O.A. Fapohunda: Conceptualisation, Methodology, Investigation, Data Curation, Formal Analysis, Writing - Original Draft, Writing - Review & Editing.C.I. Ajuwa: Methodology, Validation, Writing - Review & Editing, Supervision.O.D. Samuel: Methodology, Validation, Writing - Review & Editing, Supervision. All authors read and approved the final manuscript

ABSTRACT Background: Corrective maintenance rework is recurring but poorly quantified in LNG, undermining RAM performance despite three decades of RCM scholarship.Methods: A retrospective mixed-methods design synthesised 442,701 closed CM Work Orders across 25 annual cohorts (2000–2024) from a major LNG facility, triangulated with a four-round Delphi inquiry of 4,329 stakeholder respondents. Random-effects meta-analysis pooled annual 90-day recurrence incidence using Restricted Maximum Likelihood estimation. Mann–Kendall and Sen&apos;s slope quantified trend. Joinpoint regression, interrupted time-series and Autoregressive Integrated Moving Average decomposed structural breaks and serial dependence. Cochran&apos;s Q, I², τ² and univariable random-effects meta-regression attributed heterogeneity. A two-layer Predictive Rework-Risk Framework stratified work orders into Low, Moderate, High and Critical risk bands under rolling-origin temporal validation, evaluated by AUROC curve, calibration slope, Brier score and Decision Curve Analysis. Reporting follows PRISMA 2020 and ISO 14224 / 55000 / IEC 60300.Results of this study showed that there was a decline in Annual incidence from 24.09% to 8.55% (absolute reduction 15.54 pp; 64.5% relative). Sen’s slope −0.442 pp/year. The pooled incidence was 14.96% (95% CI 13.58–16.46%; PI 8.76–24.38%); I² = 99.29%; τ² = 0.0827. Joinpoint identified 2017 as the breakpoint year; pre-break slope of −0.566 pp/year turned +0.154 pp/year after breakpoint.. Delphi agreement was strong (Kendall&apos;s W = 0.463, p < 0.001). The AUROC of PRRF is 0.8643 and its recall is 0.9118 at threshold of 0.16.Conclusions: The framework offers a transferable governance instrument linking predicted rework risk directly to pre-closure intervention in high-reliability process industries.