In the field of drilling automation, most attention is usually focussed on the automating the mechanical side of the business and yet vast amounts of time and energy are spent by well engineers on designing and executing wells that may involve redesigns during the execution of the well due to subsurface uncertainty. This effort involves many people in the organisation, creates large amounts of documents and data that usually reside in different silos. The effort involved in pulling the right data together from different sources and checking that it's correct involves a lot of drudgery, hence it called be a ‘Cinderella’ task. There is a maxim that says that "If you can't measure you can't manage", but it's equally true that "if you can't analyse your measurements effectively and efficiently then you can't optimize", either for ROP or energy efficiency. This paper discusses the essential requirements for workflow optimization and drilling advisory systems, both as a business case in itself, but also as a reassuring steppingstone to full drilling automation.

Significant advances in AI and Cloud Computing in the last decade have led to the ability to better automate the workflow process and allow real-time drilling advisory systems that allow a human-in-the-loop to react to and modify the drilling process. The essential parts of such are a workflow/optimisation system, no matter how good the AI, is that it should be intuitive, easy to use, vendor neutral, have robust data cleansing, be backed by 24/7 support, be customisable to the particular needs of the operation and flexible enough to work on different platforms yet allow expansion as new requirements are added. Operators are consistently wanting to centralise, then standardise, data flows for greater automated analysis and management.

Using such a system, ROP improvements of over 18% have been seen in land operations, flat time reductions of 20% have been seen on offshore wells, weight to weight connection times have been reduced by over 35%, costly trips out of hole have been avoided due to MWD failure by directly comparing hole depth with offset well data in real time.

With cloud-enabled, real-time AI, there is currently no limit on how ‘joined-up’ the drilling process can be, from rig scheduling through to production. Once confidence is gained by operators with drilling advisory systems and data quality then the next steps would be to hook these systems directly to drive rigs and tools with the human now as the monitor.

You can access this article if you purchase or spend a download.