The paper addresses the parameter estimation problem in nonlinear dynamic biochemical pathways, which is formulated as a nonlinear programming (NLP) problem with differential-algebraic constraints. Traditional local optimization methods often fail due to the ill-conditioned and multimodal nature of these problems. To overcome this, the authors explore several state-of-the-art deterministic and stochastic global optimization (GO) methods. A case study involving the estimation of 36 parameters in a nonlinear biochemical dynamic model is used to benchmark these methods. Among the stochastic methods, Evolution Strategies (ES) are found to be the most successful, despite not guaranteeing global optimality. The paper highlights the importance of combining optimization with suitable simulation modules for designing metabolic pathways and calibrating models to experimental data. The results demonstrate that ES methods, particularly the Stochastic Ranking Evolution Strategy (SRES) and Unconstrained Evolution Strategy (uES), can efficiently and robustly solve the inverse problem, achieving near-exact results with a computational effort that is manageable for practical applications. The authors conclude that ES methods are the most competitive stochastic optimization techniques for large and complex parameter estimation problems in biochemical systems.The paper addresses the parameter estimation problem in nonlinear dynamic biochemical pathways, which is formulated as a nonlinear programming (NLP) problem with differential-algebraic constraints. Traditional local optimization methods often fail due to the ill-conditioned and multimodal nature of these problems. To overcome this, the authors explore several state-of-the-art deterministic and stochastic global optimization (GO) methods. A case study involving the estimation of 36 parameters in a nonlinear biochemical dynamic model is used to benchmark these methods. Among the stochastic methods, Evolution Strategies (ES) are found to be the most successful, despite not guaranteeing global optimality. The paper highlights the importance of combining optimization with suitable simulation modules for designing metabolic pathways and calibrating models to experimental data. The results demonstrate that ES methods, particularly the Stochastic Ranking Evolution Strategy (SRES) and Unconstrained Evolution Strategy (uES), can efficiently and robustly solve the inverse problem, achieving near-exact results with a computational effort that is manageable for practical applications. The authors conclude that ES methods are the most competitive stochastic optimization techniques for large and complex parameter estimation problems in biochemical systems.