Correlation Between Completion Quality and Productivity Degradation Narrows Down Uncertainty in Production Forecasts

Abstract

Abstract Productivity index (PI) degradation is a critical issue that negatively impacts hydrocarbon recovery. This problem is very common in deepwater environments such as Gulf of Mexico (GOM). In recent years, several publications have addressed this issue proposing integrated workflows to assess and mitigate this problem and providing interesting insights into the causes of this phenomenon. Our organization commissioned a study (Knobles et al. 2017) to understand the reasons for the high productivity declines observed across our assets in GOM and identify lessons learned and best practices to reduce the severity of the PI degradation in cased-hole-frac-pack (CHFP) completions in high-permeability reservoirs (50-1000 mD), which are the most common in our operations. These findings have contributed to improve our CHFP performance. This performance has been assessed based on a completion scorecard that grades and weights the key design and execution parameters of the CHFP completion and yields a score (0 to 100) for the completion execution and productivity. Knobles et al. established a correlation between the productivity score in the completion scorecard and a qualitative ranking of the well productivity performance, namely productivity tiers. The wells in scope were categorized in four groups, tier 1, tier 2, tier 3, and tier 4. Tier 1 is the highest performance level, i.e., low initial skin and low productivity degradation. Later, the completion scorecard was reviewed by a group of completion subject-matter experts (SME) to improve the correlation between the productivity score and the productivity tiers. The updated scorecard enhanced this relationship making it suitable as an indicator of long-term productivity. However, the use of this correlation in production forecasting is limited because of its categorical nature. The objective of this work is to quantify the relationship between the completion scorecard and the productivity performance and use the results to narrow down the productivity degradation uncertainty range in production forecasts. We use data science concepts and tools to analyze our well productivity database, Integral Well Productivity Tracker (IWPT), and develop a numeric correlation between the actual well productivity data observed in the field and the completion scorecard results from the post-job evaluations. The process includes IWPT data wrangling, modeling, and visualization. We performed regression analyses on the normalized productivity index and cumulative liquid production for approximately 120 wells in the IWPT dataset to calculate PI degradation rate for each well and combine them with the completion scorecard results. The results validate the conclusion from Knobles et al.'s work, i.e., a correlation exits between the productivity score from the completion scorecard and the productivity performance. The new, numerical relationship allows us to consider the improvements in completion quality. This quantitative correlation can inform the definition of the productivity-degradation-uncertainty ranges in the production forecast during planning and after the completion job when the completion scorecard results are available. Narrowing down the uncertainty in productivity degradation significantly impacts the estimated-ultimate recovery (EUR) of any GOM deepwater project.