MULTI-CRITERIA OPTIMIZATION FOR SELECTING THE BEST CONFIGURATION OF COMPUTER SYSTEMS
DOI:
https://doi.org/10.35546/kntu2078-4481.2025.4.2.23Keywords:
genetic algorithms, multi-criteria optimization, artificial intelligence, configurator, automated selection of components, budget constraints; compatibility.Abstract
Self-assembly of a computer is a justified and profitable strategy for many users, it provides significant advantages over purchasing a ready-made system. In particular, it is the best cost-effectiveness and savings, perfect adaptation to the user's needs, the ability to upgrade, repair, and control the quality of components yourself. Specialized recommendation systems – online configurators, which allow you to simplify and secure the process of selecting components as much as possible, can help you assemble a computer yourself. This article is devoted to the development of a hybrid recommendation system, the algorithm of which combines natural language processing methods, heuristic genetic search, and generative capabilities of a neural network model. Based on the analysis of scientific and technical literature, the problem of component selection is formalized as a multi-criteria optimization problem. The solution of this problem is presented using a genetic algorithm. Thanks to the genetic algorithm, the system is able to find close to optimal configurations from a large combinatorial space of options, and the use of ChatGPT to generate explanations improves the quality of interpretation of the results for the end user. The study presents the architecture of the proposed method for automated selection of components for computer systems, which consists of several interacting components, namely, a user interface, a natural language query processing module, an optimization module (genetic algorithm), a response generation module (based on ChatGPT), and a relational database of components. The article presents the main steps of the configurator algorithm, each of which describes the initial input data and output data. The results of the study can be used by higher education institutions and training centers to modernize curricula and improve the quality of training of specialists in the field of software engineering.
References
Winarno, M. (2018). Design and development of computer specification recommendation system based on user budget with genetic algorithm. International Journal of New Media Technology, 5(1), 25–29. https://doi.org/10.31937/ijnmt.v5i1.814
Mishra, S., Bane, S., Pandit, R., & Phadnis, N. (2021). PC configuration and component recommendation system. International Journal of Computer Applications, 183(1), 5–8. https://doi.org/10.5120/ijca2021921411
Veres, O., Ilchuk, P., & Kots, O. (2023). Information system for analysis of hardware computer components. In Proceedings of the 2023 IEEE International Conference on Computer Science and Information Technologies (CSIT) (pp. 1–4). IEEE. https://doi.org/10.1109/CSIT61576.2023.10324147
Cooler Master. (n.d.). Power supply calculator. https://www.coolermaster.com/en-global/power-supply-calculator/
Ehrgott, M. (2008). Multiobjective optimization. AI Magazine, 29(1), 47–57. https://doi.org/10.1007/978-0-387-76635-5_6
Березький, О. М., Теслюк, В. М., Дубчак, Л. О., Мельник, Г. М., & Батько, Ю. М. (2022). Дослідження і проектування комп’ютерних систем та мереж: навчальний посібник. Тернопіль: ЗУНУ.
Rosenberg, R. S. (1970). Stimulation of genetic populations with biochemical properties: I. The model. Mathematical Biosciences, 7(3–4), 223–257. https://doi.org/10.1016/0025-5564(70)90126-4
Rosenberg, R. S. (1970). Simulation of genetic populations with biochemical properties: II. Selection of crossover probabilities. Mathematical Biosciences, 8(1–2), 1–37. https://doi.org/10.1016/0025-5564(70)90140-9
Mitchell, M. (1999). An introduction to genetic algorithms (5th printing). Cambridge, MA: MIT Press.
Katoch, S., Chauhan, S. S., & Kumar, V. (2021). A review on genetic algorithm: Past, present, and future. Multimedia Tools and Applications, 80(5), 8091–8126. https://doi.org/10.1007/s11042-020-10139-6
Пиріг, Я., Климаш, М., Пиріг, Ю., & Лаврів, О. (2023). Генетичний алгоритм як інструмент для вирішення задач оптимізації. Інформаційно-комунікаційні технології, електронна інженерія, 3(2), 95–107. https://doi.org/10.23939/ictee2023.02.095
Wang, Q. J. (1997). Using genetic algorithms to optimise model parameters. Environmental Modelling & Software, 12(1), 27–34. https://doi.org/10.1016/S1364-8152(96)00030-8
Shevchenko, S., Zhdanova, Y., & Bilous, M. (2025). Automation of class scheduling as a necessary condition for optimization of the functioning of an educational institution. In Technical, agricultural and mathematical sciences: Scientific trends, problems and ways of their development: Collective monograph (pp. 296–309). Boston: Primedia eLaunch. https://doi.org/10.46299/ISG.2025.MONO.TECH.2
Sineglazov, V., Ryazanovskiy, K., & Chumachenko, O. (2020). Multicriteria conditional optimization based on genetic algorithms. System Research and Information Technologies, 3(3), 89–104. https://doi.org/10.20535/SRIT.2308-8893.2020.3.07
OpenAI. (n.d.). Introducing ChatGPT. https://openai.com/index/chatgpt/
Резніченко, І. В., Негоденко, О. В., & Шевченко, С. М. (2024). Дослідження програмних та технічних засобів для реалізації технології віртуальної реальності. Зв’язок, 1(1). https://doi.org/10.31673/2412-9070.2024.014147
Westerfield, G., Mitrovic, A., & Billinghurst, M. (2015). Intelligent augmented reality training for motherboard assembly. International Journal of Artificial Intelligence in Education, 25(2), 157–172. https://doi.org/10.1007/s40593-014-0032-x
