Applying Genetic Algorithm on Selecting Emergent Medical Station before Disasters

This study applying a mathematical approach to select facilities for the emergent rescue plan before the natural disasters, such as earthquakes, typhoons and floods. The hazardous area is divided into sub-area by the capacity of medical treatments and the distance between the habitants, and the emergent medical station is located at the position of the seed of this sub-area. When if any disaster happens, the wounded can be sent directly to the nearest medical station and will accept proper treatments. Such a problem can be formulated as the so-called Capacitated Clustering Problem (CCP). The CCP is to partition a group of n items (ex. the habitants) into k clusters (ex. Sub-areas) and the entities within a cluster should be as homogeneous as possible and under volume constraints. This study applies genetic algorithm (GA) to solve the CCP and the solution quality is compared with integer optimization software, LINDO. Further research of the emergency evacuation plan is another similar problem and worthy of more detailed study.