The impact of artificial intelligence on improving management efficiency in the oil and gas industry

Fastovich Vladimir Vladimirovich ORCID: 0009-0000-9104-2078 Postgraduate student; School of Public Administration; Lomonosov Moscow State University 119234, Russia, Moscow, Ramenki district, Lomonosovsky prospekt, 27 K. 4 vladimir.fastovich@yandex.ru

Artificial intelligence (AI) is an advanced and comprehensive discipline that unites various fields of knowledge (including computer science, statistics, neurology, social sciences, philosophy, psychology, bionics, etc.). The scope of AI is diverse, and covers machine learning, robotics, recognition and processing of languages and images. AI functions are also diverse: recognition, cognition, analysis of various information, its processing and, subsequently, making rational decisions on this basis, which makes it possible to simplify functional tasks in various professions. The main idea of AI is to trust machines to make independent judgments, fully or partially simplify, or even replace, the human decision-making process, as well as maximize the effectiveness and benefits of problem solving ^[1].

As such technologies are rapidly developing in the direction of digitalization, informatization and intelligence, they are finding applications in an increasing number of important applications for the national economy, for example, to maintain the country's economic stability and oil security, AI is also rapidly penetrating the oil and gas industry in areas such as finance, economics and industry.

AI is an important factor influencing the improvement of operational efficiency in the oil and gas industry, and its application is becoming more widespread, especially in the field of oil exploration and development (Fig. 1).

Fig. 1. The penetration of AI technologies in the oil and gas industry

Figure 1. Areas where AI is being implemented in the oil and gas industry

Source: calculated by the authors of: Unlocking the Potential of AI in the Oil and Gas Industry. IBM institute for business value. June, 2023.Pp. 1-25. URL: https://www.coursesidekick.com/management/4198746 (date of access: 03/19/2024).

According to the analysis of the size and market share of AI in the oil and gas industry from 2024 to 2029, it is expected that the size of the AI market in the oil and gas industry will amount to $2.98 billion in 2024 and reach $5.17 billion by 2029, during the forecast period (2024-2029), the CAGR will be 11.68%^[1].

In the exploration and development of oil and gas, which is the main area of traditional energy, geologists and exploration teams are constantly facing new challenges. For example, Heavy crude oil - crude oil illuminates 20°AP or less ^[2].

It has very viscous physical properties and is difficult to flow, but it accounts for 70 of the world's crude oil reserves, and it is difficult to develop^[2].

Currently, China is a global consumer of oil and gas resources and a producer of products based on the processing of this type of raw material ^[3]. China's proven heavy oil reserves amount to about 4 billion tons, which is more than 20% of China's total oil resources. Reserves of superheavy heavy oil with a viscosity of more than 50,000 MPa·s account for 14% of the total reserves. Currently, the recovery rate of heavy oil in China using the existing exploration and production technology is low and amounts to only 3-5%. 92% of the oil and gas fields explored in China have reserves at depths of more than 1,200 meters^[3]. The exploration and production of oil resources face challenges in terms of the quality and depth of production. The use of AI exploration in oil and gas exploration and development has become a hot topic and a trend in the industry.

The global oil and gas industry is stimulating the use of AI to improve the efficiency and effectiveness of management, as well as optimize the oil exploration process. The introduction of AI can also help to digitize records and automatically analyze the collected geological data and diagrams to perform a comprehensive analysis to identify potential problems ^[4], such as identifying oil and gas fields that are most likely to contain oil reserves to significantly improve oil and gas production, reduce costs and risks, and increase exploration and development. complex oil and gas deposits.

The industry implementation of AI technologies is rather uneven ^[5]. The main players in the AI industry are Google, Cisco, Microsoft, Intel, IBM, and Facebook. Oil and gas companies have organized a multinational collaboration with them to apply a range of AI technologies in the oil industry and have achieved results in step-by-step research and results in practical application^[4]. (Table 1).

