This study’s main objective is to underscore machine learning’s role in exploring heat transfer analysis across various engineering challenges and practical problems. The continuous development of machine learning techniques has significantly enhanced computational precision and efficiency. The present study aims to address the heat transfer enhancement due to the presence of hybrid nanoparticles and thermal radiation in a flow of Jeffrey fluid over a cylinder embedded in a porous medium. The Tiwari-Das model develops the governing Jeffrey Hybrid Nanofluid problem, which contains nonlinear Partial Differential Equations. The considered hybrid nanoparticles are \(A{l}_{2}{O}_{3}\) and \(CuO\) suspended in a base fluid of ethylene glycol. The dimensional governing equation is transformed into a dimensionless form by applying suitable non-similar transformations. The dimensionless partial differential equations are truncated using the Local Non-Similarity method up to the third level of truncation. The resulting system of ordinary differential equations is solved using MATLAB’s built-in function bvp4c to obtain the reference solution. An approximate solution is computed using an Artificial Neural Network simulation. The predicted solutions from the Artificial Neural Network and the reference solution are compared and found to be well-fitted. The best fitness is obtained at different iterations 299, 372, and 268 with performance metrics [ \({7.19\times 10}^{-10}\) , \({5.50\times 10}^{-10}\) , and \({6.74\times 10}^{-10}\) ] for the physical parameter \(Pr\) . Permeability plays a crucial role in flow through porous media, that is, soils, biological tissues, or engineered porous materials. A higher permeability means that the material allows fluids to pass through more easily, while a lower permeability resists fluid flow. Therefore, increasing the permeability leads to an enhancement in the fluid velocity owing to reduced flow resistance.
