Dynamic random walk-based glider snake algorithm for optimization of vehicle suspension components

Vehicle components need to be lightweight in order to reduce fuel consumption, lower emissions, and maintain structural integrity. This study proposes a glider snake optimizer enhanced with the dynamic random walk (DRW) technique to improve convergence performance and solution quality in engineering design optimization. The proposed algorithm incorporates periodic learning mechanisms and quasi-oppositional population initialization to achieve a better balance between exploration and exploitation while avoiding premature convergence. The developed optimization approach was applied to the lower control arm of an automotive suspension system with the objective of minimizing structural mass while satisfying stress constraints. The optimization framework integrates parametric design variables with finite element analysis. The results demonstrate that the dynamic random walk-based glider snake optimizer achieved a minimum mass of 3.311 kg while maintaining the maximum stress at 144.5 MPa. Compared with the initial design, the proposed method provided a mass reduction of 21.00%. The findings indicate that the proposed DRW-based glider snake optimizer converges efficiently and can be considered an effective tool for lightweight automotive component design and emission reduction-oriented engineering optimization problems.