Table 1 / Table 1

The current state of AI cooperation between the world's largest oil companies and oilfield service companies

The current state of AI collaboration between the world's largest oil companies and oilfield services companies

Sources: Compiled by the author based on: The Open AI Energy Initiative welcomes seven new participants. Ads. C3. AI. 2022. URL: https://c3.ai/the-open-ai-energy-initiative-welcomes-seven-new-members /. Scott. Fuzzy logic: definition, meaning, examples, and history. Automated investment. Investopedia.2023 URL: https://www.investopedia.com/terms/f/fuzzy-logic.asp.z

Sources: compiled by the author on: The Open AI Energy Initiative Welcomes Seven New Members. Announcements. C3. AI. November, 2022. URL: https://c3.ai/the-open-ai-energy-initiative-welcomes-seven-new-members/. Gordon Scott.Fuzzy Logic:Definition, Meaning, Examples, and History. Automated Investing. Investopedia. April,2023. URL: https://www.investopedia.com/terms/f/fuzzy-logic.asp.z

In 2014, the Spanish industrial company Repsol and the American technology company IBM jointly developed the world's first "Cognitive Technology" for oil industry applications. It is specifically used to optimize reservoir production and investment in new oil, and in the area of strategic decision-making opportunities^[5]. This technology can analyze information materials such as hundreds of thousands of documents and reports. By analyzing various combinations of data, such as seismic images, oil reservoirs, objects, etc., it can also introduce new factors in real time that need to be considered, such as: economic stability, political situation, natural disasters, etc., conduct targeted analysis and modeling to reduce the risks of oil and gas production^[6].

ExxonMobil cooperates with the Massachusetts Institute of Technology in the development of robots with artificial intelligence for ocean exploration, Schlumberger is developing an intelligent resource-sharing platform, the DELFI Cognitive Research and Development Environment^[7]; Cooperation between the oilfield services company BHGE and NVIDIA to promote the use of AI in the oil and gas sector shows that the role of AI technology is increasing for impact on the economic benefits of the oil and gas industry.

As of the end of 2023, Shell has deployed more than 160 AI projects in its oil and gas supply chain, achieving efficient data collection and modeling while significantly reducing natural gas production costs in exploration and drilling projects ^[6].

AI-based solutions are also being used in Russia. The cumulative economic effect of them over 6 years in SIBUR amounted to more than half of 45 billion rubles. AI is used at a wide variety of decision-making points: in 2023, SIBUR received more than 70 AI-based hypotheses that could potentially increase labor productivity, speed and accuracy of decision-making, and prevent deviations in the operation of equipment and personnel. They have also been added to solutions using large language models (LLM)^[8].

AI provides an informed analysis of demand in the oil and gas market.

The AI model can reveal complex patterns and correlations that analysts might miss by analyzing large amounts of historical data, market trends, uncertainty and a culture of geopolitical conflict, and even complex social media sentiment.

For example, using the Python language, industry researchers analyzed "What are the largest oil refining countries in 2023?" using the following code to visualize the research data:

import pandas as pd

data = {

'Country': ['USA', 'China', 'Russia', 'Saudi Arabia', 'Canada'],

'Oil Processing Volume (million barrels)': ^{[120, 90, 80, 75, 70]}

}

df = pd.DataFrame(data)

df_sorted = df.sort_values(by='Oil Processing Volume (million barrels)', ascending=False)

top_countries = df_sorted.head()

top_countries.

The data of the largest oil refining countries in 2023 can be obtained using a computer (Table 2):

Table 2 / Table 2

The leading countries in terms of oil refining in 2023

Sources: compiled by the author in Python

Source: compiled by author using Python

Thanks to the data obtained, it is known that these countries have the largest volume of oil refining in 2023, with the United States taking the first place. These data not only provide an overview of the global refining situation, but also highlight the important role these countries play in the oil industry.

The steps for using AI to integrate resources in this case are as follows:

Identification of the data source. For example, let's select the website of the U.S. Energy Information Administration (EIA) to obtain oil production data, which may include information on the rate of oil refining in each country.

Data collection. Currently, there is a global transition to the digital format of information presentation ^[7].

Extract relevant data from relevant pages and structure this data for analysis.

Data preparation. For example, you can extract data on oil refining rates by country and region from the EIA website. In situations where it is not possible to directly scan web content or interact with external websites in real time to extract and process data due to environmental constraints, tools such as Beautiful Soup or Scrapy in Python ^[8] are used to scan data from the EIA website.

Data analysis. After extracting the data, it must be cleaned and prepared for analysis. It covers data type conversion, handling missing values, and structuring data into a useful format. Includes: web page parsing, data cleaning and preparation.

Analysis and visualization. For example, analyze the data to understand the rate of oil refining by country. It includes calculating statistics, identifying trends, and visualizing data to make it easier to understand.

Visualization. Export the data table.

AI is used to predict oil prices and demand changes through data processing, which allows companies to make strategic decisions based on the integration of various resources and presented data ^[9].

The course of research and application of AI in the field of well logging, geophysical exploration, field development and ground engineering:

Well logging data collection

Drilling processes in the oil and gas industry are complex and high-tech ^[10].

