Managing pipeline integrity revolves around abundance of data and information from monitoring of safe operating limits, inspections and maintenance. Those data and information may come from real-time/on-line systems, manually input ad-hoc, manually input from inspections and maintenance carried out on a particular pipeline. The oil and gas pipeline industry have sort of mature with respect to having a software/tool in aiding and assisting personnel in performing risk, fitness-for-service, repair assessments, and executing management of change for a pipeline system; and many more assessments/analyses.
Nevertheless with the invent of analytics, there is strong need to explore the new ways of working with those abundance data and information OR maximising the utilisation of the data and information for the benefit of assessing or evaluating pipeline's risk and integrity to predict 'accurately' risk and integrity so that specific and cost-effective actions and mitigations can be deployed within a stipulated period of time.
In those regard, PETRONAS is actively working with industry to establish predictive analytics for critical offshore and onshore pipelines' threats/anomalies i.e. internal corrosion and free span for offshore; and external corrosion and geotechnical hazard for onshore. This paper will dwell on the principles, concepts and methodology of predictive analytics tools for the development. It is envisaged that eventually those threats/anomalies will be analytics-managed to eliminate unwanted incidents to PETRONAS's offshore and onshore pipelines. In addition, analytics-pipeline integrity management will also likely to provide 'accurate' prediction of 'future' pipeline integrity.