多级机加装配混合制造系统中的跨车间零部件调度策略研究

发布时间：2018-05-22 13:28

本文选题：跨车间生产 + 机加装配混合制造系统　；参考：《东北大学》2012年硕士论文

【摘要】：随着生产水平的提高,以及需求的多元化,生产模式由单工序、大批量集中生产逐渐转为多工序、多品种、变批量生产。由于单个车间的设备局限性,企业内部不同车间的生产协作现象越来越频繁,车间之间的协作计划与调度愈加困难,这就对原有的生产调度方法提出了新的挑战。本文以某机械制造企业车间运作管理实践需求为背景,作为国家自然科学基金课题的重要组成部分,在对车间调度研究现状进行分析与综述的基础上,特别针对一类具有多级机加和装配关系的混合型制造系统的跨车间调度问题开展建模和优化算法的研究,主要包括： (1)面向单级机加和装配组成关系的零部件最优调度策略研究。针对独立生产车间只有一台通用(万能)机加设备和一台装配设备场景,提出两种基本结构单元及其描述方法,在此基础上提出基于理想生产单元的理论最小完成时间计算方法；提出面向产品结构图的数学描述,并提出基于支路优先级的调度策略；分别针对单机和并行多机情形下,一台装配设备的两条支路产品结构,比较分析了三种不同的调度策略。 (2)面向多级机加和装配组成关系的零部件调度策略。分别针对零件加工顺序相同和不同情形,提出面向基本结构单元的最小影响值的调度策略；考虑多级装配,产品结构具有两条和三条支路,每种机加工设备单台或多台,装配设备两台和三台的不同场景,通过引进同步装配的思想、最大(最小)时间差值和实际时间差值的概念,分别提出了基于同步装配思想的两种不同调度策略。 (3)双跨车间零部件调度问题的两种启发式算法。针对协作(跨)车间生产情境,通过引进逻辑上的联合车间调度(简称联合调度)、参考加工起始时间、参考加工结束时间、加工延迟费用、满意度增值效益(提前)等概念,从车间之间协调的视角,提出了以最小化总生产成本为目标的双跨车间调度问题的数学规划模型,以平衡系统的总生产时间、各车间的延迟时间和提前时间(车间满意度)。针对问题的特点,提出了基于联合调度思想、嵌入遗传搜索的启发式调度方法和基于系统协调思想最小成本调度启发式算法。这两个算法将原问题分解为三个阶段；第一、二阶段采用基于两层结构图的遗传搜索算子求解零件的分批与指派问题；第三阶段采用启发式调度策略,完成最后的并行机调度。通过对不同测试算例、不同算法参数分析了两种调度启发式的性能。
[Abstract]:With the improvement of production level and the diversification of demand, the mode of production has gradually changed from single process, mass centralized production to multi-process, multi-variety and mass production. Because of the limitation of the equipment in a single workshop, the phenomenon of production cooperation among different workshops is becoming more and more frequent, and the planning and scheduling of cooperation between workshops becomes more and more difficult, which poses a new challenge to the original production scheduling methods. Based on the practical requirement of workshop operation management in a mechanical manufacturing enterprise, as an important part of the project of National Natural Science Foundation, this paper analyzes and summarizes the present situation of workshop scheduling research. In particular, the modeling and optimization algorithms for a class of hybrid manufacturing systems with multilevel machine addition and assembly relationship are studied, including: 1) Research on the optimal scheduling strategy of components for single stage machine adding and assembling. In view of the scene where there is only one universal (universal) machine and one assembly equipment in the independent production workshop, two basic structural units and their description methods are presented. On this basis, a theoretical minimum completion time calculation method based on ideal production unit is proposed, a mathematical description oriented to product structure diagram is proposed, and a scheduling strategy based on branch priority is proposed. The product structure of two branches of one assembly equipment is compared and three different scheduling strategies are analyzed. Second, the component scheduling strategy for multistage machine loading and assembly. According to the same and different machining sequence of parts, a scheduling strategy for the minimum influence value of basic structural unit is proposed. Considering multi-stage assembly, the product structure has two or three branches, and each machining equipment has one or more units. By introducing the concept of synchronous assembly, maximum (minimum) time difference and actual time difference, two different scheduling strategies based on synchronous assembly are proposed. 3) two heuristic algorithms for two-span job shop parts scheduling problem. According to the production situation of collaborative (cross) workshop, this paper introduces the concepts of logical joint workshop scheduling, referring to processing start time, processing end time, processing delay cost, satisfaction increment benefit (advance), and so on. From the perspective of coordination between workshops, a mathematical programming model for two-span job-shop scheduling problem with the goal of minimizing total production cost is proposed to balance the total production time of the system, the delay time of each workshop and the advance time (job shop satisfaction). According to the characteristics of the problem, a heuristic scheduling method based on the idea of joint scheduling and embedded genetic search and a heuristic algorithm of minimum cost scheduling based on the idea of system coordination are proposed. The two algorithms decompose the original problem into three stages: firstly, the genetic search operator based on two-layer structure graph is used to solve the part batch and assignment problem; the third stage uses heuristic scheduling strategy to complete the final parallel machine scheduling. The performance of two scheduling heuristics is analyzed by different test examples and different algorithm parameters.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH186

【共引文献】