Due to the heterogeneity of oil and gas formations, the complexity of detection objects, as well as the diversification and complexity of logging conditions, there is an urgent need to explore new measurement methods and operating modes when collecting downhole formation parameters and transmitting logging data. Implement AI to achieve more accurate ^[11], efficient and secure operations and geological information detection.

In recent years, the use of AI in logging processing and interpretation has mainly focused on automatic depth correction, automatic report generation, intelligent stratification, curve reconstruction, lithology identification, logging image interpretation, reservoir parameter prediction, oil and gas estimation, and shear waves. speed prediction, detection of filling of cracks and cavities, etc.

Currently, oil and gas companies have developed commercial products for data collection and remote data logging. As the world's leading oilfield maintenance company, Schlumberger (SLB) has developed remote logging centers, intelligent reservoir testing, and Techlog well software with intelligent processing and interpretation capabilities that have been put into commercial use. The operating environment of the Techlog software includes TechData data management: loading and managing DLIS, LIS, XTF, NTI, LAS, WITSML, CSV, CGM, ASCII, SVG, EMF, JPG, GIF, TIFF, BMP, PNG, SEG-Y and other formats ^[12]. Techlog Data Display: Uses simple and flexible tables, well logs, histograms, etc. for interactive management, display, editing, and preprocessing of drilling data, etc^.[9] (Fig. 2).

Fig. 2. The visual interface of Techlog

Figure 2. Visual interface “Techlog”

Sources: Official website of Schlumberger, software products: Techlog. URL: https://www.slb-sis.com.cn/products-services/Techlog.html (date of access: 03/26/2024).

Schlumberger installed the first control valve in Norwegian waters in May 2000. Since then, more than 2,200 control valves have been installed in 26 countries around the world. Techlog technology has enabled the deployment of 11 data server centers and 14 remote registration centers worldwide with 108 operational engineers, allowing experts to work together and make decisions remotely. 20% of registration operations are performed by remote registration centers^[10].

Geophysical exploration of oil and gas fields

The main applications of artificial intelligence in geophysical exploration equipment are: detection of vibro-seismic devices, drones, seismic instruments, etc. ^[13].

The intelligent vibrator can adjust the power, frequency range, scanning time, phase and other parameters according to the specific conditions of the surface and deep seismogeological conditions of the working area, it is safe and environmentally friendly ^[14].

Intelligent drones for collecting geophysical data can perform high-precision terrain detection, risk assessment, node monitoring, data recovery, material delivery, rescue and other tasks.

Collection of geophysical research

Due to the continuous development of AI technologies, it is safe to assume that the geophysical collection will move to the stage of intellectual development after the digital development stage with the following characteristics: sensorless digitization, a high degree of automation with feedback, "robotic" core equipment, integrated operational procedures, predictable production dynamics and the implementation of some peripheral computing capabilities for big Data processing (Big Date) ^[15].

The geophysical research technology implements digital management of construction tasks, field workers, equipment and visualization, optimizes construction procedures, simplifies work procedures, and provides intelligent stimulation, real-time quality control, remote technical support, and management and dispatching.

Due to the increase in seismic data and the need to interpret this data, the use of AI has significantly increased the efficiency of processing and interpreting seismic data, while ensuring accuracy.

AI is used in the interpretation of seismic structures (including fault identification, layer interpretation, interpretation of the top and bottom of a rocky hill, interpretation of a riverbed or cave, etc.), identification of seismic data, prediction of reservoir parameters, seismic velocity reading and modeling, as well as microseismic data analysis, detailed explanation, and other aspects.

Ground—based engineering - predicting the risks of oil leaks and machine failures, promoting high safety standards in the industry.

The main effect of the introduction of AI in the oil and gas industry is to reduce losses and production costs ^[16]. By constantly monitoring the operating status of the equipment, AI algorithms can detect security threats and potential failures during oil and gas production. By identifying threats at an early stage and issuing a warning, rig operators can plan maintenance activities in advance, which in turn will help reduce the likelihood of risky events, including accidents, thereby creating a safer working environment and reducing financial losses when equipment is suspended for repairs.

For example, it is known that oil and natural gas have a high degree of danger both during field development and during further transportation and processing of raw materials due to the threat of ignition and release of toxic gases. Artificial intelligence systems can not only monitor toxicity and leakage levels and warn users about problems, but also automatically modify systems such as cooling and heating systems in a timely manner to ensure the safety of cargo storage from the threat of seasonal temperature fluctuations. This allows the corporation to anticipate and eliminate potential equipment failures before they occur ^[17].

