Automating Oil & Gas Asset Valuation
Classic asset valuation usually runs through three specialists in sequence — a geologist, a reservoir engineer, and an economist — reconciling data in spreadsheets by hand: weeks of work, dozens of edits, and files that drift out of sync. Here's what in that process can be calculated automatically — geology, reserves, the production profile, NPV/IRR economics — and what still has to stay with a human.
Automating oil and gas asset valuation means moving geology, reserves, production profile and NPV/IRR economics out of scattered spreadsheets and into one calculation loop where every number can be checked. It doesn't replace judgment: a person picks the scenarios and signs off the result — without losing weeks to manual reconciliation or one broken cell throwing off the whole model.
Why Does an Asset Valuation Usually Take Weeks?
The classic route from "we found a block" to "we understand the NPV" runs through several specialists in sequence: a geologist interprets the data and hands the result to a reservoir engineer, who calculates the production profile and hands it to an economist, who reconciles everything into one financial model. Every handoff is a manual transfer of numbers between formats — and a chance for the methodology to drift at the seam.
SPE's own trade press describes the same path at an industry-wide level — from seismic data to a finished development plan:
«This workflow, as currently practiced, is a highly complex interaction of numerous different types of software applications combined with expert input from multiple disciplines of geoscience and engineering. It is expensive, time-consuming, and a considerable source of delay between first hydrocarbon discovery and ultimate economic exploitation of a resource.» — Thomas Halsey, JPT (SPE) — "A Grand Challenge: Digital Transformation for the Upstream Oil and Gas Industry"
Researchers at the Oil and Gas Research Institute of the Russian Academy of Sciences and Gubkin University describe the same methodology from the inside — and land on a similar list of pain points:
- Field-development investment projects run for decades — a modeling error doesn't surface right away
- There's never enough source data, so the forecast carries a high degree of uncertainty
- An error in the technical-economic forecast means a real risk of missing production targets later, in the field
Their conclusion isn't "calculate more carefully by hand" — it's to move the methodology into a system that calculates on its own:
«…the methodological provisions discussed were built into an industry-wide automated system capable of running multi-variant calculations of technical-economic indicators for investment projects.» — Sardanashvili, Bogatkina, Lyndin — Neftegaz.RU, "Technical-Economic Evaluation of Field Development. Problems and Methodological Provisions" (in Russian)
How Big Is the Risk From Manual Spreadsheets?
An asset valuation model is usually a chain of linked spreadsheets: reserves feed the production profile, the production profile feeds the financial model. A formula copied into the wrong row, or a range that didn't pick up newly pasted data, propagates through the whole chain — and it isn't always visible right away, because nobody has time to recompute everything by hand after every edit.
This isn't a hypothetical risk — it's a measured one. Prof. Pak-Lok Poon and colleagues from four universities and hospitals reviewed 35.5 years of research on spreadsheet quality:
«A recent study has found that 94% of spreadsheets used in business decision-making contain errors, posing serious risks for financial losses and operational mistakes. […] "The high rate of errors in these spreadsheets is concerning," says Prof. Poon.» — Phys.org, on Pak-Lok Poon et al., "Spreadsheet quality assurance: a literature review," Frontiers of Computer Science (2024)
A related problem isn't just the calculation itself — a model goes stale the moment it's finished. According to Dmitry Maslennikov, director of digital transformation in oil production at Tsifra Group, a manually built model doesn't update in real time, and refreshing it takes months of work from a whole institute of engineers — so the next decision on the asset often gets made on an already outdated number.
What Exactly Gets Automated in an Asset Valuation?
Automation doesn't replace any single stage of the valuation — it takes over the recalculation and reconciliation between them, leaving the choice of assumptions and the final check to a person. In practice, that's five linked calculations:
- Gathering the block's geology — from open industry data, not a folder of manually forwarded files
- Estimating reserves by analogy, within a probabilistic P90/P50/P10 range
- A year-by-year production profile — instead of manually extrapolating a decline curve in a separate spreadsheet tab
- Economics — NPV, IRR — across several price scenarios and development options at once
- Sensitivity analysis — an automatic recalculation whenever any input changes, instead of a manual pass back through the whole chain of spreadsheets
Vygon Consulting keeps two separate software modules for this — one for development-performance calculations, one for the technical-economic side — and calls out the effect of automation specifically on recalculation speed:
«A module for the automated review of the economic section of technical field-development projects» — Vygon Consulting, SMARTEC (in Russian)
«Cuts the calculation time for forecast hydrocarbon production profiles, relative to a full reservoir simulation model, by hundreds of times» — Vygon Consulting, SMARTECH (in Russian)
Where Does Human Responsibility Still Apply?
