<p>Geostatistical seismic inversion methods employ stochastic sequential simulation for model generation and perturbation. Typically, the outputs depend on iterative workflows, such as genetic algorithms, where the optimization process focuses on achieving the best correlation between synthetic and real seismic data. However, these methods can be computationally expensive due to the need for multiple property simulations, particularly when applied to regular grids, which often fail to accurately represent the true geological features of the subsurface. This study presents methodologies aimed at enhancing geostatistical inversion by integrating local and global sequential inversions. This approach reduces computational costs while effectively aligning seismic and well data. To achieve this balance of cost and benefit, several enhancements were made to the default genetic algorithm to optimize parameters. These enhanced solutions are referred to as the Historic Genetic Algorithm (HGA) and the Evolutionary Algorithm (EVO). The HGA utilizes the best results from previous iterations, incorporating them as trends, input data for simulations, and for correlation in subsequent iterations. In contrast, the EVO technique builds upon the same algorithm as the HGA but incorporates an evolving parametrization approach across generations. Additionally, these approaches can be applied to both stratigraphic and regular grids. Thus, this work represents a significant advancement in integrating geological consistency into geostatistical seismic inversion, overcoming the limitations of using regular grids to replicate complex geological patterns. The results presented highlight the advantages of this new approach, which significantly enhances computational performance by achieving high correlation coefficients and some of the shortest processing times.</p>