Manurung, Dina Magdalena (2025) Prediksi Harga Cabai Merah Keriting di Kota Medan Menggunakan Model Hybrid ARIMAX ANN. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
21083010117-cover.pdf Download (1MB) |
|
Text (Bab 1)
21083010117-bab1.pdf Download (287kB) |
|
Text (Bab 2)
21083010117-bab2.pdf Restricted to Registered users only until 19 September 2027. Download (783kB) |
|
Text (Bab 3)
21083010117-bab3.pdf Restricted to Repository staff only until 19 September 2027. Download (549kB) |
|
Text (Bab 4)
21083010117-bab4.pdf Restricted to Repository staff only until 19 September 2027. Download (1MB) |
|
Text (Bab 5)
21083010117-bab5.pdf Download (258kB) |
|
Text (Daftar Pustaka)
21083010117-daftar pustaka.pdf Download (246kB) |
|
Text (Lampiran)
21083010117-lampiran.pdf Restricted to Repository staff only Download (326kB) |
Abstract
The agricultural sector has played an important role in Indonesia's economy, one of which is through the horticulture sub-sector, which has high economic value. Curly red chili is one of the superior horticultural commodities that is in high demand among Indonesians. In addition to being used as a cooking ingredient, curly red chili also has significant health benefits and export value. Medan, as the largest city in North Sumatra, has a high demand for curly red chilies due to its growing population. However, limited local production makes chili prices in this city highly dependent on supplies from outside the region, causing fluctuations. In 2024, chili price fluctuations will have an impact on farmers, traders, and consumers. External factors that influence price fluctuations are precipitation and major holidays. To overcome these price fluctuations, a prediction model is needed that can capture price patterns and exogenous variables. The ARIMAX model is capable of capturing the linear relationship between chili prices and exogenous variables, but it is less effective in recognizing non-linear patterns. Therefore, this study uses the Hybrid ARIMAX-ANN method, in which Artificial Neural Network (ANN) is used to overcome the limitations of ARIMAX. The results show that the Hybrid ARIMAX-ANN model performs better than the single ARIMAX model by reducing prediction errors. The Hybrid model obtained a MAPE value of 5.86% and an RMSE of 3403.80. The price forecast for the next 6 days is estimated to be in the range of IDR 41,000–IDR 44,000 per kilogram, with rainfall affecting price variations, while the absence of seasonal demand keeps prices stable.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer Programming | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Dina Magdalena Manurung | ||||||||||||
| Date Deposited: | 19 Sep 2025 03:28 | ||||||||||||
| Last Modified: | 19 Sep 2025 03:28 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/43797 |
Actions (login required)
 |
View Item |