Abstract: In the actual production process of manufacturing enterprises, a production scheduling component was designed and developed to solve the problems of four-day rolling plan's dependence on personal experience, inability of Excel table planning data to directly interface with MES system, and delayed production information feedback. The "seeding machine" scheduling algorithm was adopted, combined with the characteristics of mixed-line production in enterprises, effectively solving key problems such as low scheduling efficiency, data silos, and delayed feedback. Through comparative analysis with traditional optimization algorithms such as genetic algorithm and simulated annealing algorithm, the superiority of the "seeding machine" algorithm in mixed-line production environment was verified. After more than half a year of practical application verification, the four-day plan production time was shortened from 4 hours to 1.5 hours, with efficiency improved to 60% above; the plan accuracy rate increased to 95%, providing practical tools for enterprise digital transformation and accumulating valuable experience.