Click to edit Master title style,Click to edit Master text styles,Second level,Third level,Fourth level,Fifth level,*,ACTION WORKOUT,g,Heijunka Cookbook,计划生产的精益方法论,计划生产和精益方法论-中文版课件,1,1.,选择生产线,Analyze Historical Sales Data,Determine Build Strategy,Calculate Takt Time,Level Load,Sequence,Understand what is made,明白要做什么,型号,3000,4000,16,种硬件配置,79,个零件编号,23,个供应商,每年14,000台,DASH,2493,种定单组合,1.选择生产线Analyze Historical Sal,2,2.,识别特殊型号或配置,产品有多少种不同的型号?每种型号不同在哪里?,举例:,Dash 3000/4000,病人监护仪,16,种不同的硬件配置:,3000 or 4000,有没有复写器,有没有,RF,Nellcor or Masimo,Analyze Historical Sales Data,首先搞清楚产品,Determine Build Strategy,Calculate Takt Time,Level Load,Sequence,Understand what is made,2.识别特殊型号或配置 产品有多少种不同的型号?每种型号不,3,3.,做一个产品系谱,Analyze Historical Sales Data,Determine Build Strategy,Calculate Takt Time,Level Load,Sequence,哪些零件是,每个,型号都要用上的?,哪些零件是用在,部分,型号上的?,Understand what is made,首先搞清楚产品,产品系谱不是简单的,BOM.,它在了解一种零件的组装频率上起了很大的作用,3.做一个产品系谱Analyze Historical S,4,4.,要明白哪些零件装到产品里,每个组件都包括哪些零件(规格和数量),Understand what is made,Analyze Historical Sales Data,Determine Build Strategy,Calculate Takt Time,Level Load,Sequence,首先搞清楚产品,4.要明白哪些零件装到产品里每个组件都包括哪些零件(规格和,5,5.,收集以往的销售数据,不能预测未来历史是最好的销售预言家,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,Analyze historical sales data,5.收集以往的销售数据不能预测未来历史是最好的销售预言家,6,6.,按照不同的部件分解销售 数据,Analyze historical sales data,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,6.按照不同的部件分解销售 数据Anal,7,7.,总量数据,Analyze historical sales data,计算出每个季度的总量,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,7.总量数据Analyze historical sale,8,8.,计算出组件的百分比,用总量数据来计算出每个季度已销售组件的百分比,Analyze historical sales data,举例:在2002年第四季度卖出的所有仪表板里66%的产品有,Nellcor DAS,组件.,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,8.计算出组件的百分比用总量数据来计算出每个季度已销售组件,9,9.,选择数据预测未来的销售情况,如何选择:,市场的变化?,产品的稳定性?,Stage in product life?,Analyze historical sales data,给市场变化下定义2002年第一季度的数据不再具有代表性.,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,9.选择数据预测未来的销售情况如何选择:Analyze h,10,10.Weight PSI 25%,History 75%,Analyze historical sales data,OPSI Percentages(Weighting=25%),计划百分率,Historical Percentages(Weighting=75%),Dealing with mix.,PSI for volume,Heijunka for mix,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,10.Weight PSI 25%,History 75,11,11.Multiply Subassembly Percentages,Probability of customer ordering:,3000,Nellcor,with Writer,with RF,(Part No.=420000-007),=,(.47)(.73)(.41)(.15)=2.1%,Analyze historical sales data,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,11.Multiply Subassembly Perce,12,12.Create P-Q Chart,Analyze historical sales data,2015999-011,占了,12%,的销售量,%,Understand what is made,分析以往的销售数据,Determine Build Strategy,Level Load,Calculate Takt Time,Sequence,过去4个季度的销售情况,12.Create P-Q ChartAnalyze hi,13,13.,决定生产策略,考虑因素:,检测时间,质量问题,供应商周期,80/20 rule,决定生产策略,Level Load,Calculate Takt Time,Sequence,Analyze Historical Sales Data,预测生产,Understand what is made,Determine build strategy,定单生产,%,13.决定生产策略考虑因素:决定生产策略Level Loa,14,14.,计算工作日,Determine Build Strategy,均衡生产,Calculate Takt Time,Sequence,Analyze Historical Sales Data,Understand what is made,Level load,感恩节,圣诞节,(,第四季度68个工作日)(4 个无计划生产日)=64 有效工作日,14.计算工作日Determine Build Strat,15,15.,计算每天/每周的产量,Determine Build Strategy,均衡生产,Calculate Takt Time,Sequence,Analyze Historical Sales Data,Understand what is made,Level load,每季度需求或是预测,每季度工作日,举例:,4288,件/季度,64,天/季度,=,67,件/天,335,件/周,多少?,每天产量,=,=,15.计算每天/每周的产量Determine Build,16,16.,计算出每周产品组合,335*20.2%=,68,件/周,Level load,什么型号?,Determine Build Strategy,均衡生产,Calculate Takt Time,Sequence,Analyze Historical Sales Data,Understand what is made,16.计算出每周产品组合335*20.2%=,17,17.,决定每周产量,Level load,Determine Build Strategy,均衡生产,Calculate Takt Time,Sequence,Analyze Historical Sales Data,Understand what is made,什么型号?,17.决定每周产量Level loadDetermine B,18,18.a,计算出生产线的节拍 时间,节拍时间是企业的心跳,有效时间,需求量,节拍时间,=,节拍时间是一个数学问题,但它是精益生产的基本原则.,有效时间:,安排用来生产的所有生产时间,需求量:,在一段时间内预定零件的生产数量,每个轮班的有效时间是6.6个小时(连续性的),这6.6个小时是除去午饭,休息等的连续工作时间.,举例,工厂每天两班制,顾客的需求是每个星期1000件产品,3960,分钟/周,1000,件/周,=3.96,分钟/件,节拍时间=,Calculate Takt time,计算节拍时间,Determine Build Strategy,Level Load,Sequence,Analyze Historical Sales Data,Understand what is made,18.a 计算出生产线的节拍 时间,19,仪表板生产线每天一班制,6.6,小时/天,*60,分钟/小时,*64,天/季度,=25,344,分钟/季度,顾客的需求是每个季度4288台,有效时间,需求量,节拍时间,=,4288,台/季度,25,344,分/季度,=,6,分/台,Dash Line Takt Time,=,Calculate Takt time,计算节拍时间,Determine Build Strategy,Level Load,Sequence,Analyze Historical Sales Data,Understand what is made,(,每个车间都会不同),18.b,计算出生产线的节拍 时间,仪表板生产线每天一班制有效时间 节拍时间=4288,20,我们已经计算出了生产线的节拍时间,现在我们要为我们的生产排序,概要:,工作天数,生产线的节拍时间=6 分钟,12,个不同的型号,Sequencing Production,Level Load,Determine Build Strategy,Calculate Takt Time,排序,Analyze Historical Sales Data,Understand what is made,排序,生产线的节拍时间为6分钟,接着,我们计算出每个型号的节拍时间,我们已经计算出了生产线的节拍时间,现在我们要为我们的生产