This study presents an enhanced coordinate partitioning method based on absolute coordinates, which incorporates iterative refinement into the update process of dependent coordinates. The goal is to improve both numerical stability and computational efficiency. The approach combines Newton-Raphson iteration with iterative refinement to accurately compute the dependent coordinates. To assess the method’s effectiveness, it is compared against a semi-recursive formulation based on relative coordinates through numerical simulations in MATLAB. Two multibody systems are used for validation: a standard four-bar linkage and a more complex six-body planar mechanism. The results demonstrate that the proposed global coordinates partitioning with iterative refinement strategy achieves accuracy comparable to the relative coordinates semi recursive formulation while significantly reducing computational time. These findings highlight the method’s scalability, efficiency, and suitability for simulating complex multibody systems with topologies.