Now more than ever, the industry is utilizing remote operations, with a corresponding increase in automation, in order to perform well construction in a safe and reliable manner. However, it is still challenging to maintain consistent and repeatable performance. To improve the likelihood of success when planning a new well, it is common practice to carry out offset analysis to learn from past mistakes and build on success factors. This process can be labor intensive as it relies on very experienced engineers to process and analyze the necessary data, for example, selecting appropriate offset wells and aggregating relevant information from historical drilling incidents. A set of analytics and workflows have been developed to automate the selection of relevant drilling intervals based on similar formation, equipment, and/or trajectory attributes and to apply this set of analytics to wellbore construction. Artificial intelligence and machine learning approaches have then been applied to score data quality, mitigate data gaps, isolate near misses and recorded drilling hazards, and assign similarity measures, which are then used collectively to index the historical dataset. This descriptive analytic suite then enables the drilling engineers to perform predictive analytics on the well plan, such as risk score estimation for encountering certain hazards or drilling issues along the planned well trajectory, and for residual risk determination after mitigation measures are applied. In the future, these analytics can be utilized in real-time, remote operations to improve situational awareness and mitigate common hazards by maintaining alignment with the plan and ensuring the effectiveness of mitigation measures. This presentation will showcase some examples of offset well analysis for stuck pipe and steering dysfunction. Close collaboration between operators, service companies and academia utilizing advanced machine learning and artificial intelligence frameworks will drive further advances in risk and cost reduction in the well construction space.
01:05:09