<![CDATA[One of the form of disaster is fire. Fire is a condition where buildings in a certain place or settlement, factory, market, and other places are hit by fire which cause victims and loss. Forest and land fire gives many negative effects in various fields such as biology, health, economy, and climate. Fire in Indonesia is one of the general problem which need to be noticed because it happens in every year. Fire can be observed through satellite by detecting hotspot on Earth surface. Fire occurrence rates which keep increasing every year need to be observed and prevented. That's why, there is a need for research to predict the number of hotspot which identify fire disaster in a certain time. This research proposes and creates program to predict the number of hotspot occured in Java Island using Fuzzy Inference System Mamdani. The data used is hotspot data in Java island from the beginning of 2012 to the end of 2016. The reason why Java island is chosen is because hotspots in Java island is easier to be observed because there are more civilian and lesser forests than other islands. There are three testing method in this research, with testing data of 2 years, 1 years and 6 month. Testing is done to know the accuracy of the number of hotspot prediction in 10 days period. Testing result shows that the system can predict the number of hotspot with quiet good result. 10 days period hotspot prediction has MSE value of 5432.37196 with testing data of 2 years. 10 days period hotspot prediction has MSE value of 5381.73457 with testing data of 1 years. 10 days period hotspot prediction has MSE value of 9483.8878 with testing data of 6 months.]]>
Copyrights © 2019
