为求解车间作业调度问题,提出一种基于个体差异化自学习的改进教学算法.针对教学算法局部搜索能力不高的缺陷,提出学生不仅应向能力好的学习者学习,亦应进行有差异的自我学习.通过学习者的完工时间评估学生的学习能力,提出学习次数概念,并设计自学习算子,完善学生阶段的更新,提高算法的局部搜索能力.最后,对OR-Library中的标准仿真实例进行实验,结果表明改进教学算法是有效的,其在收敛精度和鲁棒性能上均有较好的提高. In order to solve the problem of job shop scheduling, this paper proposes an improved teaching algorithm based on individual difference self-learning.Aiming at the defect that teaching algorithm has low local search ability, it is suggested that students should learn not only from capable learners but also from different learners Self-learning.According to the completion time of learners, students’ learning ability is evaluated, the number of learning times is proposed, the self-learning operator is designed, the student stage is improved, and the local search ability of the algorithm is improved.Finally, the standard simulation in OR- Experimental results show that the improved teaching algorithm is effective, which has a good improvement in convergence accuracy and robustness.