Robust-Fuzzy Optimization Model for Personalized Digital Advertising with a Digital Twin and Cognitive Intelligence Approach

This study introduces an intelligent optimization framework for personalized digital advertising that integrates robust-fuzzy modeling with cognitive insights derived from digital twin systems. The proposed model addresses the challenges of uncertainty, emotional variability, and dynamic consumer behavior by combining fuzzy demand estimation, robustness against data volatility, and real-time sentiment-driven adaptation. It optimizes resource allocation across multiple advertising channels while ensuring resilience and personalized engagement. To evaluate the model’s performance, exact and metaheuristic solution methods were applied, including GAMS for small instances, Particle Swarm Optimization, and the Bat Algorithm for large-scale cases. Experimental results demonstrate that incorporating digital twin feedback significantly enhances engagement efficiency and targeting precision, particularly under uncertain conditions. Additionally, sensitivity analyses, time-adaptive simulations, and robustness–performance frontier evaluations confirm the model’s flexibility and stability. This research contributes a comprehensive, scalable, and emotionally aware decision-support tool tailored for the complexities of modern marketing environments.