ANALYSIS OF ALGORITHMIC VULNERABILITIES AND ARCHITECTURAL LIMITATIONS OF STATIC BID EVALUATION MODELS IN PUBLIC PROCUREMENT SYSTEMS
DOI:
https://doi.org/10.32689/maup.it.2025.4.15Keywords:
decision-making algorithms, risk-oriented ranking, public procurement software systems, reputation metrics, declarative arbitrage, adaptive algorithms, software architectureAbstract
The research objective is the development and algorithmic substantiation of a risk-oriented method for optimizing the winner selection procedure in public procurement software systems. This is achieved by enhancing the mathematical model for adjusted price calculation, formalizing risk parameters, and integrating participants' reputation characteristics directly into the auction's algorithmic core to increase the system's resilience to manipulative behavior. Methodology. The research methodology is based on mathematical modeling, algorithmic analysis, decision theory, and software engineering. It also includes experimental simulation of ranking algorithm performance under conditions of information asymmetry and strategic interaction between electronic auction participants. Scientific Novelty. The scientific novelty lies in the formalization of the "declarative arbitrage" strategy as an algorithmic vulnerability of static ranking models. Furthermore, the adjusted price model is improved by integrating a risk-oriented trust multiplier, calculated based on verified contract fulfillment history, which dynamically influences the decision-making process. Conclusions. Experimental simulation results confirmed that the proposed method negates the economic benefits of manipulative strategies and makes virtuous behavior mathematically more advantageous than dumping. The practical significance of the work lies in the potential implementation of the method as a software module within the service-oriented architecture (SOA) of e-procurement systems with high reliability and scalability requirements.
References
Корнійчук О. В., Граф М. С. Дослідження типів інформаційних атак пов'язаних з блокчейном. Сучасні тенденції розвитку системного програмування : матеріали конф., м. Київ : Нац. авіац. ун-т, 2022. С. 142–143.
Корнійчук О. В., Граф М. С. Порівняльний аналіз оптимізаційних моделей в системах державних закупівель. Інформаційні технології та програмування : матеріали конф., м. Черкаси, 2023. С. 142–143.
Про публічні закупівлі : Закон України від 25.12.2015 № 922-VIII : станом на 01.01.2024. Законодавство України. URL: https://zakon.rada.gov.ua/laws/show/922-19
Artificial Intelligence in Public Procurement : OECD Public Governance Reviews. Paris : OECD Publishing, 2024. 120 p.
Bakhshi R., Endriss U. Generative AI in Principal-Agent Models: Mitigating Adverse Selection and Moral Hazard. arXiv. 2025. URL: https://arxiv.org/abs/2502.12969
Bauhr M., Czarnitzki D. Need or greed? Conditions for corruption in public procurement. Governance. 2020. Vol. 33, no. 4. P. 1–19.
Behzadian M. et al. PROMETHEE Methods in Supply Chain Management: A Systematic Literature Review. Expert Systems with Applications. 2023. Vol. 215. Art. 119321.
Dmytryshyn B., Pikhotskyi V. Digital audit as a tool for combating corruption in public procurement. Economics and Law. 2023. No. 3. P. 45–52.
Fazekas M., Blum J. R. Improving Public Procurement Outcomes: Review of Tools and the State of the Evidence Base. World Bank Policy Research Working Paper. 2021. No. 9690. 58 p.
Gallego J. et al. Corruption in Public Procurement: A Cross-National Analysis of the Literature. Journal of Public Procurement. 2021. Vol. 21, no. 1. P. 1–25.
Opricovic S., Tzeng G. H. Fuzzy Multi-Criteria Decision Making for Sustainable Supplier Selection. Cham : Springer, 2024. 280 p.
Saaty T. L., Vargas L. G. Models, Methods, Concepts & Applications of the Analytic Hierarchy Process. Cham : Springer, 2022. 475 p.
Taherdoost H., Brard A. Analyzing the Process of Supplier Selection Criteria and Methods. Procedia Manufacturing. 2019. Vol. 32. P. 1024–1034.
The Impact of ProZorro Procurement System on the Ukrainian Economy. London : European Bank for Reconstruction and Development, 2023. 45 p.
Wang R., Li C., Zhang K., Tu B. Zero-trust based dynamic access control for cloud computing. Cybersecurity. 2025. Vol. 8. Art. 12. URL: https://cybersecurity.springeropen.com/articles/10.1186/s42400-024-00320-x
Zhang Y. et al. Robust Online Learning with Private Information: The Explore-Exploit-Punish Algorithm. arXiv. 2025. URL: https://arxiv.org/abs/2505.05341
