中級(jí)統(tǒng)計(jì)過程控制教程(spc)-文庫吧在線文庫
【正文】 下一步驟是篩選 (screening)數(shù)據(jù) .這個(gè)步驟是利用簡單實(shí)驗(yàn)前統(tǒng)計(jì)工具 (statistical preexperimental tools)去消除過程中所有無效用的變化源 .這是實(shí)驗(yàn)前階段 (preexperimental stage),在此階段可以利用一些工具如柏拉圖 (pareto),因果分析圖 (cause and effect),變量控制圖 (Multivari charts),這樣當(dāng)實(shí)驗(yàn)直接涉及變異目標(biāo)時(shí)你能更準(zhǔn)確地把目標(biāo) . 假如你的過程中變量很多 ,才有必要用較復(fù)雜的 統(tǒng)計(jì)方法 ,如部分因子實(shí)驗(yàn)法去處理 .什么是解決問題策略的步驟 ？ (續(xù) )SPCadvanced training course步驟三 : 現(xiàn)在可以進(jìn)行實(shí)驗(yàn) ,目標(biāo)在于預(yù)測某種變動(dòng)對(duì)現(xiàn)行過程的作用 .這樣才能把握地預(yù)測出輸入變量 ”X”對(duì)輸出 ”Y”的影響 ,或至少能使建議的過程 (B)比現(xiàn)行過程(C)產(chǎn)生更良好的結(jié)果 .步驟四 : 繼續(xù)做實(shí)驗(yàn) .你的過程有兩個(gè)或四個(gè)變量 .你的策略是按因素設(shè)計(jì)用 t檢驗(yàn)及 f檢驗(yàn)去比較是否呈正態(tài)分布 .步驟五 : 繼續(xù)做實(shí)驗(yàn) .你的策略是有因素設(shè)計(jì)進(jìn)行方差分析 .當(dāng)你有五個(gè)或五個(gè)以上變量時(shí) ,要利用部分因子實(shí)驗(yàn)法 .SPCadvanced training course步驟六 : 你已繪上 X及 Y軸 ,進(jìn)行過程控制 .利用控制圖表去維持過程的控制 .步驟七 : 可以能過重新確定所在和重新實(shí)驗(yàn)發(fā)現(xiàn)新的解決方法來改進(jìn)你的控制方式 ,這是一項(xiàng)不重復(fù)的過程 .步驟八 : 99% 的過程變異已受你的控制 ,可以運(yùn)用不同方法去控制最后的1% 以完成任務(wù) ,這些方法包括 :響應(yīng)曲面法 (Response Surface Methodology),正交旋轉(zhuǎn)應(yīng)曲面法 (Othogonal Rotatable RSM),單純形法 (Simplex),最速上升法 (SteepeAscent)以及回歸方法 (Regression).用以上方法我們可以對(duì)一些物理模式中的參數(shù)進(jìn)行估計(jì) .SPCadvanced training courseProblem solving step by step approach:SPCadvanced training courseContinuous Improvement RoadmapProcess Characterization FlowchartPrioritize Opportunities for improvementSelect teamProcess MappingDefine the critical characteristicPerform Gauge RRCapable? ImprovementmeasurementPerform a process capability studyStable capableControlOptimizationImplement total process controlInstitutionalizationWhat? Why?To work on critical business issuesTo select a cross function teamTo map the process(as is should beTo identify define major product,process and machine parametersTo quantify the measurement errorTo assess process stability estimate process mean variationTo pare process performanceto customer requirementsTo identify cause effect relationshipsTo implement preventive procedures steps to assure a stable high performance processTo implement best practices standardsTo leverage resourcesTo expand knowledge creativityTo get a visual picture of the processTo use 80/20 principle(80% of the performance will becontrolled by 20% of the parameters)To verify the measurement variation is less than 10% of the process variation or spec.To verify the process is stableTo estimate the process mean/variationTo pare the process performance to spec.To find important input variablesTo assure proactive preventive thinking is implementedTo assure est process monitoringTo lay out C/A procedure up frontTo implement what was learned across BUEd Bedell’s CIRSPCadvanced training course統(tǒng)計(jì)知識(shí)的層次(針對(duì)一制程 )僅有經(jīng)驗(yàn) 沒有數(shù) 據(jù)數(shù) 據(jù) 采集 看看數(shù) 據(jù)整合數(shù) 據(jù) 圖 , 表描述統(tǒng)計(jì) 平均值 , 方差特征統(tǒng)計(jì) Cp/Cpk, 曲 /直線擬合控制統(tǒng)計(jì) SPC控制圖預(yù)示統(tǒng)計(jì) 變量分析 ,實(shí)驗(yàn)設(shè)計(jì) 線性或多元回歸SPCadvanced training cou
環(huán)評(píng)公示相關(guān)推薦
鄂ICP備17016276號(hào)-1