Pendekatan Machine Learning Terhadap Indikator Fisiologis Manusia Untuk Memprediksi Preferensi Pengguna Egrek (Alat Panen Tandan Buah Segar)

Azzahra, Fadhila Shafa (2025) Pendekatan Machine Learning Terhadap Indikator Fisiologis Manusia Untuk Memprediksi Preferensi Pengguna Egrek (Alat Panen Tandan Buah Segar). Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6010231015-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (3MB) | Request a copy

Abstract

Upaya untuk memprediksi tingkat stres seseorang melalui pengukuran fisiologis telah berlangsung lama. Penelitian-penelitian awal umumnya melibatkan teknik pengukuran yang invasif, seperti pemantauan detak jantung dan tekanan darah secara langsung. Akan tetapi, perkembangan teknologi wearable devices telah memungkinkan pengumpulan data fisiologis secara non-invasif dan lebih objektif bagi penelitian. Penelitian ini bertujuan untuk mengembangkan model prediksi yang dapat membantu petani kelapa sawit dalam memilih alat panen tandan buah segar (TBS) yang tepat, yakni egrek. Penggunaan egrek selalu membutuhkan tenaga yang bahkan jika dilakukan terus menerus dapat menimbulkan potensi bahaya bagi kesehatan. Dengan memanfaatkan teknologi machine learning, penelitian ini menggunakan data fisiologis petani, khususnya detak jantung. Data diperoleh selama petani menggunakan berbagai jenis egrek dalam proses pemanenan. Metode kmeans clustering dipilih sebagai algoritma machine learning karena kemampuannya dalam mengolah data untuk memprediksi kelompoknya. Selain itu model decision tree yang telah dilatih kemudian digunakan untuk mengidentifikasi fitur fisiologis yang paling berpengaruh terhadap preferensi petani terhadap suatu produk egrek. Hasil penelitian diharapkan dapat memberikan rekomendasi yang lebih objektif dalam pemilihan egrek, sehingga dapat meningkatkan efisiensi dan produktivitas petani kelapa sawit.
===================================================================================================================================
Efforts to predict a person's stress level through physiological measurements have been going on for a long time. Early studies generally involved invasive measurement techniques, such as direct monitoring of heart rate and blood pressure. However, the development of wearable device technology has enabled the collection of physiological data non-invasively and more objectively for research. This study aims to develop a prediction model that can help oil palm farmers in choosing the right fresh fruit bunch (FFB) harvesting tool, namely the egrek. The use of egrek always requires energy which, even if done continuously, can pose potential health hazards. By utilizing machine learning technology, this study uses farmers' physiological data, especially heart rate. Data was obtained while farmers were using various types of egrek in the harvesting process. The kmeans clustering method was chosen as the machine learning algorithm because of its ability to process data to predict its groups. In addition, the trained decision tree model was then used to identify the physiological features that most influence farmers' preferences for an egrek product. The results of the study are expected to provide more objective recommendations in selecting egrek, so as to increase the efficiency and productivity of oil palm farmers.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Detak Jantung, Egrek, Fisiologis, Kelapa Sawit, Pertanian, Agriculture, Heartbeat, Kmeans Clustering, Machine Learning, Oil Palm, Physiological
Subjects:	S Agriculture > S Agriculture (General) T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26101-(S2) Master Thesis
Depositing User:	Fadhila Shafa Azzahra
Date Deposited:	04 Feb 2025 10:11
Last Modified:	04 Feb 2025 10:11
URI:	http://repository.its.ac.id/id/eprint/117918

Actions (login required)

View Item