Sistem Klasifikasi Kondisi Lingkungan Kandang Bebek Pedaging Menggunakan Metode Decision Tree untuk Pengendalian Blower dan Alarm

Nuswantari, Ifda Leoni Devi (2026) Sistem Klasifikasi Kondisi Lingkungan Kandang Bebek Pedaging Menggunakan Metode Decision Tree untuk Pengendalian Blower dan Alarm. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2040221022-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Kondisi lingkungan kandang bebek pedaging dipengaruhi oleh konsentrasi gas amonia (NH₃), suhu, dan kelembapan relatif. Pemantauan yang masih dilakukan secara manual menyebabkan perubahan kondisi kandang tidak selalu dapat diketahui dengan cepat sehingga berpotensi memperlambat penanganan. Penelitian ini bertujuan merancang dan mengimplementasikan sistem klasifikasi kondisi lingkungan kandang bebek pedaging menggunakan metode Decision Tree Classification and Regression Tree (CART) pada mikrokontroler ESP32-S3. Dataset yang digunakan terdiri atas 4.820 data parameter konsentrasi gas amonia, suhu, dan kelembapan relatif yang dibagi menjadi data pelatihan dan data pengujian dengan rasio 80:20. Data pelatihan diseimbangkan menggunakan metode Random Under Sampling (RUS), kemudian model dibangun menggunakan kriteria Gini Index untuk menghasilkan klasifikasi kondisi lingkungan ke dalam kategori Aman, Waspada, dan Bahaya. Hasil pengujian menunjukkan model memperoleh Accuracy sebesar 97,76%, sedangkan nilai precision, recall, dan F1-score masing-masing sebesar 0,98. Pengujian menggunakan 5-Fold Cross Validation menghasilkan rata-rata Accuracy sebesar 98,83% dengan standar deviasi 0,00261, yang menunjukkan bahwa model memiliki performa yang konsisten dan stabil. Hasil implementasi menunjukkan sistem mampu mengklasifikasikan kondisi lingkungan secara otomatis serta mengendalikan blower pada kondisi Waspada dan blower serta speaker pada kondisi Bahaya sesuai hasil klasifikasi. Dengan demikian, metode CART berhasil diterapkan sebagai dasar pengambilan keputusan pada sistem otomatis untuk mitigasi paparan gas amonia pada kandang bebek pedaging.
===================================================================================================================================
The environmental conditions in broiler duck houses are influenced by ammonia (NH₃) concentration, temperature, and relative humidity. Since monitoring is still done manually, changes in house conditions are not always detected quickly, which can potentially delay corrective action. This study aims to design and implement a system for classifying the environmental conditions of broiler duck houses using the Decision Tree Classification and Regression Tree (CART) method on an ESP32-S3 microcontroller. The dataset consists of 4,820 data points on ammonia gas concentration, temperature, and relative humidity, divided into training and testing data in an 80:20 ratio. The training data was balanced using the Random Under-Sampling (RUS) method, and the model was built using the Gini Index criterion to classify environmental conditions into the categories Safe, Caution, and Danger. The test results showed that the model achieved an Accuracy of 97.76%, while the precision, recall, and F1-score values were all 0.98. Testing using 5-Fold Cross Validation yielded an average Accuracy of 98.83% with a standard deviation of 0.00261, indicating that the model performs consistently and stably. Implementation results show that the system is capable of automatically classifying environmental conditions and controlling the blower under “Caution” conditions and both the blower and speaker under “Danger” conditions according to the classification results. Thus, the CART method was successfully applied as the basis for decision-making in an automated system for mitigating ammonia gas exposure in broiler duck houses.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Decision Tree CART, ESP32-S3, Gas amonia, Kandang bebek pedaging, Klasifikasi, CART Decision Tree, ESP32-S3, Ammonia Gas, Broiler Duck Pens, Classification.
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.62 Decision support systems T Technology > T Technology (General) > T59.7 Human-machine systems. T Technology > TD Environmental technology. Sanitary engineering > TD883 Air quality management. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Ifda Leoni Devi Nuswantari
Date Deposited:	04 Aug 2026 01:44
Last Modified:	04 Aug 2026 01:44
URI:	http://repository.its.ac.id/id/eprint/142691

Actions (login required)

View Item