Ensemble-based well placement optimization improves infill drilling decisions for Petoro

Petoro improved well placement decision-making by leveraging ensemble-based analytics in Well Placement Advisor on the Delfi™ digital platform. The solution increased confidence in drilling outcomes, reduced simulation workloads by up to 60%, and accelerated evaluation workflows through automation and cloud scalability.

Petoro, needed to maximize value from complex reservoir assets while operating under uncertainty. Their assets are modeled using ensemble-based approaches, but prior workflows relied heavily on deterministic percentile cases—leaving up to 95% of valuable data underutilized.

The organization required a cloud-enabled, scalable solution that could:

• Rapidly analyze large ensemble datasets.
• Support data-driven well placement decisions under uncertainty.
• Leverage full ensemble information instead of simplified scenarios.
• Reduce manual workflows and computational inefficiencies.

Additionally, the high cost and risk of drilling new wells, particularly the potential for dry wells, made it critical to improve accuracy and confidence in identifying optimal infill drilling locations.

To address these challenges, Petoro partnered with SLB to implement the Well Placement Advisor within a modern reservoir engineering workspace, enabling probabilistic, AI-driven analysis of reservoir data.

By implementing Well Placement Advisor on the Delfi platform, and adopting an ensemble-based workflow, Petoro achieved significant improvements in efficiency, accuracy, and decision confidence. The solution enabled the identification of high-potential well targets using probability maps and opportunity index (OI) analysis across ensemble realizations, improving target accuracy and robustness.

By incorporating the full range of uncertainty into the analysis, Petoro reduced drilling risk and increased confidence in avoiding non-productive wells.

The workflow also delivered substantial simulation efficiency gains. Advanced ensemble reduction techniques reduced simulation runs by up to 60% while maintaining 99.5% accuracy, significantly lowering computational effort. At the same time automated simulation updates and cloud-based execution accelerated well evaluation cycles and minimized manual intervention.

Supported by cloud-native infrastructure, the approach enabled parallel simulation and improved accessibility for reservoir engineers, making workflows more agile and collaborative.

Overall, automation, clustering, and ensemble reduction collectively decreased processing time and storage requirements, making large-scale studies more practical and cost-effective. The solution enabled Petoro to fully leverage ensemble data, transforming well placement from a deterministic, manual process into a probabilistic, data-driven workflow that maximizes reservoir value while minimizing risk.