Parametric fluid mechanics study of vessel geometry and blood rheology: Structural absence of a Pareto trade-off

Authors

  • Tulus Tulus Department of Mathematics, Universitas Sumatera Utara, Medan, Indonesia https://orcid.org/0000-0002-5178-7806
  • Jonathan L. Marpaung Department of Mathematics, Universitas Sumatera Utara, Medan, Indonesia https://orcid.org/0000-0002-4645-6148
  • Nuraini A. Aziz Department of Mechanical Engineering, Universiti Putra Malaysia, Putrajaya, Malaysia https://orcid.org/0000-0002-9289-6478
  • Tulus J. Marpaung Statistics Program, Faculty of Vocational Education, Universitas Sumatera Utara, Medan, Indonesia
  • Suriati Suriati Department of Information Technology, Universitas Harapan Medan, Medan, Indonesia
  • Muhammad R. Syahputra Department of Mathematics, Universitas Sumatera Utara, Medan, Indonesia
  • Putri K. Nasution Department of Mathematics, Universitas Sumatera Utara, Medan, Indonesia

DOI:

https://doi.org/10.52225/narrax.v4i3.353

Keywords:

Blood flow simulation, stenosis, non-Newtonian fluid, reduced-order model, hemodynamic optimization

Abstract

Cardiovascular hemodynamics in narrowed arteries are heavily dictated by local geometry and complex fluid properties, but multi-objective optimization loops using full Navier–Stokes solves remain computationally prohibitive. The aim of this study was to evaluate a multi-objective parametric framework for non-Newtonian blood flow through stenosed arteries and determine whether competing trade-offs exist between geometric narrowing and fluid rheology. The framework combines a Gaussian-parametrized stenosis model with the Carreau–Yasuda rheological law under a reduced-order, lubrication-theory formulation. At each axial position, the model prescribes a parabolic velocity profile scaled by mass conservation, a kinematic approximation that replaces a coupled Navier–Stokes solve. Two separate parametric optimization analyses were conducted with an NSGA-II scheme over three engineering metrics (peak wall shear stress, pressure drop, and viscous energy loss). The first analysis varied stenosis severity and width with rheology fixed, whereas the second varied rheological parameter (μ₀, , ) with a severe stenosis fixed. Across four representative configurations, peak velocity and maximum wall shear stress depended on severity alone (0.57 m/s and 1.26 Pa at δ=0.3; 1.00 m/s and 3.80 Pa at δ=0.6), whereas pressure drop and energy loss also rose with stenosis width. Both algorithmic runs converged to a single corner of their mathematical domains rather than a spread Pareto front. Analytical arguments and uniform sampling grids confirmed that each objective changes monotonically with each isolated parameter, causing a single corner solution to dominate each domain. The framework demonstrates the structural behavior of the simplified fluid formulation rather than competing physical optima. An independent, transient 3D COMSOL simulation illustrated a proximal–distal pressure difference varying over the cardiac cycle (peak 3.5 mmHg) to highlight transient physical behaviors omitted by steady-state models. These findings suggest that reduced-order computational formulations structurally govern local fluid indices, demonstrating that geometry and rheology do not present a Pareto trade-off within this kinematic framework.

Downloads

Download data is not yet available.

Downloads

Published

2026-11-18

How to Cite

Tulus, T., Marpaung, J. L., Aziz, N. A., Marpaung, T. J., Suriati, S., Syahputra, M. R., & Nasution, P. K. (2026). Parametric fluid mechanics study of vessel geometry and blood rheology: Structural absence of a Pareto trade-off. Narra X, 4(3), e353. https://doi.org/10.52225/narrax.v4i3.353