PRISMA-Based Systematic Literature Review of Reliability-Centered Turnaround Planning for Reducing Downtime in Power Generation Facilities

Abstract

This study investigates how reliability-centered turnaround planning and complementary digital, operational, and energy-performance capabilities contribute to reducing maintenance-related downtime in power-generation facilities. The research addresses the problem that conventional turnaround planning often relies on predetermined schedules and historical maintenance routines that may insufficiently account for asset criticality, real-time equipment condition, changing grid requirements, predictive information, and energy-performance deterioration, creating risks of prolonged outages, inefficient resource allocation, delayed restart, and recurring post-maintenance failures. The purpose is to determine the individual and combined effects of Reliability-Centered Turnaround Planning, Smart Manufacturing and Predictive Maintenance Integration, Grid Integration and Operational Coordination, and Energy-Efficiency Technology Integration on Turnaround Downtime Reduction Performance. A quantitative, explanatory, cross-sectional, multiple-case-study-based design was employed across selected power-generation facility cases, including thermal and coal-fired, gas and combined-cycle, hydroelectric, and renewable or hybrid generation environments. The purposive sample comprised maintenance and reliability engineers, operations personnel, electrical and mechanical engineers, turnaround and outage managers, predictive-maintenance specialists, and grid and energy-efficiency professionals. Of 320 questionnaires distributed, 303 were returned and 291 usable responses were retained, producing a valid response rate of 90.9%. Reliability was satisfactory, with Cronbach’s alpha values ranging from .84 to .91, KMO = .895, and Bartlett’s Test significant at p < .001. Statistical analysis included descriptive statistics, Pearson correlation, multiple linear regression, VIF, tolerance, Durbin-Watson analysis, and residual diagnostics. TDRP recorded M = 4.23, SD = 0.51. SMPMI showed the strongest correlation with TDRP, r = .76, followed by EETI, r = .73, RCTP, r = .71, and GIOC, r = .69, all p < .001. The regression model explained 71.9% of TDRP variance, R² = .719, adjusted R² = .715, F (4,286) = 182.95, p < .001. SMPMI was the strongest predictor, β = .34, followed by RCTP, β = .29, EETI, β = .26, and GIOC, β = .22. The findings support H1 through H5 and imply that power-generation facilities can reduce turnaround downtime more effectively when reliability-centered planning is integrated with predictive digital intelligence, grid-aware coordination, and energy-efficiency technologies.