Oleksii Kizilov
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Oleksii Kizilov has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Andrii Kopp
National Technical University “Kharkiv Polytechnic Institute
2 shared publications - Dmytro Orlovskyi
National Technical University “Kharkiv Polytechnic Institute
2 shared publications - Mykhaylo Godlevskyi
National Technical University "Kharkiv Polytechnic Institute", Ukraine
1 shared publication - Olha Halatova
National Technical University "Kharkiv Polytechnic Institute
1 shared publication
Latest Publications
- Open Access
- 31 October 2026
- Published by Private Company Technology Center in Technology audit and production reserves
- Vol. 5 (2(91)) , 13-20
Development of information technology for AI-assisted quality assessment of business process models
The object of research is the procedure of assessing the quality of business process models using the judgments of Artificial Intelligence (AI) assistants. The problem addressed relates to ensuring the quality of Business Process Model and Notation (BPMN) models. This directly determines their suitability for analyzing, automating, and describing business processes, and remains relevant since business process modelling is a key component of Business Process Management (BPM), which provides tools for formalizing, analyzing, and improving organizational activities. This research considers the use of AI for assessing the quality of business process models, where the weights of BPMN element types are determined by AI assistants based on Large Language Models (LLMs) rather than by human experts. This paper examines the use of the use of rank-based weighting methods based on evaluations from multiple AI assistants. Information technology was implemented using the Python programming language and tested on a set of 3722 BPMN models obtained from the GitHub public repository. The experimental results confirm that the choice of weighting method has a significant impact on the final distribution of BPMN model quality ratings. The proportion of models assessed as “Very good” ranges from 6% to 21% depending on the ranking method used, except equal weights resulted in 87% of “Satisfactory” models. The results obtained demonstrate the value of using AI assistants to improve the objectivity, reproducibility and scalability of assessing the quality of BPMN models, providing a modern alternative to traditional approaches based on expert evaluations. This information technology can be used by organizations implementing or improving BPM systems to analyze the quality of BPMN models, and can be integrated into automated pipelines in BPMN repositories and decision support systems in order to assess business process modelling maturity. As this approach does not require experts to define weights, it is applicable in situations where resources are limited or where assessment results need to be revised frequently.
- 31 October 2026
- Published by Private Company Technology Center in Technology audit and production reserves
- Vol. 5 (2(91)) , 13-20
Open AccessDevelopment of information technology for AI-assisted quality assessment of business process models
The object of research is the procedure of assessing the quality of business process models using the judgments of Artificial Intelligence (AI) assistants. The problem addressed relates to ensuring the quality of Business Process Model and Notation (BPMN) models. This directly determines their suitability for analyzing, automating, and describing business processes, and remains relevant since business process modelling is a key component of Business Process Management (BPM), which provides tools for formalizing, analyzing, and improving organizational activities. This research considers the use of AI for assessing the quality of business process models, where the weights of BPMN element types are determined by AI assistants based on Large Language Models (LLMs) rather than by human experts. This paper examines the use of the use of rank-based weighting methods based on evaluations from multiple AI assistants. Information technology was implemented using the Python programming language and tested on a set of 3722 BPMN models obtained from the GitHub public repository. The experimental results confirm that the choice of weighting method has a significant impact on the final distribution of BPMN model quality ratings. The proportion of models assessed as “Very good” ranges from 6% to 21% depending on the ranking method used, except equal weights resulted in 87% of “Satisfactory” models. The results obtained demonstrate the value of using AI assistants to improve the objectivity, reproducibility and scalability of assessing the quality of BPMN models, providing a modern alternative to traditional approaches based on expert evaluations. This information technology can be used by organizations implementing or improving BPM systems to analyze the quality of BPMN models, and can be integrated into automated pipelines in BPMN repositories and decision support systems in order to assess business process modelling maturity. As this approach does not require experts to define weights, it is applicable in situations where resources are limited or where assessment results need to be revised frequently.
- Open Access
- 30 July 2024
- Published by National Technical University Kharkiv Polytechnic Institute in Bulletin of National Technical University "KhPI". Series: System Analysis, Control and Information Technologies
- No. 1 (11),p. 34-38
RESEARCH ON ERROR PROBABILITY ASSESSMENT IN USER PERSONAL DATA PROCESSING IN GDPR-COMPLIANT BUSINESS PROCESS MODELS
The only right strategy for businesses and government organizations in Ukraine and other countries that may face aggression is to recognize themselves as a potential target for cyberattacks by the aggressor (both by its government agencies and related cybercriminal groups) and take appropriate measures in accordance with the European Union’s General Data Protection Regulation (GDPR). The main purpose of the GDPR is to regulate the rights to personal data protection and to protect EU citizens from data leaks and breaches of confidentiality, which is especially important in today’s digital world, where the processing and exchange of personal data are integral parts of almost every business process. Therefore, the GDPR encourages organizations to transform their day-to-day business processes that are involved in managing, storing, and sharing customers’ personal data during execution. Thus, business process models created in accordance with the GDPR regulations must be of high quality, just like any other business process models, and the probability of errors in them must be minimal. This is especially important with regard to the observance of human rights to personal data protection, since low-quality models can become sources of errors, which, in turn, can lead to a breach of confidentiality and data leakage of business process participants. This paper analyzes recent research and publications, proposes a method for analyzing business process models that ensure compliance with the GDPR regulations, and tests its performance based on the analysis of BPMN models of business processes for obtaining consent to data processing and withdrawal of consent to user data processing. As a result, the probability of errors in the considered business process models was obtained, which suggests the possibility of confidentiality violations and data leaks of the participants of the considered business processes associated with these errors, and appropriate recommendations were made.
- 30 July 2024
- Published by National Technical University Kharkiv Polytechnic Institute in Bulletin of National Technical University "KhPI". Series: System Analysis, Control and Information Technologies
- No. 1 (11),p. 34-38
Open AccessRESEARCH ON ERROR PROBABILITY ASSESSMENT IN USER PERSONAL DATA PROCESSING IN GDPR-COMPLIANT BUSINESS PROCESS MODELS
The only right strategy for businesses and government organizations in Ukraine and other countries that may face aggression is to recognize themselves as a potential target for cyberattacks by the aggressor (both by its government agencies and related cybercriminal groups) and take appropriate measures in accordance with the European Union’s General Data Protection Regulation (GDPR). The main purpose of the GDPR is to regulate the rights to personal data protection and to protect EU citizens from data leaks and breaches of confidentiality, which is especially important in today’s digital world, where the processing and exchange of personal data are integral parts of almost every business process. Therefore, the GDPR encourages organizations to transform their day-to-day business processes that are involved in managing, storing, and sharing customers’ personal data during execution. Thus, business process models created in accordance with the GDPR regulations must be of high quality, just like any other business process models, and the probability of errors in them must be minimal. This is especially important with regard to the observance of human rights to personal data protection, since low-quality models can become sources of errors, which, in turn, can lead to a breach of confidentiality and data leakage of business process participants. This paper analyzes recent research and publications, proposes a method for analyzing business process models that ensure compliance with the GDPR regulations, and tests its performance based on the analysis of BPMN models of business processes for obtaining consent to data processing and withdrawal of consent to user data processing. As a result, the probability of errors in the considered business process models was obtained, which suggests the possibility of confidentiality violations and data leaks of the participants of the considered business processes associated with these errors, and appropriate recommendations were made.