Earthquake Disaster Mitigation Based on Peak Ground Acceleration with Multi-Sensor System

Abstract

Indonesia is a country at the meeting point of tectonic plates, so it will likely experience the natural phenomenon of earthquakes. Earthquakes are natural events that humans cannot predict and cause fatal impacts, such as the number of fatalities and damage to buildings that cause a lot of material losses. This system was created to reduce the secondary effect of natural disasters. With the existence of an earthquake mitigation system in the form of an alarm warning and auto cut-off system, it is hoped that this system can minimize damage, loss of life and increase the level of earthquake alertness. Along with the development of technology, one of the main objectives of this research is to make people aware of evacuating immediately after the alarm and cut off some crucial flows such as electricity, gas, and others. This system utilizes multi-sensor technology, the Internet of Things, and Artificial Intelligence. This system implements 2 algorithms, such as K-Nearest Neighbor (KNN) and Logistic Regression. In KNN with 3 active sensors, the resulting accuracy is 94%. While Logistics Regression with 10 active sensors, the resulting accuracy is 83%.