Perancangan Model Mcdm Fuzzy Gaussian Terintegrasi Untuk Evaluasi Penyebab Kecelakaan Fatal Di Industri Migas

Wicaksono, Fermi Dwi (2023) Perancangan Model Mcdm Fuzzy Gaussian Terintegrasi Untuk Evaluasi Penyebab Kecelakaan Fatal Di Industri Migas. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 7032201010-Dissertation.pdf] Text
7032201010-Dissertation.pdf - Accepted Version
Restricted to Repository staff only until 1 April 2025.

Download (8MB) | Request a copy


Industri migas memiliki risiko yang sangat besar terkait dengan operasi yang kompleks, tenaga kerja dan durasi kerja yang besar, serta keterlibatan berbagai disiplin kerja dalam operasi yang tidak rutin. Keadaan ini berpotensi menimbulkan kecelakaan fatal dalam operasi dan prosesnya. Sebagai upaya pencegahan kecelakaan terpadu, faktor penyebab kecelakaan fatal di masa lalu perlu diidentifikasi dan dianalisis. Penelitian ini mengelaborasi signifikansi dan nilai hubungan faktor penyebab kecelakaan fatal. Metodologi terintegrasi dari pemodelan persamaan struktural (SEM), Gaussian fuzzy DEMATEL (Laboratorium Uji Coba dan Evaluasi Pengambilan Keputusan), dan proses jaringan analitik Gaussian fuzzy Monte-Carlo diterapkan untuk menentukan nilai kuantitatif factor penyebab insiden potensial tinggi. Metodologi ini mengatasi keterbatasan evaluasi ketidakpastian dalam pengambilan keputusan Multi-kriteria konvensional (MCDM). Studi kasus dilakukan pada disiplin produksi terkait industri hulu migas, untuk menunjukkan faktor paling signifikan yang menyebabkan kecelakaan fatal. Sebagai bahan masukan dalam penelitian ini, data kecelakaan fatal dikumpulkan dari database International Oil and Gas Producer Association (IOGP’s) dari tahun 2010 hingga 2018. Statistika kecelakaan tersebut menguraikan jenis kecelakaan fatal yang paling sering terjadi selama operasi produksi minyak dan gas. Tiga factor penyebab utama yang dievaluasi: 1. Faktor perilaku pribadi; 2. Faktor kondisi proses yang menyebabkan situasi kerja tidak aman; 3. Disfungsi manajemen organisasi. Metode yang diusulkan, proses hirarki analitik fuzzy Monte-Carlo, memperhitungkan ketidakpastian dan ketidakjelasan dalam menentukan analisis penyebab kecelakaan. Penelitian kuantitatif dan analisis data dilakukan
The oil and gas industry possesses an enormous risk associated with complex operation, large manpower and working duration, and involvement of various working discipline in the un-routine operation. This circumstance potentially leads to the fatal accidents within its operation and process. As the integrated accident prevention measures, past fatal accidents causal factors need to be identified and analyzed. This research elaborates the significance and relationship values of fatal accidents causal factors. The integrated methodology of structural equation modeling (SEM), Gaussian fuzzy DEMATEL (Decision-Making Trial and Evaluation Laboratory), and Gaussian fuzzy Monte-Carlo analytic network process is applied to determine the High potential incident causal factors’ quantified values. This methodology overcome the limitation of uncertainty evaluation in conventional multi-criteria decision making (MCDM). The case study is performed at production concerned disciplines of upstream oil and gas industry, to demonstrate the most significant factor leading to the fatal accidents. As this research’s input, fatal accidents data are gathered from the International Oil and Gas Producer Association’s (IOGP’s) database during the year of 2010 until 2018. The accident statistics elaborates the most frequent type of fatal accident occurring during the oil and gas production operation. Three major causal factors in are evaluated: 1. Personal-behavioral factors; 2. Process condition factor leading to unsafe working situation; 3. Organizational management dysfunction. The proposed method, fuzzy Monte-Carlo analytic hierarchy process, takes into account the uncertainty and fuzziness in determining accident causes analysis. The quantitative research and data analysis are performed for the proposed integrated methodology to measure the evaluation’s robustness, accuracy, and uncertainty handling capability. The data analysis includes statistical testing to validate the systematic literature review that is a baseline of this research. This research discloses that Personal-behavioral factor comes as the first significant causal factors that prompt to fatal accident in the oil and gas industry with the value of 0.3015. As the second place, the organizational management dysfunction lies with the value of 0.3006. And lastly, process condition factor leading to unsafe condition with the significance value of 0.2956. In terms of risky operation, Production operations is considered as the most high-risk activity with the value of 0.136. In addition, based on the result of sensitivity analysis, it is confirmed that the causes factor alternatives’ rank does not change in spite of the working activity-attributes’ weight change for all judgments. Therefore, the evaluation of this research is able to a give consistent guidance to the determination of fatal accident causes factors. Furthermore, this research provides recommendations for organizational and managerial implications to the concrete implementations in oil and gas industry’s operations. These recommendations shall be an ultimate practical guidance for the oil and gas organizations’ accident preventive measures.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: Gaussian fuzzy Monte-Carlo analytic network process; Gaussian fuzzy DEMATEL; Fatal accident causes analysis; Structural equation modeling Gaussian fuzzy analytic network process; Gaussian fuzzy DEMATEL; Analisis penyebab kecelakaan fatal; Structural equation modeling
Subjects: T Technology > T Technology (General)
T Technology > T Technology (General) > T55 Industrial Safety
T Technology > T Technology (General) > T58.62 Decision support systems
Divisions: Interdisciplinary School of Management and Technology (SIMT) > Doctor of Technology Management (DMT)
Depositing User: Fermi Dwi Wicaksono
Date Deposited: 14 Mar 2023 01:15
Last Modified: 14 Mar 2023 01:15

Actions (login required)

View Item View Item