Pemanfaatan Model Seasonal Autoregressive Integrated Moving Average (SARIMA) Untuk Akurasi Prediksi Waktu Pada Jam Atom Cesium Di Badan X

Sirait, Rezki Ojak Taruna (2025) Pemanfaatan Model Seasonal Autoregressive Integrated Moving Average (SARIMA) Untuk Akurasi Prediksi Waktu Pada Jam Atom Cesium Di Badan X. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5009221150-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (6MB) | Request a copy

Abstract

Jam atom Cesium merupakan alat ukur waktu yang sangat akurat dan menjadi standar utama dalam menjaga waktu universal yang digunakan pada berbagai aplikasi teknologi tinggi, seperti navigasi satelit (GPS), telekomunikasi, sistem keuangan, dan penelitian ilmiah. Meskipun sangat presisi, jam atom Cesium tidak sepenuhnya bebas dari kesalahan dan anomali yang dapat disebabkan oleh berbagai faktor eksternal, sehingga memengaruhi akurasi dan kestabilan waktu yang dihasilkan. Penelitian ini bertujuan untuk meningkatkan akurasi dan prediksi kestabilan waktu pada jam atom Cesium yang digunakan di Badan X dengan menerapkan model SARIMA (Seasonal Autoregressive Integrated Moving Average). Model ini digunakan untuk menganalisis data REFGPS sebagai referensi waktu eksternal guna mengidentifikasi pola musiman dan non-musiman dalam data serta memprediksi potensi ketidakstabilan waktu. Hasil penelitian menunjukkan bahwa model SARIMA (1,1,2)(0,1,1)90 memiliki performa prediksi yang unggul dibandingkan model Autoregressive Integrated Moving Average (ARIMA). Pada model ARIMA, nilai Mean Absolute Percentage Error (MAPE) adalah 2.009052%, Mean Squared Error (MSE) sebesar 55454.08731 ns, dan Root Mean Squared Error (RMSE) sebesar 235.4869154 ns. Sementara itu, model SARIMA menghasilkan MAPE sebesar 0.006718%, MSE sebesar 1423.2724 ns, dan RMSE sebesar 37.72628318 ns.Dengan MAPE yang sangat kecil (0.006718%), model SARIMA terbukti memiliki kemampuan prediksi yang sangat baik. Penelitian ini diharapkan dapat berkontribusi dalam meningkatkan performa dan keandalan sistem yang menggunakan jam atom Cesium, sekaligus meminimalkan kesalahan waktu dalam berbagai aplikasi kritis.
=================================================================================================================================
Cesium atomic clocks are highly accurate time measurement instruments and serve as the primary standard for maintaining universal time used in various high-tech applications, such as satellite navigation (GPS), telecommunications, financial systems, and scientific research. Despite their precision, Cesium atomic clocks are not entirely free from errors and anomalies, which can be caused by various external factors, thus affecting the accuracy and stability of the generated time. This study aims to enhance the accuracy and predict the time stability of Cesium atomic clocks used in Institution X by applying the SARIMA (Seasonal Autoregressive Integrated Moving Average) model. This model is utilized to analyze REFGPS data as an external time reference to identify seasonal and non-seasonal patterns in the data and predict potential time instabilities. The research findings demonstrate that the SARIMA (1,1,2)(0,1,1)90 model exhibits superior predictive performance compared to the ARIMA model. For the ARIMA model, the Mean Absolute Percentage Error (MAPE) is 2.009052%, the Mean Squared Error (MSE) is 55454.08731 ns, and the Root Mean Squared Error (RMSE) is 235.4869154 ns. Meanwhile, the SARIMA model achieves a MAPE of 0.006718%, an MSE of 1423.2724 ns, and an RMSE of 37.72628318 ns. With an exceptionally low MAPE of 0.006718%, the SARIMA model proves to have outstanding predictive capabilities. This study is expected to contribute to improving the performance and reliability of systems utilizing Cesium atomic clocks while minimizing time errors in various critical applications.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Jam Atom Cesium, SARIMA, REFGPS, Prediksi Waktu, Cesium Atomic Clocks, SARIMA, REFGPS, Time Prediction
Subjects:	T Technology > T Technology (General) > T174 Technological forecasting
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Rezki Ojak Taruna Sirait
Date Deposited:	31 Jan 2025 07:19
Last Modified:	31 Jan 2025 07:19
URI:	http://repository.its.ac.id/id/eprint/117453

Actions (login required)

View Item