When Gas Flows and Liquids Move: Engineering Predictability into Remote Production
In remote hydrocarbon operations, two things define success: what comes out of the ground, and whether it can actually leave the field. This project redefined how performance is understood for critical midstream systems through integrated digital asset modeling.
For one upstream operator managing production across isolated clusters, the challenge was magnified by complex surface networks — where associated gas must be compressed and transported over long distances, and valuable natural gas liquids (NGLs) need stabilization and timely export. In such environments, design capacity means little if reliability is uncertain, maintenance is reactive, or bottlenecks emerge unexpectedly.
This project redefined how performance is understood for these critical midstream systems. Instead of relying on isolated equipment specs or optimistic flow models, the team built two integrated digital assets — Model 1 for a gas transmission network, and Model 2 for an NGL recovery chain — that simulate real-world behavior over time, accounting for failures, repairs, maintenance strategies, and operational dependencies.
This isn't just simulation. It's a new operating principle — where uncertainty is replaced with insight, and resilience becomes a designed outcome, not a hope.
Technical Approach
This project delivered a comprehensive, dual-system Reliability, Availability, and Maintainability (RAM) analysis for two core hydrocarbon handling networks serving a remote upstream oil production operation.
Using AspenTech Fidelis, both models were built as dynamic, probabilistic digital assets capable of simulating multi-year operational timelines under realistic variability.
The models were developed collaboratively with engineering and operations teams through structured workshops, ensuring that logic rules and system behaviors reflect actual practices, not theoretical ideals.
Rather than treating each model in isolation, they were aligned under a common methodology, data structure, and reporting format.
Each Model Incorporates:
- Equipment-specific failure rates and repair durations
- Maintenance logic (planned and unplanned)
- Operational sequencing rules and control dependencies
- Power supply links and external constraints
System Coverage:
- Model 1: Associated gas gathering and transport network
- Model 2: NGL stabilization and delivery chain
Key Achievements
Real-World Variability Simulation
Both models account for unplanned outages, scheduled shutdowns, standby activation, and cascading effects — generating statistically robust forecasts under actual operating conditions.
Quantified Lost Production
For the first time, deferred gas volumes and undelivered NGL barrels due to system unreliability were estimated and allocated to specific subsystems.
Direct Downtime Tracking
Custom measurement units were added to capture annualized NGL downtime in barrels, providing a clear link between technical performance and commercial impact.
Failure Contribution Mapping
Specific failure modes were defined, classified, and simulated, revealing which issues drive the most disruption across both systems.
Validated Operational Assumptions
Iterative review sessions ensured that logic configurations matched field reality, improving model credibility and usability.
Unified Analytical Foundation
By developing both models under the same framework, the organization now has a consistent language for assessing performance across different asset types.
Strategic Impact
The dual-model approach transforms decision-making from fragmented assessments to integrated foresight. It enables leadership to:
- Prioritize capital investments based on quantified returns in availability and throughput
- Optimize maintenance planning by focusing on "bad actor" components
- Assess risk exposure for production commitments with complex, interdependent systems
- Improve coordination between gas and liquids teams, breaking down silos
- Support contingency planning with realistic scenarios
- Align technical choices with commercial obligations
Designed as a living platform, the models are not end-state deliverables but foundational tools for continuous improvement. Their structure allows for updates as new data becomes available, and their methodology is replicable across other clusters, offering a scalable blueprint for enterprise-wide reliability modeling.
In an environment where every barrel deferred carries cost and risk, this initiative sets a new standard for how remote production assets should be analyzed, optimized, and managed.
Technologies & Methodologies
Ready to Transform Your Remote Production Assets?
Our integrated dual-system approach can be adapted to your specific remote production challenges. Let's discuss how we can help you replace uncertainty with insight and make resilience a designed outcome.