Model Pengukuran Safety Performance Pada Proyek Konstruksi Bendungan Berbasis Bayesian Belief Networks

Asrar, Muhammad Nabil (2021) Model Pengukuran Safety Performance Pada Proyek Konstruksi Bendungan Berbasis Bayesian Belief Networks. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 03111950030013_Master_Thesis.pdf] Text
03111950030013_Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (5MB) | Request a copy

Abstract

Proyek Konstruksi rentan terhadap risiko kecelakaan kerja karena sifatnya yang dinamis dan kompleks.Selain itu, sumber daya manusia juga masih menjadi sumber utama dalam proses pelaksanaan konstruksi. Oleh karena itu, Safety Performance menjadi suatu keharusan dalam pelaksanaan proyek konstruksi.Tujuan dari penelitian ini adalah mengusulkan model probabilistik untuk mengukur Safety Performance pada proyek konstruksi bendungan berbasis Bayesian Belief Network (BBN). BBN dapat digunakan untuk mengakomodasi ketidakpastian (risiko) yang dapat menyebabkan kecelakaan kerja.Variabel risk event,source of risk dan conditional probability table (CPT) dalam membuat permodelan Safety Performance diperoleh dari studi literatur,data sekunder dan wawancara terstruktur dengan pakar, baik dari akademisi maupun praktisi konstruksi. Analisa sensitivitas juga digunakan untuk melihat variabel yang paling dominan mempengaruhi Safety Performance. Hasil penelitian ini menunjukan bahwa model usulan ini dapat digunakan untuk mengukur Safety Performance.Pada pelaksanaan proyek konstruksi bendungan terdapat 12 variabel source of risk dan 7 variabel risk event yang memiliki pengaruh terhadap Safety Performance. Hasil uji validitas model menggunakan test only pada proyek konstruksi Bendungan Sepaku Semoi di Kalimantan Timur menunjukan akurasi 73.86%. Analisa sensitivitas menunjukan bahwa variabel risiko yang terjadi yaitu "terpeleset jatuh dari ketinggian","kejatuhan material dari perancah", dan "tertimbun galian longsoran" merupakan variabel yang sangat berpengaruh terhadap kondisi Safety Performance berdasarkan kejadian kondisi di lapangan. Selain dapat digunakan sebagai early warning system dalam rangka mendukung pengambilan keputusan sehingga dapat meningkatkan keselamatan kerja, khususnya di proyek konstruksi bendungan.
=================================================================================================
Construction projects are Venerable to the risk of work accidents due to their dynamic and complex nature. In addition, human resources are still as the main source in the construction process. Therefore, safety performance becomes a necessity in the implementation of construction projects. The purpose of this study is to propose a probabilistic model measure safety performance in Bayesian belief network-based dam construction project. BBN is used accommodate the uncertainty of risks that can lead to accidents. Variable risk event, source of risk and Conditional probability table (CPT) in making safety performance modeling obtained from literature studies, secondary data and structured interviews with experts, both from academics and construction practitioners. sensitivity analysis is also used to look at the variables that most dominantly affect safety performance. The result of this study shows that this proposed model can be used to measure safety performance. The implementation of dam construction projects included 12
variables source of risk and 7 variables risk event that has an influence on safety performance. The results of the model validity test using test only at the Sepaku Semoi Dam Project in East Kalimantan showed an accuracy of 73.86%. Sensitivity analysis shows that the risk variables that occur are "slipped falling from a height", "Fall of scaffolding material from above" and "Buried avalanche" is a variable that greatly affects the condition of safety performance based on the occurrence of conditions in the field. In addition to being able to be used to measure safety performance, this model can also be used as an Early warning system in order to support decision making therefore it can improve work safety, especially in dam construction projects.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Pengukuran, Safety Performance, Proyek Konstruksi Bendungan, Bayesian Belief Networks, Ketidakpastian Risiko Measurement, Safety Performance, Dam Construction Projects, Bayesian Belief Networks, Risk Uncertainty.
Subjects: Q Science > QA Mathematics > QA336 Artificial Intelligence
Q Science > QA Mathematics > QA9.58 Algorithms
T Technology > TA Engineering (General). Civil engineering (General)
T Technology > TA Engineering (General). Civil engineering (General) > TA169 Reliability (Engineering)
Divisions: Faculty of Civil Engineering and Planning > Civil Engineering > 22101-(S2) Master Thesis
Depositing User: Muhammad Nabil Asrar
Date Deposited: 15 Aug 2021 05:34
Last Modified: 15 Aug 2021 05:34
URI: http://repository.its.ac.id/id/eprint/86757

Actions (login required)

View Item View Item