| dc.description.abstract | Over the last decade, there have been major changes in casing design requirements in the oil and gas industry. The changes were necessary to withstand the loads and stresses from the unconventional reservoir. However, there had been numerous reports of casing failure, indicating that the current casing design for the unconventional wells might not be adequate, resulting in a higher-than-expected failure rate in the production strings. Many factors are considered in casing design, as it goes through different stages of well completion and production. For this reason, it is difficult to address the cause of the failure. This study is carried out using statistical analysis methods to identify the most influential factors leading to the casing failure, and recommend what can be done to improve casing design by using the data analysis method.
This study focuses on a dataset from the Granite Wash play located in Texas and Oklahoma. Statistical analysis methods, including basic descriptive analysis method and logistic regression method, are employed to analyze the correlation between the failed wells and the possible factors contributing to the failure. The data analysis shows four main factors: casing size, average water used per fracture stage, average proppant used per fracture stage and casing weight, which are correlated to the casing failure in this study region.
The study concludes that data analysis can be employed in making an improvement in casing design for the wells that experience higher than usual rate of casing failure. By iii following the recommendations on using data analysis, the causes of the casing failures can be found which can be used for the casing design improvements | en |
