Winaldi , Putra Jaya (2025) MULTIVARIATE TIMESERIES FORECASTING HARGA ASPAL DENGAN ALGORITMA GATED RECURRENT UNIT (GRU). FAKULTAS TEKNIK, UNIVERSITAS LAMPUNG.
|
Text
1. ABSTRAK - Winaldi Putra Jaya.pdf Download (4MB) | Preview |
|
![[img]](http://digilib.unila.ac.id/style/images/fileicons/text.png) |
Text
2. SKRIPSI FULL - Winaldi Putra Jaya.pdf Restricted to Repository staff only Download (4MB) | Request a copy |
|
|
Text
3. SKRIPSI TANPA PEMBAHASAN - Winaldi Putra Jaya.pdf Download (4MB) | Preview |
Abstract
Fluktuasi harga minyak global menyebabkan volatilitas harga aspal yang tinggi dan ketidakpastian dalam perencanaan bisnis. Penelitian ini bertujuan mengembangkan model multivariate forecasting harga aspal menggunakan algoritma Gated Recurrent Unit (GRU). Metodologi yang digunakan mencakup tahapan Business Understanding, Data Understanding, Data Preprocessing, Data Modelling, Evalution, dan Deployment. Data yang digunakan mencakup harga aspal dan harga minyak global periode 2016–2023, dengan jumlah data 371 baris dan 3 kolom aspal low dan high, serta minyak (close). Pada tahap eksplorasi, variabel harga minyak dianalisis sebagai prediktor utama perubahan harga aspal. Hasil evaluasi model menunjukkan GRU mampu memprediksi tren harga dengan sangat baik, dengan memiliki rata-rata MAE 6,2441, RMSE 8,2880, dan R- Squared 96,05%. Model GRU terbukti mampu menangkap dan mengenali pola deret waktu. Model ini diimplementasikan dalam dashboard interaktif berbasis Power BI untuk memudahkan visualisasi tren data dan pengambilan keputusan strategis secara cepat dan efisien. Kata kunci: Harga Aspal, Forecasting, GRU Global oil price fluctuations cause high asphalt price volatility and uncertainty in business planning. This study aims to develop a multivariate asphalt price forecasting model using the Gated Recurrent Unit (GRU) algorithm. The methodology used includes the stages of Business Understanding, Data Understanding, Data Preprocessing, Data Modeling, Evaluation, and Deployment. The data used includes global asphalt and oil prices for the period 2016–2023, with a total of 371 rows and 3 columns of low and high asphalt, and oil (close). In the exploration stage, the oil price variable is analyzed as the main predictor of asphalt price changes. The results of the model evaluation show that GRU is able to predict price trends very well, with an average MAE of 6.2441, RMSE of 8.2880, and R- Squared of 96.05%. The GRU model is proven to be able to capture and recognize time series patterns. This model is implemented in an interactive dashboard based on Power BI to facilitate data trend visualization and strategic decision making quickly and efficiently. Keywords: Asphalt Price, Forecasting, GRU
| Item Type: | Other |
|---|---|
| Subjects: | Universitas Lampung > Fakultas Teknik > Prodi Teknik Informatika |
| Divisions: | Fakultas Teknik > Prodi Teknik Informatika |
| Depositing User: | UPT . Siswanti |
| Date Deposited: | 29 Oct 2025 08:00 |
| Last Modified: | 29 Oct 2025 08:00 |
| URI: | http://digilib.unila.ac.id/id/eprint/92327 |
Actions (login required)
 |
View Item |