Umm Shaif Field in ADNOC

Overview

The Umm Shaif Field, operated by ADNOC, is an offshore asset with complex reservoir heterogeneities and a mix of oil and gas development challenges. Traditional full-physics reservoir simulations for field development planning (FDP) were taking 9–12 months, limiting the ability to evaluate multiple scenarios and reducing decision-making confidence.

Objective

Deploy HawkEye FDP™— AI/ML-driven hybrid modeling platform and optimize current FDP to see if we can get more oil with less wells.

Optimize well number, placement, and trajectories.
Reduce CAPEX while maintaining or have a higher recovery.

Approach

HawkEye FDP™ integrates Fast Marching Method (FMM), hybrid machine learning models, and advanced optimizers to identify optimum well locations and trajectories using only a single simulation file as input. The workflow:

Preprocess field data and history-matched dynamic models.

Generate Relative Opportunity Ranking maps to identify high-potential zones.

Run hybrid optimization integrating geological, surface, and operational constraints.

Validate optimized solutions via reservoir simulation.

Results

Optimized Well Count

Fewer new wells compared to the base case (e.g., -9 OP / +3 WI in Arab D5 scenario).

Higher Recovery

Up to 17% additional oil for new reduced well cases.

Improved Plateau

Extended production plateau with higher reservoir pressure.

CAPEX Savings

Significant reduction through drilling fewer, higher-impact wells.

Key Advantages

Speed

Reduced FDP optimization cycle from 9–12 months to 4 weeks.

Flexibility

Ability to propose alternate well trajectories delivering equivalent oil recovery in case of drilling challenges.

Accuracy

Validated by ADNOC’s Thamama Excellence Center (TEC) through a “blind test” on Arab D5 model.

Impact

"The POC demonstrated ADNOC’s ability to move from static, simulation-heavy workflows to a fast, automated, and data-driven FDP process. The “More with Less” concept—higher recovery with fewer wells—proved both technically and economically viable, paving the way for broader deployment across ADNOC’s asset portfolio."