DETEKSI JARAK PANDANG AMAN SEBAGAI ACUAN UNTUK KESELAMATAN PENERBANGAN DENGAN MENGGUNAKAN METODE BACKPROPAGATION

Abstract

Delay is a term used when the flight schedule late while depart or arrive. Delay causes due to bad weather, technical problems on the plane, uncompleted crew, and others. Delay is procedure to avoid unexpected condition for example is accidents. Bad weather is the the one of delay reason difficult to estimated or detected because relativity variable of each hour. Bad weather can be detected by the air visibility. According to the official standart of transportation the minimum distance of visibility is 5 kilometers. While distance less than for flight safety, can be categorized as danger. Therefore, the porpose of this research is to know early when the flight has to be delayed or not. Therefore to detect a weather condition can be use backpropagation method which can be identified through several factors such as wind speed, humidity, and temperature in the air. Based on this research using 720 data, obtained a high accuracy of 95.71%, recall of 97.81%, and precision of 94.98%. With learning rate of 0.1 and hidden layer as much as 100, which shows that the detection of visibility has a very good performance. Therefore the model can be used for the detection of the visibility of the safe.