The Virtual Corrosion Engineer (VCE)(1) platform in Shell uses data from online monitoring of industrial plant process conditions as the input to the best available models for the damage mechanisms applicable to the equipment or components of interest. The primary output is then the corrosion rate or susceptibility to failure, displayed on a continuously updated dashboard in real time. However, for furnaces or heat exchangers with hundreds or thousands of tubes, most components of the input data are properly considered as distributions rather than single point value estimates, and hence the remnant life of the tubes is also best represented as a distribution. This paper shows how Monte-Carlo simulation techniques can be used to address this problem and improve the quality of decision making, using practical examples related to creep damage of furnace tubes in service.