Machine Learning and Natural Language Processing for Automated Analysis of Drilling and Completion Data
Bibliographic record
Abstract
Abstract During Drilling and Completion (D&C) operations large volumes of data are typically collected in oil and gas fields. These datasets typically contain hidden (valuable) information that could be used to improve D&C performance (e.g., by identifying bottlenecks in drilling operations, analyzing non-productive time, optimizing rig schedule based on key indicators, etc.). Unfortunately, D&C datasets are typically not well suited for data mining: they are not structured, they are text heavy and they often contain numerous gaps and errors that hinder automated pre-processing techniques. In this paper, an innovative method to automatically extract smart analytics and opportunities from D&C reports is presented. Initially, a combination of Natural Language Processing, Data Mining, and Machine Learning algorithms are used to quality check a large volume of drilling data (including the text in the daily drilling reports), extract crucial information, and predict the non-productive time and its type. This results in a significant reduction of the labor-intensive quality check task for thousands of datasets and also the unbiased classification of the events. Then, the D&C datasets are integrated with other data sources such as production, geology, reservoir, etc. to generate a set of crucial drilling and reservoir management metrics. The proposed method, which was successfully applied to fields in North and South America, is applied here to two onshore fields located in the Middle East. By applying the developed tool, the data processing and integration time that used to take months to accomplish could be reduced to only a few days. In addition, analyzing metrics such as the Drilling Efficiency Index, normalized drilling days for each field and well type, cost analysis, detailed analysis of non-productive time, effect of completion parameters on production, design efficiency, etc. enabled us to quickly identify the D&C bottlenecks in each field and provide customized solutions to diagnose each problem. In addition, the historical data was used to improve future rig scheduling and resource allocation by applying advanced optimization algorithms with cumulative oil production, net present value, and operation time as the objective functions. In the final stage, the results were vetted by the experts to assure it meet the best D&C practices. The novelty of the presented method lies in using advanced technologies such as Natural Language Processing, Data Mining and Machine learning to QC, mine, integrate and analyze large volumes of D&C data in a very short time, find the bottlenecks and optimize the future plan with evident benefits of improving D&C performance and capital efficiency from a global reservoir management perspective.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".