Abiansyah, Ghaotama Akbar (2026) IMPLEMENTASI KEBUTUHAN IRIGASI TANAMAN BERBASIS MACHINE LEARNING TERHADAP DATA KONDISI TANAH DAN MIKROKLIMAT PADA SISTEM IOT. S1 thesis, Universitas PGRI Madiun.
This is the latest version of this item.
![[img]](https://eprint.unipma.ac.id/style/images/fileicons/text.png) |
Text
HALAMAN DEPAN GHAOTAMA.pdf Download (604kB) |
|
Text
ABSTRAK GHAOTAMA.pdf Download (249kB) |
|
Text
BAB I GHAOTAMA.pdf Download (295kB) |
|
Text
BAB II GHAOTAMA.pdf Download (826kB) |
|
Text
BAB III GHAOTAMA.pdf Restricted to Repository staff only Download (476kB) |
|
Text
BAB IV GHAOTAMA.pdf Restricted to Repository staff only Download (1MB) |
|
Text
BAB V GHAOTAMA.pdf Restricted to Repository staff only Download (281kB) |
|
Text
DAFTAR PUSTAKA GHAOTAMA.pdf Download (270kB) |
|
Text
LAMPIRAN GHAOTAMA.pdf Restricted to Repository staff only Download (4MB) |
Abstract
Conventional plant watering, which still relies on farmers’ habits without considering soil and environmental conditions, leads to inefficient water use. This study aims to design and develop an automated irrigation system based on the Internet of Things (IoT) and machine learning that is capable of predicting plant water requirements based on soil and microclimate conditions in real time. The system was developed using the Prototype method, which consists of five stages: Communication, Quick Plan, Modeling Quick Design, Construction of Prototype, and Deployment Delivery & Feedback. The IoT device was built using an ESP8266 NodeMCU microcontroller that reads air temperature and humidity, soil temperature, and soil moisture data, and controls a water pump through a relay. Irrigation requirement prediction was carried out using the Decision Tree algorithm, trained on the secondary Stuard Tomato Dataset, then deployed on a FastAPI server hosted on Hugging Face Space and accessed by the IoT device through an API. The test results show that the machine learning model achieved an accuracy of 96.72%, a precision of 83.33% for the ON class, a recall of 75.00% for the ON class, an F1 Score of 78.95% for the ON class, and an AUC value of 0.8788, which falls into the good category. Functional system testing using the black box technique on all test scenarios produced valid results, including a threshold-based fallback mechanism that maintains irrigation continuity when the server connection is disrupted. Therefore, the developed system is proven capable of supporting automated and efficient irrigation decision-making based on soil and microclimate conditions.
| Item Type: | Thesis/Skripsi/Tugas Akhir (S1) |
|---|---|
| Kata Kunci: | Irigasi Otomatis; Internet of Things; Machine Learning; Decision Tree; Mikroklimat |
| Subjects: | S Agriculture > S Agriculture (General) T Technology > T Technology (General) |
| Divisions: | Fakultas Teknik > Teknik Informatika |
| Depositing User: | ABIANSYAH AKBAR GHAOTAMA |
| Date Deposited: | 14 Oct 2026 02:24 |
| Last Modified: | 14 Oct 2026 02:24 |
| URI: | http://eprint.unipma.ac.id/id/eprint/10077 |
Available Versions of this Item
-
IMPLEMENTASI KEBUTUHAN IRIGASI TANAMAN BERBASIS MACHINE LEARNING TERHADAP DATA KONDISI TANAH DAN MIKROKLIMAT PADA SISTEM IOT. (deposited UNSPECIFIED)
- IMPLEMENTASI KEBUTUHAN IRIGASI TANAMAN BERBASIS MACHINE LEARNING TERHADAP DATA KONDISI TANAH DAN MIKROKLIMAT PADA SISTEM IOT. (deposited 14 Oct 2026 02:24) [Currently Displayed]
Actions (login required)
 |
View Item |
