Corrosion management system encompassing the various stages of an asset life starting from design, construction, through to operation and decommissioning remains the key focus in ensuring integrity and safe operation of the asset. Corrosion study is conducted during the initial design phase, followed by multiple reviews during the operational stage as part of the overall corrosion management process. These studies aim to identify all damage mechanisms that can be present, including both non-age-related and age-related mechanisms. Currently in the oil and gas industry, corrosion rate predictions for age-related mechanisms are generated via mathematical equations or correlations as outcome from laboratory testing and analyses which may not be representative of the actual operating condition. These predictions impose limitations with regards to utilizing inputs produced from big data. Application of artificial intelligence to predict corrosion rate offers advantages where real high frequency data streams from IoT sensors are analyzed via machine learning algorithm thus providing prediction based on historical experience of specific asset. Data preprocessing is an important step in machine learning that involves transforming raw data from various parameters so that issues owing to the incompleteness, inconsistency, and/or lack of appropriate representation of trends are resolved to arrive at a data set that is in an understandable format. Feature engineering will then be performed which analyze the parameter correlation to obtain the most suitable combination and the best features and data characteristics. For corrosion rate prediction, the supervised learning algorithm is applicable such as logistic regression, naive bayes, support vector machines, artificial neural networks, and random forests. The final step of the machine learning modelling is the model validation. The predicted corrosion rates will be verified with actual thickness measurement at site. To date, we have covered 30 process units which includes different trains, 120 corrosion groups selected from a total of about 3800 corrosion groups for the whole facility. 700 customized machine learning models were developed. Success is defined by best highest accuracy (>80%) with an optimum model run time. Recent validation has shown the ability to predict an anomaly in future trend which coincides with actual increase in corrosion rate.