Russia Deploys a New AI System for Handling Petroleum Products

A Preventive System

Tomsk State University of Control Systems and Radioelectronics (TUSUR) has created Russia’s first system that processes live video feeds from cameras using machine learning to monitor personnel actions and the condition of equipment in real time.

The system automatically detects deviations from operating procedures, alerts personnel and generates short video clips of each event – 15 seconds before and after the incident. This allows operators to quickly confirm a violation and make informed decisions. The product was tested in the fall of 2025 and has already been deployed at Oilteam. During its first month of operation, the system identified more than 200 procedural deviations of varying severity.

The development is part of the Priority-2030 program, aimed at big data analysis, anomaly detection and prediction of technological failures. It represents a significant step in the digitalization of Russia’s oil production industry, particularly in improving the safe operation of automated oil loading stations and mitigating human-factor risks. The system strengthens safety oversight and reduces the likelihood of procedural violations during petroleum product handling. A decision-support module has also been created for tanker truck loading operations. In abnormal situations – if the automated station begins to malfunction – the system automatically blocks or stops oil flow, preventing hazardous incidents. Developers plan further enhancements, and a demo version is currently on display at the university’s Sovereign Technologies Technopark, which is open to the public.

Transforming Processes Through Digitalization

The adoption of digital technologies is having a profound and irreversible impact on Russia’s oil and gas industry. The most significant changes are occurring in data analytics and process automation. Internet of Things solutions are being deployed widely across the sector. In the medium term, the industry expects exploration and production costs to fall by 10–15% and project commissioning timelines to shorten by up to 40%. At the same time, according to Russia’s Ministry of Energy, the digital maturity of the oil and gas sector still lags behind that of pharmaceuticals, food processing and metallurgy.

By 2025, the pace and scale of digitalization across all industrial sectors in Russia have increased significantly and reached peak activity. The key growth drivers are higher production volumes and the withdrawal of foreign software and equipment suppliers from the market. Additional catalysts include the need to meet import substitution targets and comply with critical information infrastructure requirements. Traditional automation goals remain relevant: improving business efficiency, reducing costs, and ensuring product quality and equipment reliability

Dmitry Vasilyev

senior partner at the League of the Digital Economy

Aggregated market experience shows that digitalization and automation can reduce emergency shutdowns by around 20% and cut operating costs by 10–20%, while improving resource management. Russian companies have repeatedly reported positive outcomes from implementing AI-, IoT- and big data-based solutions.

Practical Industry Examples

Analysts often point to the Sakhalin-2 project, where a predictive system developed by AVEVA has been deployed. It monitors the condition of 100 units of rotating equipment in real time. As a result, the operator shifted from scheduled preventive maintenance to condition-based maintenance. The system compares current data with historical benchmarks, enabling personnel to make timely decisions on servicing and spare-parts procurement.

In early 2024, Transneft announced the start of testing artificial intelligence and machine learning technologies. The trials assessed their suitability for identifying patterns, anomalies and forecasting oil volumes in storage tanks and pipelines by analyzing changes in primary data. AI is already being used to support long-term forecasting.

Another notable example is a project by Gazprom Nedra, which won the Oil 4.0 competition last year. The digital platform was developed by TetraSoft. Before implementation, drilling management relied on basic models and data analysis algorithms. The project introduced AI algorithms, predictive analytics, big data analysis and modeling technologies. As a result, drilling processes were optimized and production losses linked to equipment failures were significantly reduced.

Overall, the Russian development reflects a broader shift toward smart production sites in oil extraction, demonstrating a clear example of applied AI delivering tangible industrial benefits. The manual-operations monitoring system enhances safety and has the potential to become a standard for automated oil loading stations.