This study explores the application of data-driven simulation (DDS) in resource allocation (RA) within high-mix, low-volume (HMLV) smart manufacturing systems. The authors develop a DDS-based decision support system (DDS-DSS) that incorporates two RA strategies: simulation-based bottleneck analysis (SB-BA) and simulation-based multi-objective optimization (SB-MOO). To enhance the performance of SB-MOO, a modified version of the NSGA-II algorithm, named NSGA-II-ML, is introduced, featuring a unique meta-learning mechanism. The proposed DDS-DSS also includes a post-optimality analysis using the DBSCAN clustering algorithm to derive actionable insights. A real-life marine engine manufacturing application is presented to demonstrate the effectiveness of the proposed system. The results show that NSGA-II-ML outperforms other algorithms in terms of the number of non-dominated solutions and hypervolume, particularly in medium and large-sized problems. Additionally, NSGA-II-ML achieved a 24% improvement in the best throughput found in the real case problem, outperforming SB-BA, NSGA-II, and DE. The post-optimality analysis provided valuable knowledge about the key decision variables influencing throughput.This study explores the application of data-driven simulation (DDS) in resource allocation (RA) within high-mix, low-volume (HMLV) smart manufacturing systems. The authors develop a DDS-based decision support system (DDS-DSS) that incorporates two RA strategies: simulation-based bottleneck analysis (SB-BA) and simulation-based multi-objective optimization (SB-MOO). To enhance the performance of SB-MOO, a modified version of the NSGA-II algorithm, named NSGA-II-ML, is introduced, featuring a unique meta-learning mechanism. The proposed DDS-DSS also includes a post-optimality analysis using the DBSCAN clustering algorithm to derive actionable insights. A real-life marine engine manufacturing application is presented to demonstrate the effectiveness of the proposed system. The results show that NSGA-II-ML outperforms other algorithms in terms of the number of non-dominated solutions and hypervolume, particularly in medium and large-sized problems. Additionally, NSGA-II-ML achieved a 24% improvement in the best throughput found in the real case problem, outperforming SB-BA, NSGA-II, and DE. The post-optimality analysis provided valuable knowledge about the key decision variables influencing throughput.