Abstract: Delivery on time was one of the important goals of the production operations in make-to-order industries. When the production load of the manufacturing system was heavy， overtime work was the most commonly used effective measure to ensure the delivery time. In order to realize the no-tardiness delivery of order tasks and the optimal use of overtime hours， a multi-objective no-tardiness job shop scheduling problem considering overtime was proposed， and a mathematical model with the objective of minimizing the total overtime hours and the makespan was established. To solve the problem effectively， a multi-objective golden eagle optimizer was proposed. In the algorithm， a coding method was designed to map the continuous solution space to the discrete solution space of the problem， a two-stage decoding scheme was proposed， an elitist no-dominated sorting selection operator was integrated， and an adaptive offspring individual generation strategy was introduced. The effectiveness and superiority of the proposed algorithm were verified by comparison experiments on 23 modified job shop scheduling problem benchmarks as experimental objects. 

Key words: job shop scheduling, no-tardiness, overtime, multi-objective golden eagle optimizer

摘要： 按期交货是面向订单型制造企业生产运作所追求的重要目标之一。当制造系统生产负荷较重时，加班作业是保证交货期最常采用的有效措施。为实现订单的无拖期交付和加班时间的合理优化使用，以面向订单型制造系统为背景，提出了一种考虑加班的多目标无拖期作业车间调度问题，建立了以最小化总加班时间和最大完工时间为目标的数学规划模型。为有效求解该问题，提出了一种多目标金鹰优化算法。在算法中设计了连续解空间映射至问题离散解空间的编码方法；提出了两阶段解码方案；融入了基于精英保留策略的非支配排序选择算子；引入了自适应子代个体生成策略。以23个作业车间调度问题基准算例为实验对象进行了对比实验，实验结果验证了所提出算法的有效性和优越性。

关键词: 作业车间调度, 无拖期, 加班作业, 多目标金鹰优化算法