<p>The challenge of intelligent manufacturing necessitates the effective alignment of production capacity with customer demands to optimize production planning. Traditional scheduling methods often struggle with computational efficiency and balancing multiple objectives in large-scale production settings. To address these challenges, this paper introduces a dual-archive integrated Multi-objective Grey Wolf Optimizer (DIMGWO) to solve the hybrid flow shop problems with missing operations (HFSMO). The DIMGWO enhances traditional Multi-objective Grey Wolf Optimizer by incorporating a Bernouilli shift method, dynamic population segmentation, and subpopulation leader selection to improve convergence and diversity. Additionally, the algorithm includes innovative mechanisms such as reinitialization of inferior individuals, local search enhancement for superior individuals, adaptive strategy adjustment and population division. The effectiveness of DIMGWO is demonstrated through empirical evaluations on benchmark test functions, multi-objective algorithm performance metrics and its application in real world manufacturing scenarios, particularly in coke production scheduling. The experiments show the algorithm’s performance superiority, which not only demonstrates its potential in complex production environments but also increases the availability of high-quality scheduling schemes for decision makers. The algorithm shows superior performance metrics that highlight its potential for broader adoption in complex production environments and increases the availability of high-quality scheduling schemes for decision makers.</p>