris flow. The identification model was implemented in C# programming language incorporated with Google Map API. Using the Morakot Typhoon, Haitang Typhoon, Gaemi Typhoon and Songda Typhoon as the demonstration cases, the identification model was shown to successfully identify the isolated island regions and display the districts that can be passed through and the isolated regions on the Google Map for the users. With the final evaluation of feasibility and usability by five experts from National Taitung University of Disaster Science Reduce Center, it showed that the isolated island identification model was recognized among the experts by the evaluation results of totally average scoring 4.3 in technical feasibility, scoring 4.1 in economic feasibility, scoring 4.3 in organizational feasibility with three assessments.