Citation

Ehfidha, Ashraf M.A. and Thiagarajah, Siva Priya (2025) Improved Operational Hazard Risk Management for Oil and Gas Automation Systems with Advanced Voting Logic. In: 2025 Multimedia University Engineering Conference, MECON 2025, 21 July 2025 - 23 July 2025, Cyberjaya, Malaysia.

Text 12.pdf - Published Version Restricted to Repository staff only Download (903kB)

Abstract

Fire and gas detection systems play an important role in ensuring safety in high-risk environments such as the oil and gas industry. However, existing systems suffer from high false alarm rates and limited reliability, which impacts operational efficiency. This study presents an enhanced detection framework that integrates a voting logic of 2-out- of-2 (2oo2) and 2-out-of-3 (2oo3) and calibrated thresholds for Hydrogen Sulfide (H₂S) gas, flame, and heat-sensitive cable systems. The system was validated through simulation-based testing. Results showed an 18% reduction in false alarm detections for H₂S gas detection, and improved H₂S detection efficiency from 72% to 93%, leading to a 58% decrease in unnecessary shutdowns, and reduced shutdown occurrences by 29.17%. The flame detection system achieved 83% accuracy, while the heat-sensitive cable system improved to 82% accuracy. A risk assessment done on the proposed system using a 5×5 risk matrix classified all the systems as low risk. This research established a scalable framework for future integration of adaptive voting and IoT technologies, enhancing safety, increasing reliability and reducing system shutdowns which lead to poor productivity in industrial settings.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Human-machine interface (HMI), hazard risk management
Subjects:	T Technology > TP Chemical technology > TP670-699 Oils, fats, and waxes
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	18 Mar 2026 08:23
Last Modified:	19 Mar 2026 02:27
URII:	http://shdl.mmu.edu.my/id/eprint/15592

Downloads

Edit (login required)