Development of a Risk-Based Maintenance Optimisation Model for Critical Rotating Machinery in Nigerian Liquefied Natural Gas Processing Trains

The maintenance of critical rotating machinery in Nigerian liquefied natural gas processing trains, including main refrigerant compressors, expanders, and booster pumps, demands optimisation frameworks that quantitatively balance failure risk consequences with maintenance intervention costs and availability impacts over the long-term operational horizon, a level of theoretical and computational sophistication not yet systematically applied in the Nigerian LNG sector despite its direct implications for safe and profitable operation. This research develops a risk-based maintenance optimisation model for critical rotating machinery in LNG processing trains, integrating stochastic equipment reliability modelling, consequence severity assessment from process hazard analysis, maintenance cost optimisation, and operational availability maximisation within a coherent theoretical decision framework. The model employs a continuous-time Markov chain reliability model parameterised from failure mode and effects analysis data and operational experience records from Nigeria LNG operations, capturing multi-state degradation including incipient degradation, minor fault, major fault, and failure states. A quantitative risk assessment module computes risk exposure as the product of failure probability and consequence severity, with consequence severity quantified through process simulation of loss of containment scenarios and economic impact modelling. Multi-period dynamic programming is applied to derive optimal maintenance decision policies over a 20-year operational horizon under both deterministic and stochastic cost scenarios. Monte Carlo simulation quantifies the sensitivity of the optimal policy to parameter uncertainty. Case study application to three representative rotating machine trains demonstrates the framework's ability to reduce total risk-adjusted cost by 19 percent relative to interval-based maintenance policies while improving availability by 1.4 percentage points. Keywords: risk-based maintenance, rotating machinery, LNG processing, Markov model, maintenance optimisation Nigeria.