Automated Well Time Estimation via Statistical Model Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for well construction and hydrocarbon extraction face challenges in accurately estimating well construction activity times and non-productive times due to uncertainties and reliance on human expertise, leading to inefficiencies and increased costs.
Innovation Solution
A system that employs parametric and nonparametric estimation methods with hypothesis testing to automatically select the most suitable probability distribution models for well construction activities, using data from offset wells to predict construction times and account for uncertainties, thereby optimizing well construction processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automated distribution model selection with hypothesis testing is implemented, then measurement precision of well construction time estimation is improved, but device complexity increases
Solution Approach 1:
The system performs self-service by automatically selecting appropriate probability distribution models through hypothesis testing without requiring manual expert intervention. The automated model selection process evaluates multiple distribution models and selects the most suitable one based on statistical criteria, thereby improving measurement precision while reducing reliance on human expertise.
Solution Approach 2:
The patent replaces manual expert judgment and mechanical estimation processes with automated statistical computing methods. Hypothesis testing and parametric estimation algorithms substitute human analysis, enabling more precise and objective well construction time predictions through computational rather than manual evaluation.
2Productivity
If automated distribution model selection is used, then productivity of well construction planning is improved, but loss of time for data analysis increases
Solution Approach 1:
The system performs preliminary actions by pre-configuring multiple probability distribution models and establishing hypothesis testing frameworks before actual well construction planning begins. This preparation allows the automated system to quickly evaluate and select appropriate models during the planning phase, improving productivity without excessive analysis time during execution.
Solution Approach 2:
The patent utilizes parameter changes by adjusting statistical parameters and thresholds in the hypothesis testing process to optimize the balance between analysis thoroughness and computational efficiency. By tuning these parameters, the system achieves accurate model selection while minimizing the time required for data analysis.
Data Source
AI summary
A method can include accessing data associated with a well and one or more offset wells; based on at least a portion of the data, generating a set of distributions via parametric estimation, where the distributions are associated with a well-related activity and time; analyzing individual distributions in the set of distributions with respect to at least a portion of the data to pass or fail each of the individual distributions; and, for one or more passed individual distributions, outputting one of the passed individual distributions for the well.