This study examines electroosmotic-driven fractional Jeffrey blood flow with aggregated ZrO \(_2\) nanoparticles through a permeable time-varying multi-stenotic inclined artery. Critical influences such as thermal radiation, metabolic heat generation, buoyancy effects, and electroosmotic forces are integrated with nanoparticle aggregation to analyze their impact on blood flow behavior. The Caputo fractional derivative is employed to account for blood’s viscoelastic memory effects. The electric potential within the electric double layer is modeled by solving the Poisson-Boltzmann equation. Analytical solutions to the governing equations, transformed into dimensionless form, are derived using Laplace and Hankel transforms and expressed via Lorenzo-Hartley and Robotonov-Hartley special functions. Results show that fractional parameters reduce blood velocity and temperature, while higher stenotic height and electroosmotic forces enhance velocity. Nanoparticle aggregation decreases velocity and wall shear stress but raises temperature and the heat transfer coefficient. Artificial neural networks are used to predict wall shear stress and heat transfer coefficient with exceptional accuracy, achieving over 99.9% in both testing and cross-validation. These findings contribute to advancements in medical applications, including nanoparticle-based drug delivery, magnetically guided devices, thermal therapies, and precision blood flow monitoring.
