PENERAPAN LONG SHORT TERM MEMORY UNTUK MEMPREDIKSI JUMLAH PENUMPANG PESAWAT DAN KARGO DI BANDARA SOEKARNO-HATTA

ABDURAHMAN, SIGIT (2021) PENERAPAN LONG SHORT TERM MEMORY UNTUK MEMPREDIKSI JUMLAH PENUMPANG PESAWAT DAN KARGO DI BANDARA SOEKARNO-HATTA. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (HAL COVER) 01 COVER - SIGIT ABDURAHMAN.pdf Download (941kB) | Preview
	Text (BAB I) 02 BAB 1 - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (161kB)
	Text (BAB II) 03 BAB 2 - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (382kB)
	Text (BAB III) 04 BAB 3 - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (553kB)
	Text (BAB IV) 05 BAB 4 - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (110kB)
	Text (BAB V) 06 BAB 5 - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (185kB)
	Text (BAB VI) 07 BAB 6 - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (535kB)
	Text (DAFTAR PUSTAKA) 08 DAFTAR PUSTAKA - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (80kB)
	Text (LAMPIRAN) 09 LAMPIRAN - SIGIT ABDURAHMAN.pdf Restricted to Registered users only Download (690kB)

Abstract

The COVID-19 pandemic has had a considerable impact on the world of aviation, especially at Soekarno-Hatta Airport, Indonesia. Based on data obtained from the daily reporting application at the Directorate of Air Transport, the flow of passengers and cargo at Soekarno-Hatta Airport tends to fluctuate during this COVID-19 pandemic. Therefore, it is necessary to predict the flow of aircraft passengers and cargo in the future so that regulators or airports have readiness when there is a surge in passengers. In this study, predictions of the number of aircraft passengers and cargo at Soekarno-Hatta Airport were carried out using Long Short Term Memory modeling. The data used in this study is based on data from the Directorate of Air Transport, where the dataset used in this study is data on scheduled domestic and international aircraft passengers, both arrivals and departures at Soekarno-Hatta Airport with a period of January 2019 to December 2020. The results of the study This shows that the Long Short Term Memory (LSTM) model shows good results in making predictions with scenarios of 90% training data and 10% testing data, by producing prediction accuracy with a root mean squared error (RMSE) of 12,184 on training data and 12,402 on testing data. Meanwhile, the scenario with 80% training data and 20% test data resulted in a Root Mean Square Error (RMSE) value of 12.781 for training data and 13,382 for test data Key words: Air Transport Traffic, Long Short Term Memory, Predictions, Deep Learning Masa pandemi covid-19 memberikan dampak yang cukup besar bagi dunia penerbangan khususnya di Bandara Soekarno-Hatta, Indonesia. Berdasarkan data yang didapatkan dari aplikasi pelaporan harian di Direktorat Angkutan Udara, arus penumpang dan kargo di Bandara Soekarno-Hatta cenderung fluktuatif di masa pandemi covid-19 ini. Maka dari itu diperlukanlah suatu prediksi arus penumpang pesawat dan kargo di masa yang akan datang agar regulator atau pihak bandara memiliki kesiapan ketika terjadinya lonjakan penumpang. Pada penelitian ini dilakukan prediksi jumlah penumpang pesawat dan kargo di Bandara SoekarnoHatta dengan menggunakan pemodelan Long Short Term Memory. Data yang digunakan dalam penelitian ini berdasarkan data dari Direktorat Angkutan Udara, dimana dataset yang digunakan dalam penelitian ini merupakan data penumpang pesawat domestik dan internasional berjadwal, baik itu kedatangan ataupun keberangkatan di Bandara Soekarno-Hatta dengan periode waktu Januari 2019 hingga Desember 2020. Hasil penelitian ini menunjukan model Long Short Term Memory (LSTM) menunjukan hasil yang baik dalam melakukan prediksi dengan skenario 90% data training dan 10% data testing, dengan menghasilkan ketepatan prediksi dengan nilai root mean squared error (RMSE) sebesar 12.184 pada data training dan 12,402 pada data testing. Sedangkan untuk skenario dengan 80% data training dan 20% data testing menghasilkan nilai Root Mean Square Error (RMSE) sebesar 12.781 pada data training dan 13.382 untuk data testing. Kata kunci: Lalu Lintas Angkutan Udara, Long Short Term Memory, Prediksi, Deep Learning.

Actions (login required)

View Item