Operators who have embraced a proactive continuous improvement approach to managing their Pipeline Integrity Management Programs are now including elements of data science to efficiently assess large volumes of data for trends and outliers. With structured and normalized data, Data Scientists can construct machine learning models that expose previously unavailable insights and automation.
This white paper showcases a machine-learning approach used to predict the location and severity of metal loss, with the intent of providing a non-destructive screening tool that can quickly and reliably aid pipeline operators in assessing the threats of internal and external corrosion on pipelines without ILI data.
Our white paper provides information on machine learning methodologies, case studies where machine learning was used to develop a corrosion prediction model, recommendations for future developments and more.
Download a copy of our white paper to learn more.