Data model improves risk management in the North Sea
PROBLEM:
• Having invested in R2S digital twin technology to help manage its North Sea assets, an oil major required further insights to improve the efficiency and productivity on the asset.
• During the course of a James Fisher Asset Information Services (AIS) Design Thinking workshop, a key problem identified was the lack of efficiency in resolving anomalies. This is partially due to the lack of visibility of the anomalies across the asset as a whole, therefore accumulating risk.
• The maintenance team had no easy way of visualising multiple jobs to be completed within the vicinity of one another, and where the high risk areas lie.
SOLUTION:
• Whilst the Canadian company had no means of collating and visualising the information, they did have silos of raw data stored across the business.
• AIS first deployed deep learning (DL) to extract the data, then consolidated it into R2S.
• Extracted data was used to build a spatio-temporal risk model for the prioritisation of anomalies. This model was delivered within R2S to give a visual representation of where the greatest risks lie on the asset.
• Risk accumulation across the asset was highlighted, allowing the company to focus its attention and activity accordingly.
RESULTS:
With the implementation of this data model, the customer has been able to:
• Extract additional value from otherwise dormant data silos.
• Contextualise anomalies based on risk, location and other domain and business-specific parameters, allowing the company to plan work more efficiently, meaning higher resource utilisation and fewer trips to the asset. This reduces impact on POB and cost of mobilisation.
• Allow for data and risk driven decision making, ensuring the asset is a safer environment for all contractors.