Automation removes manual labor — copying numbers between tabs, recalculating everything after every edit, reconciling the formats of three specialists into one file. It doesn't remove the decisions: which fields to treat as analogs, which price scenario to use as the base case, what to do when new data doesn't reconcile with the existing model. That's still a person's call, not the system's.
A good illustration of that boundary isn't asset valuation itself, but a neighboring problem with the same logic: a digital twin of a producing asset. Dmitry Maslennikov of Tsifra Group describes exactly this principle — the system runs on its own while the data lines up, and hands the decision to a person only when something doesn't:
«The system should see that a new well has appeared, pull in its data, and connect it to the whole system… and call in an engineer only in exceptional cases, when inconsistencies show up in the data — for example, a new well was added, but total oil output went down» — Dmitry Maslennikov, Tsifra Group — IA Devon, "What's Holding Back Full Digitalization in Oil and Gas" (in Russian)
The same principle applies to asset valuation: it's not "the system decides instead of the person," it's "the system doesn't let a person burn time where there's nothing to decide," and it flags exactly where the data disagrees and needs a human look. Being checkable doesn't come from a person having typed the number by hand — it comes from being able to see its source and formula, and recalculate it at any time.
How Does AVP AI Automate Valuation Without Losing Traceability?
AVP AI — an AI platform for oil and gas asset evaluation. A coordinator and eight specialised agents — search, geology, reservoir simulation, surface facilities, economics, wells, business case and data standardisation — together run 34 calculation tools and cover the whole path from a block's coordinates to NPV without manual spreadsheet work.
Given coordinates or a license number, the platform pulls the block's geology from open industry data, estimates reserves by analogy within a probabilistic P90/P50/P10 range, builds the production profile and surface-facilities layout, and calculates the economics — NPV, IRR — across several price scenarios at once. The result is a report in minutes, not weeks, and every step is logged: what data was used, from which source, and what formula produced the result. The final number can be checked, not just trusted.
For what the individual agents calculate, see agentic systems in oil and gas and AI in oil and gas. Once a block is bought or a license is granted, the same calculation loop becomes the basis for oil and gas asset due diligence — now with real terms in hand instead of hypothetical ones.
Frequently Asked Questions on Asset Valuation Automation
Because the data passes through several specialists in sequence — a geologist, a reservoir engineer, an economist — and every handoff is a manual transfer of numbers between formats. SPE's JPT describes this path from seismic data to a development plan as "expensive, time-consuming." Russian researchers studying technical-economic field evaluation cite the same causes: long project timelines and a high degree of uncertainty from a shortage of source data.
Yes, and it's a measured fact, not a guess. Research by Pak-Lok Poon and colleagues (Frontiers of Computer Science, 2024) found that 94% of spreadsheets used for business decisions contain errors. In asset valuation these spreadsheets are usually chained together — reserves → production → economics — so one wrong formula propagates through the whole model, and without recomputing from scratch it isn't always visible.
Five linked calculations: gathering a block's geology from open data, estimating reserves by analogy within a P90/P50/P10 range, a year-by-year production profile, economics — NPV, IRR — across several price scenarios, and sensitivity analysis for every input. A coordinator and eight specialised agents do this in minutes instead of weeks of manual reconciliation.
No. Automation removes the manual recalculation and the transfer of numbers between tabs, not the choice of assumptions — which fields count as analogs, which price scenario is the base case, what to do with data that doesn't reconcile with the model. A person still makes that call. The system's job is to show the source and formula behind every number, not to hide that boundary, so the decision can be checked rather than taken on faith.
Sources
- Research JPT (SPE) ↗ A Grand Challenge: Digital Transformation for the Upstream Oil and Gas Industry
- Industry media Neftegaz.RU ↗ Technical-Economic Evaluation of Field Development. Problems and Methodological Provisions (in Russian)
- Industry media IA Devon ↗ What's Holding Back Full Digitalization in Oil and Gas (in Russian)
- Research & consulting Vygon Consulting ↗ Digital Transformation — Automating Economic Evaluation and Field-Development Calculations (in Russian)
- Research Phys.org ↗ Study finds 94% of business spreadsheets have critical errors
We will automate the valuation of one of your assets — free of charge
Give us a license number or coordinates — the platform will pull the geology, estimate reserves, production and NPV without manual spreadsheet work, in minutes.