【作者】 刘文平；

【导师】 李兆前；

【作者基本信息】 山东大学 ， 机械制造及其自动化， 2009， 博士

【摘要】 多品种混合装配是在基本不改变或较少改变现有生产设施的前提下,通过对装配生产线的合理组织与排产优化,实现多品种共线装配,以最大限度地挖掘生产线的潜能,用大批量生产的制造成本和响应速度,向客户提供定制的个性化产品和服务。汽车产品需求多样化和汽车市场细分化,促使越来越多的汽车制造商将品种混合装配作为增强其竞争能力的有效手段。论文以日照某企业集团的汽车生产为背景,剖析了汽车装配生产过程中亟待解决的技术问题,利用相关优化理论和邻域搜索算法,研究了混合品种汽车装配线的平衡与排序问题。该研究对类似汽车生产企业的混合装配生产具有较强的参考价值。平衡与排序是对混合装配线的高效运行产生重要影响的定量决策问题。随着问题规模的增大,其求解的难度呈现阶乘级的增长。基于平衡与排序问题的求解复杂性,本文中采用邻域搜索启发式算法进行研究。研究了混合品种装配线的平衡问题。提出了装配线在规划设计和投产运行两个阶段的优化目标,建立了装配线生产率最大化和平滑指数最小化的平衡优化模型。提出了装配优先约束关系模型和约束关系模型的优先矩阵表达法与数据文件表达法。在Memetic算法框架下,设计了装配线平衡模型的求解算法。通过实例,采用Lingo11.0优化软件包,验证了平衡模型与Memetic求解算法的有效性。研究了混合品种装配线的排序问题。探讨了混合品种的产品投放顺序对装配线的物料流动、装配线总调整时间、工作站负荷和最小生产循环周期等的影响。建立了物料消耗速率波动最小化模型、品种切换时装配线总调整时间最小化模型、因工作站超载所需的线外总替补任务量最小化模型和最终装配总费用最小化模型。通过8组算例验证了Memetic算法对于求解混合品种装配线投产排序多目标优化问题的有效性。研究了多约束条件下混合品种汽车装配线的综合优化问题。结合汽车制造企业的生产实际,将汽车装配生产中的优化问题从总装配车间向前延伸至涂装车间,分析了涂装车间车身喷涂颜色变换和总装配车间负荷与物流均衡化的需求。根据对优化问题影响程度的不同,将多约束条件分为软约束类和硬约束类。它对优化问题的描述更加符合生产实际,且可以解决传统的约束满足问题在优化过程中难以求得可行解的问题。根据汽车选装件的装配作业时间和工作站的装配能力,提出了比例约束控制的产品投放规则。比例约束是对各装配工作站上负荷与物流的一种限定,它通过更加简单的方式来控制装配线上的装配负荷与物流瓶颈。构建了总附加成本最小化的整数线性规划模型,并采用改进的邻域搜索算法对优化模型进行求解。本文研究中采用的各单元优化技术和系统优化技术,对于一般装配生产线平衡与排序问题的优化具有理论指导意义,对于类似混合品种汽车装配生产线的合理组织与调度具有实际参考价值。更多还原

【Abstract】 Without any substantial modification to existing production facilities, mixed-model assembly lines(MMALs) are capable of producing a variety of product models similar in construction and configuration concurrently,provided that MMALs are well balanced and sequenced.MMALs yield customised products and services with the manufacturing cost and response speed which are usually required under mass production.With the diversification and subdivision of the automobile market demand,automakers choose to use MMALs to sharpen their competitive edge.Focusing on the manufacturing in an automaker located in Rizhao,Shandong,the thesis examines the key technological issues that need urgent treatment.Further,the balancing and sequencing problems in MMALs are studied using applicable optimisation algorithms of VNS.The study may benefit automakers with similar MMALs and similar production context.Balancing and sequencing involve in the quantitative decision-making process aiming at improving the performance of MMALs.As the problem size grows,the computational complexity grows on a factorial basis.Thus, VNS-based heuristics were used to solve the following balancing and sequencing problems in MMALs.Balancing of MMALs.Optimisation models maximising production rate and smoothing index are formulated for the designing and operating stage of MMALs, respectively.The PG(precedence graph) model for describing assembly precedence relations is proposed,and precedence matrix and data file are used to implement PG models.Based on the theoretical framework,a Memetic algorithm is devised for solving the proposed balancing model.To verify the balancing model and Memetic algorithm,a case study is given together with a comparative test with Lingo 11.0 package. Sequencing of MMALs.Based on the analysis of the consequence analysis of launching permutation on logistics,setup time,workload and minimal production cycle of MMALs,optimal models minimising material consumption variation, setup time incurred by adjustment for model changing,utility work necessitated by overload of workstations,and total assembly cost are formulated.Finally,8 problems are tested to verify the proposed Memetic algorithm used to solve the multi-criteria model.Systematical optimisation of mixed-model automobile assembly lines.The scope of optimisation is expanded forward from the assembly shop to the paint shop,and the practical constraints including color changes and workload/material balancing in both shops are analysed.According to the impact on optimal solutions,the constraints are classified into soft constraints and hard constraints.The classification enables the model formulation to be more congruent with the practical manufacturing environment.It also facilitates the searching process for feasible solutions that is basically hard with the traditional constraints satisfaction method.To compare the assembly time of optional parts and the assembly capacity of workstations,the concept of ration constraint, which is actually the launching rule for controlling workload bottlenecks,is proposed.At the end of this part,an integer linear programming model with the objective of minimising the total assembly cost is formulated and solved with the improved neighborhood search algorithm.To sum up,the methodology adopted in each optimisation problem is meaningful to the balancing and/or sequencing optimisation of general assembly lines,and is of guiding significance for the optimal configuration and scheduling of similar mixed-model automobile assembly lines.更多还原