Gazprom Neft: the 'Digital Rig' cut non-productive drilling time by 15%

• Проект «Цифровая буровая» (6 технологий, снижение НПВ от осложнений ~15%) — Platforms.su, n/a
• «Газпром нефть» развивает проект «Цифровая буровая»: подробности и перспективы (хронология, партнёры, ЭРА.ПИК, «безлюдная буровая») — Integral Russia, 2025-04-13
• Gazprom Neft courts Middle East NOCs with digital technologies (сбои за минуты, stick/slip за 1,5+ часа, геонавигация ~50 скважин и 3–5%, CyberFrac) — JPT (Society of Petroleum Engineers), 2021-12-01
• «Цифровая буровая» — большой проект (ОПР: −6 суток, +28,5% метражные операции, +40% к нормам наращивания; цитаты Бариева, Бажина, Калимуллина) — Управление производством, n/a
• «Газпром нефть» запустила первую в России роботизированную буровую установку — Neftegaz.ru, n/a
• Цифровая трансформация «Газпром нефти» дала экономический эффект (пилот «Актив будущего» — 1,2 млрд ₽; геологоразведка 5–6 млрд ₽/год с 2025) — ComNews, 2020-08-06
• Парк нефтяного бурового оборудования в России (контекст: ~1 360 действующих установок) — Dprom.online, n/a
• Центр управления бурением «ГеоНавигатор» (с 2011, 700+ скважин/год, проводка по пласту 60% → 90%+) — Фонтанка.ру, n/a

Background

Gazprom Neft is one of Russia's largest oil companies and among the most consistent players in drilling digitalization. The industry context matters here: a drilling rig is a factory that earns money only while the bit is deepening the well. Every hour of downtime or accident recovery is pure loss, and Russia's entire active rig fleet was estimated by the Industry Ministry at roughly 1,360 units (2020) — meaning the reserve for growth by 'just drilling more' is physically limited, and the efficiency of each rig becomes the industry's main currency.

The company's digital drilling story began long before the AI hype. Back in 2011, the 'GeoNavigator' Drilling Control Center opened on the Moika embankment in St. Petersburg — an engineering headquarters that remotely supervises construction of the company's most complex wells across the country: from Orenburg and Yamal to the Arctic shelf and the Sea of Okhotsk. Today the center handles more than 700 high-tech wells a year, and high-tech wells account for about 60% of all wells the company drills annually (Fontanka).

The 'Digital Rig' project is the next turn of that spiral. It has been developing since 2017 in partnership with oilfield services company NSH Asia Drilling, with technological participation from IBM and Skoltech. The idea is to combine six technologies into one system on the rig itself: automated monitoring of drilling parameters, automatic weight-on-bit control, trajectory regulation, an automated hydraulic wrench (the 'Iron Roughneck') for pipe make-up, IoT sensors on the rig's main assemblies, and the INTROS-AVTO automated flaw detector for inspecting the traveling cable. Pilot field trials began in February 2019 at Gazpromneft-Noyabrskneftegaz assets, and a regional drilling control center opened there in September of the same year.

A distinctive trait of the project is its bet on in-house software. The company's geosteering software, per the industry journal JPT, became the first drilling software developed entirely in Russia, and its own ERA.PIK platform is being built for drilling design and management. AI models are trained on data from real assets — the Novoportovskoye, Vostochno-Messoyakhskoye, and Orenburgskoye fields. This is not a 'pilot for the press release' but a multi-year program with measurable checkpoints — which is exactly why the case rewards a detailed look.

Problem

The main enemy of drilling economics is non-productive time (NPT): downtime and dealing with complications when the well isn't getting deeper but money is being spent. Complications like stuck pipe and stick/slip vibrations (the bit periodically 'sticks' and snaps free, destroying the bottomhole assembly) develop gradually, yet a human at the console notices them too late — when a stop is already inevitable. Meanwhile the wells themselves keep getting harder: the company drills long horizontal sections that must be kept inside a productive layer roughly ten feet thick — a task where an error of degrees means meters of wellbore outside the oil-bearing zone.

The second part of the problem is people and distances. Experienced drilling engineers are scarce, and the fields are scattered across Western Siberia, Yamal, Orenburg, and the shelf; putting a strong engineering team on every rig is physically impossible. Before support centers appeared, knowledge was concentrated in specific people at specific sites, and the quality of well placement depended on who happened to be on shift.

The third part is routine that eats speed. Tripping operations, pipe make-up, monitoring the traveling cable's condition — repetitive manual operations, each taking minutes, but across thousands of repetitions adding up to days of lost time. There is a safety factor too: pipe make-up and traveling-cable work are operations where a person stands in the hazard zone, and automation physically removes them from it.

Finally, a rig generates a continuous stream of data from dozens of sensors that no human can process in real time. That is why the project's goal was framed in terms of time — cutting complication-related NPT by about 15% — rather than in abstract 'digitalization percentages'.

Solution

The company combined two loops: automation on the rig itself and centralized AI analytics in support centers.

Loop one — the 'Digital Rig' on site. Six technologies operate as a single system. A hardware-software complex monitors drilling parameters and automatically applies weight on bit — more smoothly than a human can. The trajectory regulator keeps the wellbore on plan. The 'Iron Roughneck' — an automated hydraulic wrench — makes up pipe: in the words of NSH Asia Drilling chief technologist Andrey Bazhin, 'the machine performs every operation twice as fast as a human,' and pipe make-up norms were exceeded by 40% in the field trials. IoT sensors on the rig's main assemblies feed predictive equipment maintenance. The INTROS-AVTO flaw detector continuously inspects the traveling cable via magnetic fields — drilling control center head Tagir Kalimullin noted its cost is 'incommensurate with the degree of risk reduction' it delivers.

