3双目标蚁群算法设计-NSFC.docVIP

论文编号：A2015-0450 PAGE 1 混流无等待流水线批次批量生产计划方法 蒙秋男收稿日期：2015-07-15 收稿日期：2015-07-15 基金项目：国家自然科学基金, The National Natural Science Foundation of China; 国家科技支撑计划项目(2013BAF02B03), National Key Technology R D Program(2013BAF02B03). 作者简介：蒙秋男(1969-)，女，黑龙江省齐齐哈尔市人。大连理工大学副教授，博士。主要从事计划调度和成本管理等研究。E-mail: HYPERLINK mailto:code_name@163.com code_name@163.com （大连理工大学管理与经济学部，辽宁大连，116024） 摘要:在两阶段混流无等待流水装配环境下，为解决任务组批生产导致的订单按期交付能力弱，以及产品需求与部件供应无法准确衔接的问题，以最小化在制品库存成本和产品提前拖期惩罚为目标，建立了两阶段批次批量生产计划数学模型。设计了双目标蚁群求解算法，构造了每只蚂蚁对不同目标的偏重算子，以及基于批次的可行解生成方法和信息素更新机制，提高了解的局部和全局搜索能力。通过与NSGA-II进行对比，验证了本算法在同等时间内计算精度优于后者，为提高两阶段多条生产线组批生产计划的可执行性提供方法支持。 关键词:混流无等待流水线; 两阶段生产计划; 批次; 双目标蚁群算法 中图分类号：F406.2 文献标识码：A Two-stage Batch Production Planning for Mixed Flow of No-wait Assembly Lines Meng Qiunan, Bai Xue, Zhao Cong （Faculty of Management and Economics, Dalian University of Technology, Dalian 116024） Abstract: In mixed flow of no-wait assembly lines, it is difficult to deliver orders timely and supply parts accurately to the demands of the second stage in batch production. To solve above problems, a mathematical model of two-stage batch planning is formulated to minimize both WIP inventory cost and earliness/tardiness penalty of orders. Then a bi-objective ant colony optimization algorithm (BOACO) is designed to solve this model. An operator of preferring to different targets for each of ants, construction of feasible solutions and mechanism of updating pheromone based on the batch are presented to improve the local and global search ability of BOACO significantly. A series of experiments with different sizes indicate that BOACO outperforms NSGA-II within the same computation time. The results of the study show that the proposed method of batch production planning provides support for enhancing the performability of production planning in multiple production lines of two stages under batch production condition. Key Words: mixed flow of no-wait assembly line; two-stage production planni