【作者】 黄瑶；

【导师】 邓乾旺； KONG Zhengjun；

【作者基本信息】 湖南大学 ， 工业工程（专业学位）， 2020， 硕士

【摘要】 随着现代制造技术的飞速发展,产品更新迭代的频率越来越高,适用于多批次、小批量组装不同型号产品的混流装配线已广泛存在于汽车行业。在激烈的市场竞争环境下,如何保证生产线的高效实时零件供应成为企业发展的着力点。越来越多的汽车制造商采用物料超市作为中间仓储区域,临时储存来自中央仓库的物料,主要通过AGV、牵引车等工具为混流装配线提供灵活、可靠的物料供应。然而,线边库存容量有限,且须保证装配线不发生缺货,这对物料配送来说是个巨大挑战。基于物料超市的物料配送大多采用循环配送和单车点对点的配送模式,但是这两种配送模式在配送任务规模较大的情况下,运输效率较低且无法平衡运输车任务量,因此本文围绕汽车混流装配线的物料配送问题,着重于多车协同点对点物料配送方式的调度研究,具体研究内容如下:(1)以最小化装配线线边库存为目标,建立一种汽车混流装配线多车物料准时化配送的数学模型,综合考虑工位线边库存容量、运输车装载能力以及运输车数量的多重约束,基于预定的组装顺序,通过计算获得多车物料准时化配送方案,包括运输车序号、运输物料数量和每次运输的出发时间,在保证装配线不发生缺货的前提下实现多辆运输车资源的均衡调配,进而提升物料配送效率。(2)提出一种用于求解汽车混流装配线多车物料准时化配送模型的粒子群算法,并通过标准算例仿真分析的方式与基于经典回溯法的向后回溯法做对比,结果表明向后回溯法只能有效解决小规模实例的物料配送。相较之下,即使求解大型实例,粒子群算法也能在有限时间内快速收敛,得到使线边库存水平最小的物料配送方案,且各运输车的总工作时长和总运输量较为均衡,运输车在途率也达到了较高水平,实现对运输车资源的均衡调配,保证汽车混流装配线的物料准时化供应。(3)基于正交实验法探究多车物料配送方案的关键参数对实验结果影响。通过定性分析和正交实验验证,发现线边库存容量过小或运输车的数量过少可能导致无法找到可行的物料配送方案,除此之外,待装配产品数量和工位数量也对实验指标有一定影响。更多还原

【Abstract】 Along with the rapid development of modern manufacturing technology,the frequency of product renewal iteration is getting higher.Mixed-flow assembly lines suitable for assembling different types of products in multiple batches and small batches have been widely used in the automobile industry.In the fierce market competition environment,enterprises focus on how to ensure parts supply to production lines efficiently.More and more car manufacturers use material supermarkets as intermediate storage areas,which stores materials from central warehouses temporarily.What’s more,AGV,tractor and other tools are used to provide just-in-time parts supply for its mixed flow assembly line flexiblely and reliablely.However,it is a great challenge for material distribution that the inventory capacity is limited and the assembly line must not be out of stock.Material distribution based on material supermarkets mostly adopts milk-run distribution and single vechicle point-to-point distribution.However,these two distribution modes have low transportation efficiency and can not balance the task amount of the transport vehicle in the case of large distribution tasks.The material distribution problem of mixed flow assembly line focuses on the scheduling research of multi-vehicle coordinated point-to-point material distribution method.The specific research contents are as follows:(1)A mathematical model for the just-in-time delivery of multi-vehicle materials in the automobile mixed-flow assembly line is established to minimize the line-side inventory of each assembly line,taking into account the multiple constraints of the line-side inventory capacity of the station,the loading capacity of the transport vehicle,and the number of transport vehicles.Based on the predetermined assembly sequence,a multi-vehicle material just-in-time delivery plan is obtained through calculation,including the serial number of the transport vehicle,the number of transport materials,and the departure time of each transport.On the premise of ensuring that the assembly line will not be out of stock,the resources of multiple transport vehicles will be evenly allocated to improve the efficiency of material distribution.(2)Compared with the classical backtracking method,a particle swarm optimization(PSO)algorithm is proposed to solve the just-in-time distribution model of multi vehicle materials in the automobile mixed flow assembly line.The simulationanalysis through standard calculation examples shows that the backtracking method can only effectively solve small-scale examples Material distribution.In contrast,even in large instances,the particle swarm algorithm can quickly converge in a limited time to obtain a material distribution solution that minimizes the the inventory level along the line.Moreover,the total working time and total transportation volume of each transport vehicle are relatively balanced.And the in-transit rate of transport vehicles has also reached a high level,so as to realize the balanced allocation of transport vehicle resources and ensure the just-in-time supply of materials for the mixed flow assembly line.(3)Based on the orthogonal experimental method,the influence of the key parameters of multi vehicle material distribution scheme on the experimental results is explored.Through qualitative analysis and orthogonal experimental verification,it is found that too small a line-side inventory capacity or too few transport vehicles may lead to the inability to find a feasible material distribution scheme.In addition,the number of products to be assembled and the number of work stations also have certain effects on the experimental indexes.更多还原