【作者】 胡锦峰；

【导师】 苏海涛；

【作者基本信息】 南昌大学 ， 工业工程， 2012， 硕士

【摘要】 随着计算机技术的发展,建模与仿真越来越多的被应用到优化领域之中。Flexsim建模与仿真也已成为解决由高度离散的事件组成的系统的优化问题的重要工具。本文以W公司涡旋空压机生产系统为研究对象,对其加工工艺流程进行了分析；同时,应用新的数据拟合工具ExpertFit对现场搜集的数据进行了分析拟合,得到了Flexsim代码；然后,采用Petri网对系统的各加工单元进行了建模,经初步分析,了解了系统之中的顺序与冲突事件；最后,在Petri网模型和ExpertFit拟合结果的基础上,建立了Flexsim二维与三维模型,通过仿真与结果分析,发现了w公司生产系统中存在的不足与瓶颈：操作员平均工时空闲率很高、三坐标检测处存在严重的瓶颈。针对分析得到的生产系统的不足与瓶颈,本文通过修改原有模型,建立了Flexsim实验器,并分别以操作员数目和三坐标台数为控制变量,进行了仿真实验；同时,辅以恰当的绩效指标对实验的结果进行了评价：最后,得到了最优化的资源配置方案,即操作员数目减少为3个,设置4台三坐标。通过改进,在W公司涡旋空压机的产量基本保持不变的情况下,操作员的平均工时空闲率由71.6%降至45.1%,三坐标检测处暂存区中临时实体的平均等待时间由原来的5459s降至723s；在消除瓶颈的同时,也大幅度地节省了公司的直接人工成本。本文采用Petri网建模与Flexsim仿真相结合的方法对离散系统进行建模分析与改进,取得了良好的效果,为企业合理配置资源和类似的研究提供了一种思路。更多还原

【Abstract】 With the development of computer technology, modeling and simulation have been applied into the field of the optimization more and more widely. Modeling and simulation based on Flexsim has been an important tool in solving optimization problem of the systems that made up of highly discrete events.This paper took scroll air compressor machine production system of W company as research object, and analyzed its processing process. Meanwhile, a new data fitting tools which named ExpertFit was used to analyze and fit the data which collected on the spot, and then Flexsim code was acquired. Then this paper used Petri network to model all processing units of the system. Via preliminary analysis, the order and conflict events were grasped. At last, two-dimensional and three-dimensional models of Flexsim were established based on the Petri network model and ExpertFit’s fitting results. After simulation and analysis, this paper found shortcomings and bottlenecks of W company in production systems. One is that the average idle rate of operators is very high, and another is that the Coordinate Measuring Machine (CMM) has a serious bottleneck.With the inadequate and bottlenecks taken in consideration, this paper modified the original model, established the Flexsim experimental tool, and carried out simulation experiments, operator number and CMM number taken as control variables. Meanwhile, evaluation was made base on the experiment after appropriate performance indexes were done. At last, the most optimization of resources configuration scenarios were gained, namely, set three operators and four CMMs.Via improving, on the circumstance of almost the same output of W scroll air compressor, the average idle rate of operators was cut down from71.6%to45.1%, and the average waiting time of temporary entities in the CMM queue was reduced from5459seconds to723seconds. The company’s direct labor cost was substantially saved with the elimination of bottlenecks at the same time.The method that combines Petri net modeling with the Flexsim simulation for modeling and analysis in this paper achieved good results, and provided a good idea for the company’s rational allocation of resources and some similar research.更多还原