This paper proposes and compares different techniques for maintenance optimization based on Genetic Algorithms (GA), when the parameters of the maintenance model are affected by uncertainty and the fitness values are represented by Cumulative Distribution Functions (CDFs). The main issues addressed to tackle this problem are the development of a method to rank the uncertain fitness values, and the definition of a novel Pareto dominance concept.The GA-based methods are applied to a practical case study concerning the setting of a condition-based maintenance policy on the degrading nozzles of a gas turbine operated in an energy production plant.

This paper presents the statistical characterization of the oxidation degradation mechanism affecting the nozzles of turbines operated in Oil and Gas utilities. The degradation mechanism is modeled as a four-state, continuous-time semi-Markov process with Weibull distributed transition times. Maximum likelihood estimation is used to infer the parameters of the model from an available set of field data, whereas a numerical approach to estimate the Fisher information matrix is used to characterize the uncertainty in the estimates. The estimates obtained are, then, utilized to compute the probabilities of occupying the four degradation states over time and the corresponding uncertainties. A case study is shown, dealing with real field data