Knowledge Generation in Raw Material Industries
Abstract— The article presents a mathematical toolkit for forecasting the development of innovative technologies that can be used for cost-effective development of hard-to-recover hydrocarbon reserves. The proposed approach is a symbiosis of agent-based models, Bayesian networks, learning curves, statistical analysis, and numerical simulation modeling methods. These techniques have not been used as a set in research related to the oil and gas industry ever before. The practical significance of the modeling toolkit is based on the prioritization of measures aimed at innovative growth in the oil and gas sector. To test the approach, calculations are performed for three objects: complex structures of the Predpatomsky trough in eastern Siberia and the Republic of Sakha (Yakutia), deposits of the Pre-Jurassic complex in the south-east of western Siberia, and natural bitumen in the Republic of Tatarstan. The calculations confirm the working hypothesis about the practicability of forming technological clusters for efficient extraction of hard-to-recover hydrocarbons.