Increasing profits by minimizing costs is a fairly weighty argument for oil and gas operators in determining their priorities. The widespread adoption of AI will also, in our opinion, lead to an increase in total production through successful mining optimization. The offshore oil and gas industry uses AI and data science to simplify access to the large amounts of complex data needed for oil and gas exploration and production. This allows for more accurate data analysis and improved use of existing infrastructure in order to avoid production shutdowns and downtime, which would inevitably lead to significant financial costs.

Sensors controlled by artificial intelligence monitor environmental conditions, equipment operation, and personnel movements during offshore drilling operations. The data is analyzed in real time, and in case of any deviations in the security system, alerts and corrective actions are activated to prevent potential incidents and improve emergency response.

An illustrative example of the above is the use of computer vision to optimize production, which allows for faster analysis of seismic and geological data, understanding reservoirs, and modeling to predict oil corrosion risks to reduce maintenance costs.

Especially for offshore drilling platforms: data reading of pumps, compressors and drilling equipment. The rig's sensors continuously monitor the condition of these critical components to improve risk warning capabilities.

AI is able to analyze satellite images, aerial photographs, and remote sensing data to identify signs of oil spills or pipeline leaks in the marine environment. All this allows us to act quickly by detecting these incidents at an early stage in order to mitigate the impact on the environment and prevent the spread of pollutants, as well as improve environmental safety and accelerate the implementation of emergency response measures.

Improving supply chain efficiency in the oil and gas industry

In supply chain management, Robotic Process Automation (RPA) robots can collect data from various sources such as suppliers, inventory systems, and demand forecasts to optimize purchasing decisions and maintain optimal inventory levels^[11].

Oil refiners can use AI models in their operations to predict consumer demand for petroleum products such as gasoline, diesel fuel, and jet fuel, leading to production optimization and efficient inventory management.

By using machine learning algorithms to continuously analyze data from the recycling process or monitor the structural integrity of pipelines, companies can optimize maintenance plans, extend the service life of pipeline equipment, and improve pipeline safety standards ^[18].

Strengthening audit

The impact of AI on the financial aspects of the oil and gas industry goes beyond its application to specific business tasks, such as credit scoring, fraud detection, financial report analysis, pricing and hedging, marketing, consumer behavior analysis, and algorithmic trading ^[19].

In particular, it helps financial management in planning production operations, such as the development and evaluation of financial services, and also maintains the link between production, sales and marketing operations ^[20].

Responding to fluctuations in market prices

Price volatility caused by market dynamics, which can change rapidly due to geopolitical events, economic factors, and supply and demand imbalances, is an inherent problem in the oil and gas industry. In particular, the analysis and forecasting of fluctuations in oil and gas prices using predictive modeling methods for decision-making in order to reduce operational risks (by receiving large amounts of data on prices in different periods, market trends and geopolitical indicators).

Improving the efficiency of market operations

Oil and gas companies can gain valuable data-driven insights to improve the business results of their production processes by integrating artificial intelligence software into their operations. Big data and AI can help companies better understand customer needs and preferences. By analyzing customer behavior data and market feedback, product positioning and marketing strategies can be optimized. For example, Abu Dhabi-based technology company Oil and Gas Holding Co. (nogaholding) has partnered with technology company AIQ to integrate and implement digital solutions and artificial intelligence into its mining operations. This will allow nogaholding to use the latest AIQ AI technology to increase operational efficiency^[12].We have improved management models, especially in the customer service area, by updating the software of carefully selected information, including structured documents, PDF files, handwritten notes, as well as audio and video files. Using chatbots and automated services, we have increased customer satisfaction and efficiency, optimized resource usage, reduced operating costs and, as a result, increased overall profitability.

The problems faced in the oil and gas sector due to the use of AI

Currently, the application of artificial intelligence in oil and gas exploration and development is facing many challenges, such as high data collection costs, serious data quality problems, complex business scenarios that cannot be driven solely by data, an immature R&D ecosystem, short usage time, slow response of support mechanisms ^[21].

Suggestions for improving the impact of AI on efficiency in the oil and gas industry

National management departments should formulate a plan for the development of the use of AI technologies. For example, the Chinese government has launched a series of strategic plans for artificial intelligence. For a more rational application of artificial intelligence in the oil industry, in order to promote the development of the industry, it is also necessary to formulate plans for digital transformation and development of AI technology applications based on national policies and plans related to the development of AI, combined with the business characteristics of the oil industry, as well as clarify the goals of intellectual development, clarify the content of basic technological research, areas of application, implementation plans and security measures , etc .

