Pilot Test for Continuous Production Optimization Using a Digital Solution on Permian Basin Wells

Objective/Scope
Optimizing production across unconventional assets requires rapid identification of well performance anomalies, efficient artificial lift optimization, and scalable evaluation of intervention opportunities such as acid jobs and lift-system transitions. Traditional surveillance workflows struggle to keep pace with high well counts, changing lift designs, and evolving reservoir conditions. This pilot study focuses on a digital production optimization system deployed on 441 wells equipped with Gas Lift and ESPs in Permian basin. The scope includes daily surveillance for production optimization, evaluation of artificial-lift transition scenarios, defining surface injection-pressure management criteria for multi-well gas-lift systems, and systematically assessing acid-job performance to determine optimal implementation conditions.

Methodology
A web-based production optimization platform, integrating comprehensive physics-based modeling with advanced AI-driven analytics, was used to continuously process historical and daily production data. The system employs a novel transient reservoir pressure estimation based on dynamic drainage volume computation with multiphase well modeling to characterize reservoir inflow performance, artificial-lift behavior, and deviations from design operating envelopes. Daily computations include productivity-index forecasting, bottomhole pressure tracking, and opportunity identification for lift adjustments (e.g., gas-lift injection tuning, ESP frequency optimization). Statistical analysis of historical acid-job interventions was conducted to correlate treatment success with inflow performance constraints identified and the chemical composition of produced fluids, particularly indicators of solids-related deposition risk. Multi-well gas lift modeling was used to evaluate injection-pressure requirements across groups of wells sharing same compressors and determine suitable transitions between high-pressure gas lift (HPGL) and low-pressure gas lift (LPGL) across shared facilities.

Case Study Results and Observations
Implementation of the platform’s daily optimization recommendations yielded a measurable and repeatable impact across the 441-well asset. Those wells in which recommendations were adopted delivered an average of 6% incremental oil production, primarily driven by optimized gas-lift injection rates and ESP operating frequencies. Concurrently, the field achieved over 20% reduction in average gas-lift usage, reflecting more efficient allocation of lift gas. The acid-job evaluation workflow identified the most favorable PI opportunities by tracking the PI trends associated to inflow issues for treatment success, providing operators with predictive criteria to avoid treatments likely to result in insufficient inflow improvement. Multi-well gas lift network analysis produced a clear guideline for managing surface injection-pressure constraints, including the timing and operational triggers for transitioning wells from HPGL to LPGL compressors to maximize field-wide lift efficiency.

Novelty and Significance
This work demonstrates how an integrated hybrid modeling system—combining physics-based flow dynamics with data-driven techniques, can transform daily surveillance and optimization workflows into unconventional asset management. Unlike traditional manual review processes, the platform delivers continuous, scalable, and objective recommendations for lift-control adjustments and conversions, well interventions, and facility-level gas-lift management. The structured analysis of acid-job performance provides a reliable framework for diagnosing treatment potential from both reservoir productivity and fluid-chemistry perspectives, minimizing ineffective interventions. The simultaneous optimization of ESPs, gas lift, and multi-well injection pressure management highlights the system’s ability to coordinate decisions across diverse lift systems and facility bottlenecks. The pilot results confirm the value of deploying automated, physics-informed digital solutions that enhance operational efficiency, reduce resource consumption, and support proactive field-wide production management.