Description
Format:
Lunch and Learn (lunch is provided) followed by networking.
Abstract:
This paper aims to characterize the under-developed Wolfcamp C formation in the northern Delaware Basin for potential inclusion in long-range asset planning. The study combines multidisciplinary efforts, integrating geochemistry, geology, geophysics, and reservoir engineering data. Two methods for normalizing rock and fluid properties across large areas are outlined: the "Analog Approach" with binary cut-offs and the "Integrated Machine Learning (ML) Approach" using supervised ML algorithms.
The team defined geologically similar areas (GSAs) through manual integration and grid-to-grid unsupervised ML algorithms. Reservoir engineering techniques quantified production interference within each GSA, validated by multiple data sources. The results highlight the success of the integrated ML approach, which optimizes field development using digital technologies and informed business decisions. The proposed workflow contributes to optimized full-field development and can be automated for continuous improvement with additional data.