Drilling engineers use a group of offset wells as a starting point for well planning. These offsets can be used to calculate a composite well or a technical limit (minimum operating and non-operating time) and average well (P50 operating and non-operating time) to help understand the best possible and the likely performance of the upcoming well respectively. Comparing the composite well with the average well can help identify the well activities that have the largest potential for optimization. Statistical percentiles of the activities from historical wells can also be used to calculate a target for the upcoming well.
Gathering and analyzing the offset well data to manually calculate composite and average wells is a tedious time-consuming process and prone to human error. Multiple scenarios might have to be analyzed depending on the percentiles associated to each drilling phase of the offsets. This can greatly increase the cycle time.
This paper introduces a cloud-based algorithm to automate and speed up the process involved in calculating the composite, average and the target wells.
Well planning or well design is a process or workflow where various aspects of the well construction life cycle are planned out in advance based on offset wells and experience. There are various steps involved in well planning and one of the important steps is time and cost estimation which eventually leads to an AFE generation.
The deterministic approach of giving a single point estimate for time and cost for well construction can be unreliable considering the uncertainties involved in every step of well construction (Loberg et al. 2008). Broken equipment, unexpected weather and other events can contribute to non-productive time and greatly increase the overall cost for well construction. Based on geological factors different hole sections can also have different levels of uncertainty. Non-deterministic methods utilizing the statistical variation in elapsed time of each activity in the historical offset wells can be used to reduce the impact of this uncertainty.