Loop two — predictive analytics in drilling support centers. AI advisors predict sudden failures several minutes before they occur, and developing stick/slip vibrations are recognized more than an hour and a half before a forced stop (JPT/SPE). The company's in-house geosteering software — per JPT, the first drilling software developed entirely in Russia — had steered nearly 50 wells by early 2021 and keeps the wellbore inside layers about 10 feet thick, saving 3–5% of rig time. The flagship GeoNavigator center in St. Petersburg remotely supervises more than 700 complex wells a year; after it appeared, the efficiency coefficient of placing horizontal wells within the oil-bearing layer rose from 60% to over 90% — of a kilometer-long horizontal section, about 900 meters now run through the pay zone (Fontanka).

The rollout chronology is notable for its gradualism — and for the decision points built into it. 2017 — project start. February 2019 — field trials at Gazpromneft-Noyabrskneftegaz: six wells, with targets achieved by the fifth — meaning the system reached its design regime before the trials even ended. September 2019 — launch of the regional drilling control center, which closed the loop between 'Digital Rig' data and round-the-clock engineering support. Then — a second trial phase on mobile and echelon-type rigs, the addition of Slavneft-Megionneftegaz, training AI models on data from the Novoportovskoye, Vostochno-Messoyakhskoye, and Orenburgskoye fields, development of the company's own ERA.PIK platform for drilling design and management — and the launch of Russia's first robotic drilling rig. The declared vector is the 'unmanned rig'.

Alongside drilling, the company builds an adjacent digital stack: the CyberFrac hydraulic fracturing simulator, created with St. Petersburg Polytechnic University after a 2017 government competition, is per JPT 10–20% more accurate than Western commercial analogues, enables well designs with 15% lower capital costs and a 5–10% production uplift, and simulates a full multistage stimulation in about 3 minutes.

There was also an external maturity test for this stack: per JPT, the company offered its digital drilling technologies to Middle Eastern national oil companies — a 2019 agreement with ADNOC (Abu Dhabi) included digital technology sharing, and partnerships with Dragon Oil and Mubadala Petroleum were under discussion. Roman Sokolov, manager of international technology scouting, framed the position as: 'We are not a sales company, we are an oil and gas company' (JPT). Technologies built for the company's own fields proved competitive enough to become the subject of international negotiations.

Result

Published results of the field trials at Gazpromneft-Noyabrskneftegaz: well construction time fell by 6 days versus norms (17.5% for complex wells), the speed of footage operations rose 28.5% and non-footage operations 16.3%, pipe make-up time dropped 3.7% while make-up norms were exceeded by 40% — all without increasing well cost (Up-Pro, Integral Russia). The project's planned target was a 15% cut in complication-related NPT; in the words of trial participant Ramil Bariev, 'we planned for 15% growth when testing the project, and achieved much more.'

Support-center effects are published separately: geosteering saves 3–5% of rig time (JPT/SPE), and the efficiency coefficient of placing horizontal wellbores within the pay zone rose from 60% to over 90% after GeoNavigator launched (Fontanka). Company-wide digitalization economics are also disclosed: the 'Asset of the Future' pilot delivered a 1.2 billion ruble effect, and the company expected 5–6 billion rubles per year from digital technologies in exploration starting in 2025 (ComNews).

Important framing: all the figures above come from the company and its contractors, with no independent audit. The trial results were obtained on six wells of a single asset — pilot statistics, not fleet statistics; '−15% NPT' is the program's planned value, described by participants as 'exceeded' but without a published fleet-wide actual. Geosteering and 'Digital Rig' effects cannot be added together — they partly overlap within the same rig time.

In our view, the most valuable thing about this case is its metric discipline: the company measures impact in days per well, NPT percentages, and meters of wellbore inside the pay zone — quantities that convert directly into cost per meter drilled. That distinguishes the project from the typical 'we deployed AI on a rig' story with no operational numbers. A second observation: eight years of evolution — from the 2011 center through the six technologies of 2017 to a robotic rig — show that the 'unmanned rig' is built in layers on top of a decade of data work, not bought as a finished product; replicating the top layer without the ones beneath it will not, in our view, work. An indirect confirmation of maturity is the interest of Middle Eastern national oil companies in this stack: technologies proven on the company's own fields became an argument in international negotiations before 'sovereign oilfield software' became a mainstream topic.

Lessons learned

1. The right digital-drilling metrics are non-productive time and days per well, not abstract 'uptime': −6 days per well in field trials converts directly into cost per meter drilled.
2. Prediction horizon depends on the complication type: minutes for sudden failures, 1.5+ hours for developing stick/slip vibrations — claims of 'predicting everything hundreds of hours ahead' contradict the physics.
3. Centralized support centers solve the talent problem: GeoNavigator engineers steer 700+ complex wells a year from across the country — Yamal to the Sea of Okhotsk — and pay-zone placement rose from 60% to 90%+.
4. Combining on-rig automation with centralized predictive analytics beats either part alone — but their effects overlap, so the percentages must not be added together.
5. A six-well pilot with targets achieved by well five is an honest trial format: target metrics are fixed before the start, not curve-fitted afterward.
6. Honest digitalization economics in oil and gas run to billions of rubles per pilot asset (1.2B RUB for 'Asset of the Future'), not tens of billions per single technology.
7. The robotic rig is the top of a layered cake: beneath it lie a decade of support centers, in-house geosteering software, and years of IoT data; the top layer doesn't work without the ones below.

Frequently asked questions