Pramanda, Ryan (2026) Model Ergo-Safety Untuk Keselamatan Kerja Proyek Konstruksi Menggunakan Bayesian Belief Network (Studi Kasus: Gedung Bertingkat Tinggi). Doctoral thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 7010221002_Doctoral.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
7010221002_Doctoral.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy |
Abstract
Penerapan Keselamatan dan Kesehatan Kerja (K3) merupakan aspek krusial dalam proyek konstruksi gedung bertingkat yang memiliki kompleksitas dan risiko kecelakaan kerja yang tinggi. Data BPJS Ketenagakerjaan menunjukkan peningkatan kecelakaan kerja di Indonesia selama periode 2017–2023 dengan rata-rata pertumbuhan sebesar 14,29%, sehingga diperlukan pendekatan analisis keselamatan yang lebih sistematis, prediktif, dan berbasis data. Penelitian ini bertujuan mengembangkan dan menguji model ergo-safety untuk menganalisis serta memprediksi keselamatan kerja pada proyek konstruksi gedung bertingkat menggunakan pendekatan Bayesian Belief Network (BBN) melalui integrasi kerangka Human Interface System (HIS) dan Human Factors Analysis and Classification System (HFACS). Model ergo-safety dikonstruksi menggunakan 15 variabel keselamatan kerja dengan 33 hubungan kausal disusun berdasarkan sintesis literatur dan validasi pakar, termasuk variabel sistem pembayaran yang masih relatif jarang dikaji dalam konteks keselamatan konstruksi. Penelitian ini melibatkan 28 ahli yang menilai kondisi kerja 277 pekerja pada proyek konstruksi gedung bertingkat PT Adhi Karya CW1 ITS Surabaya. Inferensi probabilistik menggunakan Gaussian Linear Bayesian Belief Network menunjukkan bahwa seluruh variabel memenuhi kriteria validitas probabilistik dengan nilai probabilitas marginal yang stabil. Hasil analisis menunjukkan bahwa dari 33 hubungan kausal, terdapat 20 hubungan positif dan 13 hubungan negatif yang merefleksikan mekanisme risiko dalam sistem keselamatan kerja berdasarkan struktur hubungan antarvariabel. Analisis koefisien menunjukkan bahwa pelatihan (X7) berpengaruh positif kuat terhadap sistem pembayaran (X14) dengan koefisien 0,835, sedangkan manajemen dan supervisi (X6) berpengaruh negatif lemah dengan koefisien −0,382. Sistem pembayaran (X14) juga berpengaruh negatif lemah terhadap kecelakaan kerja (X15) dengan koefisien −0,220. Kondisi fisik dan mental pekerja (X10) teridentifikasi sebagai faktor paling krusial dalam pengendalian risiko kecelakaan dan peningkatan probabilitas pencapaian Zero Accident sebagai outcome keselamatan kerja. Secara praktis, model ergo-safety berbasis BBN ini dapat dimanfaatkan sebagai alat analisis dan monitoring keselamatan kerja untuk mendukung perancangan strategi pencegahan kecelakaan yang lebih efektif dan kontekstual pada proyek konstruksi gedung bertingkat di Indonesia.
====================================================================================================================================
The implementation of Occupational Safety and Health (OSH) is a critical aspect of high-rise building construction projects, which are characterised by high levels of complexity and accident risk. Data from BPJS Ketenagakerjaan indicate a sustained increase in occupational accidents in Indonesia during the period 2017–2023, with an average annual growth rate of 14.29%, highlighting the need for a more systematic, predictive, and data-driven safety analysis approach. This study aims to develop and validate an ergo-safety model to analyse and predict occupational safety performance in high-rise construction projects using a Bayesian Belief Network (BBN) through the integration of the Human Interface System (HIS) and the Human Factors Analysis and Classification System (HFACS). The proposed ergo-safety model comprises 15 occupational safety variables connected by 33 causal relationships derived from a synthesis of the literature and expert validation, including a pay system variable that remains relatively underexplored in construction safety research. The study involved 28 experts who assessed the working conditions of 277 workers on a high-rise construction project at PT Adhi Karya CW1 ITS Surabaya. Probabilistic inference using a Gaussian Linear BBN confirmed that all variables satisfied probabilistic validity criteria, with stable marginal probability values. The results show that, of the 33 causal relationships, 20 are positive and 13 are negative, reflecting the underlying risk mechanisms embedded within the occupational safety system as revealed by the BBN structure. Coefficient analysis indicates that training (X7) has a strong positive influence on the pay system (X14) with a coefficient of 0.835, whereas management and supervision (X6) exert a weak negative influence (−0.382). The pay system (X14) also shows a weak negative effect on occupational accidents (X15), with a coefficient of −0.220. Furthermore, workers’ physical and mental condition (X10) is identified as the most critical factor in accident risk control and in increasing the probability of achieving Zero Accident as the safety outcome. Practically, this BBN-based ergo-safety model can be utilised as an analytical and monitoring tool to support the development of more effective and context-specific accident prevention strategies for high-rise construction projects in Indonesia.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Uncontrolled Keywords: | Faktor Manusia, K3, Ergo-Safety, HIS, HFACS, Proyek Konstruksi, BBN Human Factor, OHS, Ergo-Safety, HIS, HFACS, Construction Projects, BBN |
| Subjects: | T Technology > T Technology (General) > T55 Industrial Safety T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.62 Decision support systems T Technology > TA Engineering (General). Civil engineering (General) > TA169 Reliability (Engineering) |
| Divisions: | Faculty of Industrial Technology > Industrial Engineering > 26001-(S3) PhD Thesis |
| Depositing User: | Ryan Pramanda Pramanda |
| Date Deposited: | 26 Jan 2026 08:27 |
| Last Modified: | 26 Jan 2026 08:27 |
| URI: | http://repository.its.ac.id/id/eprint/130566 |
Actions (login required)
 |
View Item |