The $2 Billion Proof Point

Half of all oil and gas companies have now adopted digital twin technology as a core operational system. Not a pilot. Not a proof of concept. A production system that drives daily decisions about maintenance, operations, and capital allocation.

Shell’s digital twin program is the industry benchmark: $2 billion in annual savings from 20% less equipment downtime and 25% lower maintenance costs. BP, Chevron, and ADNOC have similar programs at scale. The technology works. The ROI is proven. The question isn’t whether digital twins belong in refinery operations — it’s what they’re still missing.

What Digital Twins Actually Model

Today’s refinery digital twins excel at physical asset modeling. Turbines, compressors, heat exchangers, distillation columns, relief valves — each piece of equipment gets a virtual counterpart that ingests real-time sensor data and predicts future behavior.

A digital twin of a shell-and-tube heat exchanger can track fouling rates, predict when the exchanger will drop below minimum heat transfer efficiency, and recommend cleaning schedules that optimize the balance between production uptime and maintenance cost. That’s valuable. It works because the physics is well-understood and the sensor data is continuous.

Fouling models, vibration analysis, catalyst deactivation curves, corrosion rate tracking — these are the workhorses of the current digital twin ecosystem. They answer the question: “when will this equipment need attention?”

The Layer That’s Missing

Here’s what most digital twins don’t model: the engineering standards that define what “safe” and “compliant” mean for the equipment they’re monitoring.

Every piece of refinery equipment has a design basis rooted in specific engineering standards. A pressure relief valve’s set point is determined by ASME standards. A distillation column’s shell thickness is governed by ASME Section VIII. Piping design references ASME B31.3 Process Piping. Instrument setpoints reference ISA standards.

These standards update on their own schedules — typically every 3–5 years. When ASME B31.3 publishes a new edition that changes a stress allowable or adds a material requirement, every facility that designed piping to the previous edition has a potential compliance gap. Under OSHA’s PSM standard (29 CFR 1910.119), facilities must use RAGAGEP — Recognized and Generally Accepted Good Engineering Practices. That means current standards, not the edition that was current when the plant was built.

A digital twin that monitors a piping system’s vibration, temperature, and pressure in real time but doesn’t know that the piping’s design basis references a superseded standard is only seeing half the risk picture.

Why This Gap Exists

The gap isn’t accidental. It’s architectural.

Digital twin platforms are built by software engineers optimizing for sensor data ingestion and physics-based modeling. Engineering standards compliance lives in a completely different system — typically a document management platform, a standards tracking spreadsheet, or a binder on someone’s shelf.

The two systems don’t talk to each other because they were never designed to. The digital twin vendor doesn’t sell standards compliance. The standards management system doesn’t integrate with OPC-UA data historians. And the engineering team that understands both domains is too busy running day-to-day operations to build the bridge.

The result: a $2 billion digital twin platform that can predict when a relief valve needs maintenance but can’t tell you that the valve’s design basis references API 521 6th Edition while the current standard is the 7th Edition — and that the delta includes changes to relief system sizing methodology that could affect the PHA.

What Closing the Gap Looks Like

The compliance layer that digital twins are missing isn’t another sensor feed. It’s an AI-assisted standards analysis engine that maps every piece of equipment to its design-basis standards, tracks those standards for currency, identifies gaps when new editions publish, and feeds compliance status into the digital twin’s risk model.

That’s what NORMEX Standards AI is designed to do. Not replace the digital twin — complement it. The digital twin tells you the equipment’s physical condition. NORMEX tells you whether the engineering basis underneath that equipment is still current, compliant, and defensible.

In practice, that means a facility running a Shell-class digital twin could integrate a standards compliance feed that flags when a RAGAGEP gap affects equipment the twin is actively monitoring. The turnaround planner doesn’t just see “heat exchanger E-201 needs cleaning in Q3” — they see “heat exchanger E-201 needs cleaning in Q3, and the design basis standard for this exchanger updated in January, and the new edition includes a change to tube material requirements that should be evaluated before the maintenance window.”

That’s the difference between a maintenance optimization tool and a process safety management system.

The Market Opportunity Is the Seam

The digital twin market in oil and gas is mature and well-funded. The standards compliance market is fragmented, manual, and underserved. The seam between them — where equipment health data meets engineering standards currency — is where the next generation of process safety technology will be built.

The facilities that figure this out first won’t just save money on maintenance. They’ll have audit-ready documentation that traces every equipment decision back to current RAGAGEP, every PHA assumption back to a valid engineering basis, and every compliance claim back to data the digital twin can verify in real time.

That’s not a feature request for the digital twin vendors. That’s a market opportunity for the tools that sit between the twin and the standards library.

---

Timothy Porritt is founder of Porritt Inc., building AI-powered tools for heavy industry including NEXUS CAD and NORMEX Standards AI. A petroleum engineer by training, Timothy writes about the intersection of industrial engineering, AI, and entrepreneurship.