Abstract
<jats:p>The coal industry is at its critical stage where implementation of advanced automation technologies becomes essential to ensure the competitiveness and safety of mining operations. This study analyzes the potential for adapting automated systems to control drilling processes that were originally developed for the oil and gas industry to the specific needs of coal mining. The global market for drilling automation is showing steady growth, with the volume of $4.66 billion in 2024 and a projected expansion to $9.19 billion by 2034, at an average annual growth rate of 7.32%. At the same time, the AI market in the oil and gas sector has grown from $5.305 billion in 2024, with a forecast to reach $15.010 billion by 2029, at an annual growth rate of 23.12%. Implementation of the machine learning technologies in oil companies’ drilling equipment has reduced unplanned downtime by 20%, lowered the operating costs by 10–15%, and increased the drilling speed by 26–30% in shale plays. A critical contribution of this research is the development of a methodology to assess the technological readiness of coal mining operations for integration of automated systems through a comparative analysis of geomechanical drilling parameters in various geological environments. It has been established that the correlation coefficient between the automation parameters in the oil and gas drilling and in coal mining is 0.78 for the Measurement While Drilling (MWD) systems, 0.83 for the SCADA platforms, and 0.62 for the machine learning algorithms, which determines a differentiated strategy for the technology transfer. The Chinese coal companies that integrated automated AI-based drilling systems have achieved a 40% profit margin while reducing their energy consumption by 2 million kWh per year at a single operation. The study identified a critical dependence of the implementation efficiency on the scale of operations, i.e. companies with the annual drilling volume of less than 80,000 m demonstrate a payback period of over 8 years, whereas for the volumes exceeding 150,000 m, this period is reduced to 3.8–4.5 years. A three-level model of barriers to the technology transfer has been developed with its criticality assessed using a five-point scale: cybersecurity of industrial control systems (index 4.9), lack of data to train the algorithms (4.7), and shortage of qualified personnel (4.5).</jats:p>