Mechanical Integrity as a Pillar of Process Safety Management: An RCM Approach to Preventing Loss of Containment in Aging Oil, Gas, and Power Generation Infrastructure
Keywords:
Mechanical Integrity, Reliability-Centered Maintenance, Loss-of-Containment Prevention, Process Safety Management, OT/ICS SecurityAbstract
This study examines mechanical integrity as a central pillar of process safety management by evaluating how integrated reliability, monitoring, material-assessment, and cyber-physical capabilities contribute to Loss-of-Containment Prevention Performance in aging oil, gas, pipeline, petrochemical, and power-generation infrastructure. The research addresses the problem that equipment degradation, incomplete condition information, delayed release detection, fragmented maintenance and inspection practices, and vulnerabilities in OT/ICS environments can collectively weaken containment reliability, while limited empirical evidence exists regarding how these capabilities operate together. The purpose is to determine the individual and combined effects of Reliability-Centered Mechanical Integrity, Hazardous Release Detection and Characterization Capability, Material Property Characterization and Lifecycle Assessment, and OT/ICS Supply Chain Security Resilience on Loss-of-Containment Prevention Performance. A quantitative, cross-sectional, case-study-based design was employed across multiple aging industrial facility and asset cases involving pipelines, pressure vessels, storage systems, pumps, compressors, boilers, turbines, valves, and associated monitoring and control infrastructure. The purposive sample comprised reliability and mechanical-integrity engineers, maintenance and inspection specialists, corrosion and materials engineers, process and process-safety professionals, operations personnel, instrumentation and control engineers, OT/ICS specialists, asset-integrity managers, and technical supervisors. Of 320 questionnaires distributed, 302 were returned and 291 valid responses were retained, yielding a 90.9% usable response rate. Reliability coefficients ranged from .86 to .92, KMO = .901, and Bartlett’s test was significant at p < .001. LOCPP recorded M = 4.22, SD = 0.50, followed by RCMI, M = 4.19, SD = 0.54; HRDCC, M = 4.16, SD = 0.56; MPCLA, M = 4.13, SD = 0.58; and OICSSR, M = 4.10, SD = 0.60. Correlations with LOCPP were RCMI, r = .77; HRDCC, r = .74; MPCLA, r = .71; and OICSSR, r = .68, all p < .001. The regression model explained 72.4% of LOCPP variance, R² = .724, adjusted R² = .720, F (4,286) = 187.59, p < .001. RCMI emerged as the strongest predictor, β = .34, followed by HRDCC, β = .29, MPCLA, β = .24, and OICSSR, β = .20. The findings support H1 through H5 and imply that stronger loss-of-containment prevention depends on coordinated mechanical reliability, timely release detection, lifecycle degradation assessment, and resilient industrial control systems rather than isolated technical interventions.


