Power plant megaprojects are characterized by substantial safety risks that arise from high technical complexity, concurrent and sometimes overlapping operations, a wide spectrum of high-hazard activities, and the involvement of multiple contractors and subcontractors operating within shared work environments. These conditions generate a dynamic and often uncertain operational context in which the boundaries of responsibility, communication, and coordination become critical determinants of safety outcomes. In contemporary safety research, sustained safety performance is no longer understood solely as the successful control of hazards or the achievement of favorable lagging indicators. Instead, there is growing recognition that durable safety performance is associated with the proactive use of leading indicators, the capacity for organizational learning, and the quality of safety governance that shapes decision-making across the project lifecycle. This conceptual shift from reactive control to proactive and systemic management has important implications for large-scale energy infrastructure projects.

This study analyzed the key success factors associated with achieving 53 million fatality-free work hours in the power plant megaprojects of MAPNA Development 3. The primary aim was to explain the organizational and managerial mechanisms associated with sustained safety performance rather than merely describe the reported record. Accordingly, the study sought to identify which safety-related activities and governance arrangements were most closely related to the observed stability of safety outcomes over an extended period. This emphasis reflects the need to move beyond descriptive reporting and toward an analytical understanding of how megaprojects maintain safe performance over many years of operation.

Methodologically, this applied study adopted a descriptive-analytical approach, a longitudinal time frame, and a case-study strategy. The descriptive component provided a systematic account of the evolution of safety indicators over time, while the analytical component examined the statistical relationships among key safety management variables. The longitudinal dimension enabled the tracking of changes and trends across consecutive operational years, and the case-study strategy allowed an in-depth examination of the phenomenon within its real organizational and operational context.

Registry data from 27 projects were collected through a census of official HSE repositories, covering the period from the beginning of 2012 to the beginning of 2025. This period corresponds to 13 complete operational years from 2012 to 2024. No sampling was conducted; instead, a full enumeration of available project records was employed to ensure comprehensive coverage. The collected data included official HSE reports, daily, monthly, and annual records, meeting minutes of technical safety and occupational health committees, near-miss reports, training records, and work-permit documentation. To enhance the reliability of the records, the data were cross-checked against relevant organizational documents and discrepancies were reconciled.

A total of 58 project-year observations were analyzed using descriptive statistics and Spearman’s rank-order correlation, which was selected as an appropriate nonparametric technique given the ordinal and non-normally distributed nature of the registry data. All analyses were performed in the Python environment. During the study period, 91 occupational injuries resulting in at least three days away from work were recorded. This corresponds to a cumulative frequency rate of 1.72 injuries per million person-hours, a value that provides a meaningful basis for comparison with international safety benchmarks.

The analysis revealed significant positive correlations among several leading indicators and safety management activities. A strong correlation was observed between safety training and near-miss reporting (ρ=0.826, p<0.001), indicating that increased training activity was associated with higher reporting of near-miss events. Safety training was also significantly correlated with work-permit issuance (ρ=0.738, p<0.001). In addition, preventive actions were positively correlated with near-miss reporting (ρ=0.706, p<0.001) and with work-permit issuance (ρ=0.561, p<0.001). These findings suggest that proactive safety management activities tend to operate in a mutually reinforcing manner.

The findings indicate that sustained fatality-free performance was associated with safety governance, the use of leading indicators, organizational learning, and the management of interface risks. Safety governance appeared to provide the structural and decision-making framework within which training, prevention, reporting, and control activities could function coherently. Leading indicators served as early signals of system performance and supported the detection of emerging issues before they escalated into serious outcomes. Organizational learning enabled the continuous improvement of safety practices based on accumulated experience. The management of interface risks, particularly at the boundaries between contractors, units, and activities, was identified as a critical mechanism for preventing the escalation of latent hazards in complex and concurrent operations.

Although the results provide valuable insight, it is important to acknowledge the observational nature of the data. The study is based on registry records and does not permit causal inference. The reported correlations describe associations rather than causal relationships. Future research could employ complementary qualitative methods to explore the underlying mechanisms in greater depth and could extend the analysis to other megaprojects to enhance the generalizability of the findings.

Strengthening the links between HSE data, decision-making processes, and organizational learning may support comparable megaprojects in achieving and sustaining safe performance. The integration of these elements within a coherent governance framework appears to be a promising direction for both research and practice in the safety management of large-scale energy infrastructure.