Equipment failures result in costly unplanned downtime of operations and can also lead to serious safety and environmental incidents. Therefore, predicting failures in advance and pinpointing the source unlocks significant value. Application of machine learning to real-time measurements and physics simulations provides a strong predictive maintenance framework for avoiding catastrophic failures and reducing costs.
SPE-GCS Data Analytics Study Group is pleased to collaborate with MathWorks and offer a hands-on workshop on Predictive Maintenance using Machine Learning. In this four-hour, instructor led online workshop, participants will tackle an oilfield equipment failure problem. Participants will build data-driven models in MATLAB using time series data and predict faults.