Browsing by Author "Khomenko, Volodymyr L."

Now showing 1 - 2 of 2

Formation of an Optimal Portfolio of Venture Projects
(Dnipro University of Technology, 2021) Korhina, Inna A.; Petrenko, Vitaliy O.; Khomenko, Volodymyr L.; Kulyk, Volodymyr O.
ENG: Purpose. Development of a method for forming an optimal portfolio of venture projects taking into account risks, uncertainty in initial data and limited financial resources. Methodology. To calculate the accuracy of forecasting prices, which are necessary for calculating the parameters of the stochastic optimization model for the formation of a portfolio of projects, we used the theory of random variables and regression analysis. The problem of choosing the optimal portfolio of venture projects was solved using stochastic mathematical programming. Findings. A model for creating an optimal portfolio of venture projects has been developed. It is a stochastic mathematical programming model that can be used to solve problems of investing in venture projects in the extractive industry. This model takes into account the risks associated with obtaining the expected income from the implementation of each venture project, the uncertainty in the initial data for calculating the income from the projects selected in the portfolio, as well as the limited funds required to finance the project portfolio. Originality. The stochastic optimization model for the formation of an optimal portfolio of projects, taking into account the peculiarities of venture projects, in particular, their high riskiness, has been significantly improved and adapted. Practical value. The proposed model for the formation of an optimal portfolio of venture projects can be used at mining enterprises, whose development strategy involves the implementation of innovative, high-risk projects. The use of this model in strategic planning will allow an enterprise to receive the maximum income from venture projects in the face of a lack of financial resources, as well as instability of the innovation market.
Using Machine Learning to Model Mechanical Processes in Mining: Theory, Practice, and Legal Considerations
(Engineered Science Publisher LLC, Knoxville, USA, 2025) Ratov, Boranbay; Pavlychenko, Artem; Kirin, Roman; Pashchenko, Oleksandr; Khomenko, Volodymyr L.; Tileuberdi, Nurbol; Kamyshatskyi, Oleksandr; Sieriebriak, Stanislav; Seidaliyev, Askar; Muratova, Samal
ENG: Artificial intelligence (AI) technologies, though critical for economic development, also pose risks of unpredictable outcomes and loss of control. Thus, a legal framework is necessary to regulate their use. International and state oversight is required to establish clear rules of conduct for all parties involved in AI relations, ensuring these technologies remain human-oriented and secure. In geological studies, AI can enhance the accuracy of predictions, such as improving the understanding of rock behavior during drilling. Machine learning methods, including linear regression and gradient boosting, have proven effective in predicting the mechanical properties of rocks, which helps optimize drilling operations and minimize risks like equipment damage. However, models must be fine-tuned to account for more complex dependencies, such as mineralogical characteristics. Despite the effectiveness of AI, challenges remain, including the need for high-quality data and the potential for overfitting in some methods. Incorporating AI studies into the geological code is crucial for effectively managing these technologies. By enhancing transparency, security, and accountability in AI systems, governments can mitigate risks while fostering innovation. In geology, AI’s potential for reducing drilling costs and improving safety, as well as its application to other areas like mining and construction, will drive significant advancements in scientific and industrial fields.