Article · Eastern-European Journal of Enterprise Technologies · June 25, 2025
Multi-criteria optimization of digital marketing for enterprises in the agro-industrial complex based on NSGA-III algorithm and machine learning
Olena Kryvoruchko, Zhansaya Abildaeva, Valeriy Lakhno, Цюцюра Микола Ігорович, Svitlana Tsiutsiura, Anna Kharchenko
Abstract
The object of this study is the process of optimizing digital marketing for agro-industrial enterprises under conditions of multi-criteria and uncertainty. A formal statement of the problem of optimizing marketing strategies for agricultural production has been given by using the genetic algorithm NSGA-III. A hybrid method was devised to solve the task of multi-criteria optimization of marketing strategies for agro-industrial enterprises. The method is based on the NSGA-III algorithm in combination with the XGBoost software library and adapted to industry constraints for marketing strategies in the agricultural markets of Ukraine and Kazakhstan Republic. This allows for the generation and interpretation of Pareto-optimal strategies taking into account such criteria as efficiency, coverage, return on investment (ROI), costs, and engagement. A cluster analysis of solutions has been performed; three characteristic scenarios were identified – balanced, cautious, and aggressive. Empirical validation by regression analysis demonstrated the high accuracy of the model, as well as its ability to extrapolate new solutions. In particular, the mean square error on the test sample was 0.0316 with the achieved coefficient of determination of 0.9041. The results confirm the effectiveness of the devised method to support decision-making under conditions of multi-criteria and limited resources. The proposed method was used as the basis for the development of software implemented in practice at enterprises of the agro-industrial complex. However, the scope of method application also includes the activities by other business entities that devise marketing strategies to achieve the efficiency of their activities
More by Valerii Lakhno
- Network-Aware Cloud Integration with Uncertainty-Gated Pareto Decisions: A Formal Cognitive Architecture with Exact Computational VerificationFuture Internet · 2026
- Cloud-Integration MCDM: Network-Aware Extension and Uncertainty-Gated AdmissionZenodo · 2026
- Developing the Methodological Competence of Future IT Engineers in the Republic of Kazakhstan Through Problem-Oriented LearningEducation Sciences · 2026
- АДАМИ ФАКТОРДЫ ЕСКЕРЕ ОТЫРЫП, АҚПАРАТТЫҚ-БІЛІМ БЕРУ ЖҮЙЕЛЕРІНІҢ АҚПАРАТТЫҚ ҚАУІПСІЗДІГІНІҢ ТӘУЕКЕЛДЕРІН БАҒАЛАУ МОДЕЛІНаучный журнал "Вестник Актюбинского регионального университета имени К Жубанова" · 2026
