急件订单干扰下考虑批量分割的虚拟单元重调度

发布时间：2018-06-12 18:52

本文选题：虚拟单元 + 重调度　；参考：《江苏科技大学》2017年硕士论文

【摘要】：虚拟单元动态调度和重调度是虚拟单元新的研究方向,并日益得到国内外学者的关注和重视。然而,许多学者在研究此类问题时,并未充分考虑重调度对于初始调度的干扰程度,且并未有效利用批量分割的方法缩减工件的完工时间和生产总成本。鉴于此,本文在深入研究虚拟单元重调度、干扰管理和批量分割相关国内外研究现状和相关理论的基础上,对急件订单干扰下融入干扰管理方法和批量分割思想的虚拟单元重调度问题进行了较为细致的研究。本文急件订单干扰下融入干扰管理方法和批量分割思想的虚拟单元重调度问题的研究可分为两个阶段,第一阶段为急件订单干扰下考虑序位相似性的虚拟单元重调度问题研究,在此阶段,着重解决急件订单干扰管理问题;第二阶段为虚拟单元模式下批量策略的确定问题研究,在此阶段,将批量分割的思想融入到急件订单干扰下的重调度问题当中,以期较优的柔性分批调度方案,确保在最大化资源利用的同时,缩减工件的完工时间和完工总成本。关于急件订单干扰下考虑序位相似性的虚拟单元重调度问题,本文结合急件订单特性,赋予急件订单更高的生产优先级。在急件订单干扰出现时,将急件订单与原虚拟单元常规订单的产品族和设备组进行相似性判断,在急件订单能够与原虚拟单元融合生产的情况下进行虚拟单元重调度。同时,设置了各工件工序可用机器集合和相应的加工时间集合,构建了以急件订单完工时间、系统总流程时间最短和序位相似性最大为目标的多目标非线性整数规划模型。针对模型自身特征,设计了遗传--蚁群相结合的组合优化算法求解模型。此外,借助干扰管理的方法,降低了急件订单干扰对生产调度的影响。关于虚拟单元模式下批量策略的确定问题,本文充分考虑时间和成本因素,提出了考虑批次启动时间的以总完工时间、总加工成本、总运输成本以及总拖期惩罚成本最小为目标的多目标非线性混合整数规划模型。为有效的求解模型,将遗传算法与工序优先级调度算法相结合,提出了基于工序的两阶段编码的优化遗传算法,该算法可以根据实际调度的需要将工件分割成具有柔性批量的多个子批,而且可以同时优化子批的加工工艺路径和加工排序方案。最后,本文在对HDZH机加工车间的生产现状及存在的问题进行详细分析的基础上,将所提出的急件订单干扰下融入干扰管理方法和批量分割思想的虚拟单元重调度问题策略应用到机加工车间舵系、轴系等产品的生产过程中,以验证本文所提出的方案在实际船舶企业生产中的应用价值和意义。研究结果表明,本文所提出方法具有较高的可行性和优越性。同时,本文的研究对船舶等大型生产企业优化生产流程具有一定的指导意义,在急件订单干扰时,为如何在生产成本和生产扰动之间的权衡决策提供了一定的理论支持。
[Abstract]:Virtual cell dynamic scheduling and rescheduling is a new research direction of virtual cell, and has been paid more and more attention by scholars at home and abroad. However, many scholars do not fully consider the interference degree of rescheduling to the initial scheduling, and do not effectively use the method of batch partitioning to reduce the completion time and total production cost of the jobs. In view of this, on the basis of in-depth research on virtual cell rescheduling, interference management and batch segmentation, this paper studies the current situation and related theories at home and abroad. In this paper, the rescheduling problem of virtual cell with interference management and batch segmentation is studied in detail. In this paper, the research of virtual cell rescheduling with interference management method and batch segmentation under urgent order interference can be divided into two stages. The first stage is the virtual cell rescheduling problem which considers the sequence similarity under the urgent order interference. In this stage, the interference management problem of urgent orders is solved emphatically. In the second stage, the determination of batch policy in virtual cell mode is studied. In this stage, the idea of batch segmentation is integrated into the rescheduling problem under the interference of urgent orders. It is expected that the flexible batch scheduling scheme can reduce the completion time and the total cost of the work while maximizing the utilization of resources. For the problem of virtual cell rescheduling considering sequence similarity under urgent order interference, this paper gives a higher production priority to urgent order by combining the characteristics of urgent order. When the emergency order is disturbed, the product family and the equipment group of the emergency order and the normal order of the original virtual unit are judged, and the virtual unit rescheduling can be carried out when the urgent order can merge with the original virtual unit. At the same time, the available machine sets and the corresponding processing time sets are set up for each workpiece process, and a multi-objective nonlinear integer programming model aiming at the completion time of urgent orders, the shortest total process time and the largest sequence similarity is constructed. According to the characteristics of the model, a combined genetic-ant colony optimization algorithm is designed to solve the model. In addition, by means of interference management, the influence of urgent order interference on production scheduling is reduced. With regard to the determination of batch strategy in virtual cell mode, this paper takes full account of the factors of time and cost, and puts forward the total completion time and total processing cost, which take into account the starting time of batch. A multiobjective nonlinear mixed integer programming model with minimum total transportation cost and total tardiness penalty cost. In order to solve the model effectively, an optimized genetic algorithm based on two-stage coding is proposed by combining genetic algorithm with process priority scheduling algorithm. The algorithm can divide the jobs into multiple sub-batches with flexible batches according to the actual scheduling requirements and optimize the process path and scheduling scheme of the sub-batches at the same time. Finally, based on the detailed analysis of the production status and existing problems of HDZH machining workshop, The virtual cell rescheduling strategy, which integrates the interference management method and batch partition idea, is applied to the production process of the rudder system and shafting system in the machining workshop. In order to verify the application value and significance of the proposed scheme in actual shipbuilding production. The results show that the proposed method is feasible and superior. At the same time, the research in this paper has certain guiding significance for the ship and other large manufacturing enterprises to optimize the production process, and provides a certain theoretical support for how to balance the production cost and the production disturbance when the urgent order is disturbed.
【学位授予单位】：江苏科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F426.4;F273

【参考文献】