The scheduling of mechanical job shops may be optimized, which is a significant approach to boosting production effectiveness. Based on the description of the job shop scheduling problem in this paper, a mathematical model is built with the objective function of minimizing the maximum completion time. The vector evaluation genetic algorithm, which samples the edge region, and the adaptation function, which completes the sampling of the core region, are both offered as improvements for the differential evolutionary algorithm focused on job shop scheduling. After the sampling has been encoded, the best scheduling solution is sought using a sequential differential strategy. The modified HEA-SDDE algorithm’s maximum completion time for the actual scheduling scenario of K’s job shop is decreased by 12.4%, and the posting rate of the best solution to the ideal solution reaches 0.516.