多品种混合装配车间关键资源建模与优化

发布时间：2018-04-05 09:30

本文选题：仿真模型　切入点：AGV数量配置　出处：《上海交通大学》2011年硕士论文

【摘要】：企业面临着交货期提前、客户需求多样化和产品生命周期缩短的压力，因此缩短交货期、提高资源利用率、降低生产成本、提高生产运作的灵活性，，已成为保证企业市场竞争力的重要手段。本文以日本大金现有装配车间为实际应用背景，对该装配车间存在的问题进行有针对性的研究。通过解决这些问题来提高生产线效率，缩短生产周期，提高市场反应能力。具体来说，论文主要内容如下： (1)多品种混合装配线仿真建模。针对该企业的空调装配线是多品种混合装配、车间调度逻辑复杂等特性，建立了面向对象的空调装配车间仿真模型。利用该仿真模型可以对生产计划、车间布局方案等进行验证，也为后续研究的优化方案验证提供一个平台。 (2)配料区布料及并行机器调度优化。由于装配线节拍很快，对配料区的物料分拣速度有满足节拍要求，提出以工人操作距离最短为目标，对配料区的物料分布进行优化。对多并行机器组成的CellStation单元，提出了4类规则来对Cell Station进行优化调度，比较不同规则下的完工时间。 (3)多种类型AGV配置数量优化。以生产线产能最大化为目标，以生产工艺、AGV路径、调度规则等为约束条件，建立了以AGV的数量为决策变量的数学模型来描述AGV的数量配置问题。针对该问题的复杂性，提出了基于仿真的优化方法，其利用仿真模型模拟实际车间的AGV运行状态，并通过遗传算法优化AGV的配置数量，从而有效地提高了优化结果的精度与优化的速度。 (4)AGV环线路径优化。提出利用区域环思路设计无交叉、无碰撞堵塞的AGV路径。由于区域环AGV路径存在环间流以及环内流平衡问题，以环间流最小为目标，以环与环之间环内流平衡等为约束，建立了以工位划分为决策变量的数学模型。针对该问题是NP难问题，设计模拟退火算法来求解该模型。 (5)应用系统开发。通过对日本大金现有多品种混合装配车间进行仿真建模，基于QUEST二次开发，将仿真与优化方法相结合进行应用，应用结果表明研究优化方法的有效性。
[Abstract]:Enterprises are faced with the pressure of early delivery, diversification of customer demand and shortening of product life cycle. Therefore, it is necessary to shorten delivery time, improve resource utilization, reduce production cost, and improve flexibility of production operation.Has become an important means to ensure the competitiveness of enterprises in the market.In this paper, the existing assembly workshop in Japan is used as the practical application background, and the existing problems of the assembly workshop are studied.Solve these problems to improve production line efficiency, shorten production cycle, improve market responsiveness.Specifically, the main contents of the paper are as follows:Simulation modeling of multi-variety hybrid assembly line.An object-oriented simulation model of air conditioning assembly shop is established in view of the characteristics that the air conditioning assembly line of the enterprise is multi-variety hybrid assembly and the workshop scheduling logic is complex.The simulation model can be used to verify the production plan, workshop layout and so on. It also provides a platform for the optimization scheme verification of the following research.Distribution and parallel machine scheduling optimization in batching area.Because the assembly line beat is very fast and the speed of material sorting in the batching area can meet the requirements of the beat, it is proposed to optimize the distribution of materials in the batching area by taking the shortest operating distance of the workers as the goal.For the CellStation unit composed of multi-parallel machines, four kinds of rules are proposed to optimize the scheduling of Cell Station, and the completion time under different rules is compared.Optimization of the number of multiple types of AGV configurations.Taking the production line capacity maximization as the goal, taking the production process path and scheduling rules as the constraint conditions, a mathematical model with the quantity of AGV as the decision variable is established to describe the quantity allocation of AGV.Aiming at the complexity of the problem, an optimization method based on simulation is proposed. The simulation model is used to simulate the running state of AGV in the actual workshop, and the number of AGV configurations is optimized by genetic algorithm.Therefore, the precision and speed of optimization are improved effectively.Path optimization of AGV loop.This paper presents the design of AGV path without cross and collision by using the idea of region loop.Due to the problem of inter-annular flow and flow equilibrium in the AGV path of a region ring, a mathematical model is established in which the work station is divided as a decision variable, taking the minimum flow between the rings as the objective and the flow balance between the rings as the constraint.In view of the NP-hard problem, simulated annealing algorithm is designed to solve the model.Application system development.Based on the simulation modeling of the existing multi-variety hybrid assembly workshop in Dagin Japan and based on the secondary development of QUEST, the simulation and optimization methods are combined for application. The application results show the effectiveness of the research on the optimization method.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH186

【参考文献】