Workload Analysis In The Oil And Gas Industry To Detect Fatigue Using Machine Learning Model

Kartoyudo, Wildan Husein (2025) Workload Analysis In The Oil And Gas Industry To Detect Fatigue Using Machine Learning Model. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5010211131-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

The oil and gas industry is regarded as one of the most demanding work environments, with high safety risks and heavy workloads, particularly among field personnel. The goal of this project is to monitor worker workload and detect probable weariness by combining subjective and physiological evaluation methods with a machine learning approach. Subjective workload was assessed using the NASA Task Load Index (NASA-TLX), while physiological workload was measured using heart rate monitoring to determine Cardiovascular Load. Data were gathered from 30 field workers with varying operational tasks. The results show that the General Affairs, Operator, and Maintenance departments had the highest workload levels, as evidenced by high NASA-TLX scores (more than 80 in most dimensions) and elevated %CVL values (more than 30%), indicating a strong cardiovascular load. PCA successfully reduced the dataset while retaining critical workload variables, and K-Means clustering classified the workers into three categories of low, moderate, and high, depending on workload intensity. The clustering results show that job intensity, particularly activities that require both physical and cognitive effort, is highly associated with fatigue risk. This study provides a data-driven paradigm for workload evaluation that may be utilized to optimize task distribution, eliminate fatigue-related risks, and increase operational safety and efficiency in the oil and gas industry.
====================================================================================================================================
Industri minyak dan gas dianggap sebagai salah satu lingkungan kerja yang paling menuntut, dengan risiko keselamatan tinggi dan beban kerja berat, terutama di antara personel lapangan. Tujuan dari proyek ini adalah untuk memantau beban kerja pekerja dan mendeteksi kemungkinan kelelahan dengan menggabungkan metode evaluasi subjektif dan fisiologis dengan pendekatan pembelajaran mesin. Beban kerja subjektif dinilai menggunakan Indeks Beban Tugas NASA (NASA-TLX), sementara beban kerja fisiologis diukur menggunakan pemantauan detak jantung untuk menentukan Beban Kardiovaskular. Data dikumpulkan dari 30 pekerja lapangan dengan berbagai tugas operasional. Hasilnya menunjukkan bahwa departemen Urusan Umum, Operator, dan Pemeliharaan memiliki tingkat beban kerja tertinggi, sebagaimana dibuktikan oleh skor NASA-TLX yang tinggi (lebih dari 80 di sebagian besar dimensi) dan nilai %CVL yang tinggi (lebih dari 30%), yang menunjukkan beban kardiovaskular yang kuat. PCA berhasil mengurangi kumpulan data sambil mempertahankan variabel beban kerja yang kritis, dan pengelompokan K-Means mengklasifikasikan pekerja ke dalam tiga kategori rendah, sedang, dan tinggi, tergantung pada intensitas beban kerja. Hasil pengelompokan menunjukkan bahwa intensitas kerja, terutama aktivitas yang membutuhkan upaya fisik dan kognitif, sangat berkaitan dengan risiko kelelahan. Studi ini menyediakan paradigma berbasis data untuk evaluasi beban kerja yang dapat dimanfaatkan untuk mengoptimalkan distribusi tugas, menghilangkan risiko terkait kelelahan, dan meningkatkan keselamatan serta efisiensi operasional di industri minyak dan gas.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Workload Analysis, NASA TLX, Machine Learning, Cardiovascular Load, Fatigue.
Subjects:	T Technology > T Technology (General) > T385 Visualization--Technique T Technology > T Technology (General) > T57.84 Heuristic algorithms. T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > T Technology (General) > T58.6 Management information systems T Technology > T Technology (General) > T59.7 Human-machine systems.
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26201-(S1) Undergraduate Thesis
Depositing User:	Wildan Husein Kartoyudo
Date Deposited:	30 Jul 2025 07:47
Last Modified:	30 Jul 2025 07:47
URI:	http://repository.its.ac.id/id/eprint/123998

Actions (login required)

View Item