To create a joint innovative research center for the development of AI in the oil and gas industry. The list of responsibilities of the Center should include: development of big data and AI management systems and technologies for the oil and gas industry, and provision of consulting services, provision of system solutions in software and hardware products, technical support and services for oilfield companies for digital transformation and intellectual transformation.

Conclusion

AI helps oil and gas companies better understand the market environment, optimize solutions, and increase competitiveness. AI has developed rapidly in recent years and has applications in practice and in research of oil and gas fields. Some countries consider the development of AI as one of the means to increase the competitiveness of the oil and gas industry. The intellectualization of oil and gas exploration and development has become a hot topic and a trend in the industry. It is expected that it will significantly improve the efficiency and quality of oil and gas exploration and development operations in the fields of logging, geophysical exploration, drilling, field development, surface engineering, etc., reduce costs and risks, as well as increase the efficiency of complex oil and gas field development, the level of exploration and development in general.

The growth trends and forecasts of AI applications in the oil and gas industry continue to expand. AI is involved in the analysis of demand, oil and gas supply chains in order to reduce production risks, maximize efficiency and revenues from the management and operation of oil and gas enterprises.

Oil and gas companies have developed various commercial products for data collection and remote data logging. The use of these technologies will continue to drive the digital transformation and innovative development of the industry.

Overall, AI is bringing revolutionary changes to the oil and gas industry, increasing efficiency and reducing costs, as well as supporting the goals of environmental protection and sustainable development. However, this also entails a redefinition of skills and job roles, as well as new challenges related to data security and confidentiality.

^[1] Artificial Intelligence Report for the Oil and Gas Industry // Mordor Intelligence. URL: https://www.mordorintelligence.com/zh-CN/industry-reports/ai-market-in-oil-and-gas (date of access: 02/20/2024).

^[2] Heavy oil // Neftegaz. 2013. URL: https://neftegaz.ru/tech-library/energoresursy-toplivo/147780-tyazhelaya-neft /(date of access: 12/20/2023).

^[3] Sun Huanquan. Limited resources, unlimited innovation. Interview with the Chinese Association of Petroleum Enterprises.2022.URL: http://www.zgsyqx.com/list.asp?id=9230 (date of request: 12/22/2023).

^[4] Shell, C3 AI, Baker Hughes and Microsoft jointly launch the "Open Artificial Intelligence Energy Initiative" to jointly create an AI solution ecosystem that promotes the transformation of the energy industry // Businesswise. 2021. URL: https://www.businesswire.com/news/home/20210203006063/zh-CN / (date of access: 02/22/2024).

^[5] Schlumberger, Chevron, and Microsoft will team up to digitalize the oil industry. TACC. URL: https://tass.ru/ekonomika/6900650 (date of access: 03/20/2024).

^[6] IBM - Artificial Intelligence Breathes Life into the Oil and Gas Value Chain // IBM Institute for Business Value. June 9, 2023. Pp. 1-25.

^[7] Schlumberger: Collaboration with Yandex. Cloud will accelerate the digital transformation of the Russian energy sector. Russia Oil and Gas Magazine. April 12, 2021. URL: https://www.rogtecmagazine.com/category/russia-oil-gas-magazine-ru/?lang=ru (date of access: 11/20/2024).

^[8] SIBUR received more than half of the economic effect from digital transformation thanks to artificial intelligence // Cnews. March 14, 2024.URL: https://www.cnews.ru/news/line/2024-03-14_bolee_poloviny_ekonomicheskogo (date of request: 03/22/2024).

^[9] Techlog is a software package for petrophysical interpretation. Platforms and Application Software // Techlog. URL: https://slb-sis.com.cn/products-services/Tec/1.html (date of request: 03/22/2024).

^[10] Intelligent well completion is a harbinger of the era of digital well completion. URL: https://www.slb.com/zh-cn/resource-library/case-study/case-study-china/2023/20230921-ps-intelligent-completion (date of request: 11/15/2024).

^[11] Sudeep Srivastava. How AI Is Revolutionizing the Oil and Gas Industry – Nine Use Cases and Benefits // Appinventiv. February 13, 2024. URL: https://appinventiv.com/blog/artificial-intelligence-in-oil-and-gas-industry (date of access: 03/25/2024).

^[12] ADNOC’s AIQ collaborates with nogaholding for digital upstream solutions // Oil and Gas Middle East. December 09, 2022. URL: https://www.oilandgasmiddleeast.com/products-services/adnocs-aiq-collaborates-with-nogaholding-for-digital-upstream-solutions (date of access: 03/25/2024).