SPCIntroductionI基礎(chǔ)知識(shí)資料

1、SPCIntroductionI基礎(chǔ)知識(shí)資料SPC-Introduction-I資料六點(diǎn)原則六點(diǎn)原則知識(shí)要轉(zhuǎn)換成行動(dòng)，只收集資料及繪圖是知識(shí)要轉(zhuǎn)換成行動(dòng)，只收集資料及繪圖是不夠的。不夠的。MSMS必須確認(rèn)是受控、穩(wěn)定及不是重要必須確認(rèn)是受控、穩(wěn)定及不是重要變異來源。變異來源。變異的研究及統(tǒng)計(jì)訊號(hào)及工具的應(yīng)用范變異的研究及統(tǒng)計(jì)訊號(hào)及工具的應(yīng)用范圍，不限於工廠生產(chǎn)。圍，不限於工廠生產(chǎn)。SPCSPC每個(gè)字都是重點(diǎn)。不用統(tǒng)計(jì)沒每個(gè)字都是重點(diǎn)。不用統(tǒng)計(jì)沒S S。只。只對(duì)產(chǎn)品，不對(duì)過程沒對(duì)產(chǎn)品，不對(duì)過程沒P P。都不是。都不是SPCSPC。六點(diǎn)原則六點(diǎn)原則實(shí)做，才可以真知。每個(gè)人都必須實(shí)際實(shí)做，才可以真知

2、。每個(gè)人都必須實(shí)際進(jìn)行個(gè)案。進(jìn)行個(gè)案。持續(xù)深入研究，以滿足客戶需求。尤其是持續(xù)深入研究，以滿足客戶需求。尤其是統(tǒng)計(jì)學(xué)。統(tǒng)計(jì)學(xué)。個(gè)案設(shè)計(jì)個(gè)案設(shè)計(jì) RP01RP01 PPAPPPAP樣品分組習(xí)題樣品分組習(xí)題The need for process control Detection-tolerates waste偵測(cè) Prevention-avoids waste預(yù)防Process control system model with feedback FigureFigure FourFour elementselements ofof thethe system(system(四要素）四要素）

3、TheThe processprocess（過程）（過程） InformationInformation aboutabout performanceperformance（過程績效資訊）（過程績效資訊） 過程本身、過程的內(nèi)在變異性過程本身、過程的內(nèi)在變異性 選定過程特性、設(shè)定目標(biāo)值、監(jiān)控實(shí)際值、差異比較、是否需要行動(dòng)選定過程特性、設(shè)定目標(biāo)值、監(jiān)控實(shí)際值、差異比較、是否需要行動(dòng) ActionAction onon thethe processprocess（針對(duì)過程的行動(dòng)）（針對(duì)過程的行動(dòng)） ActionAction onon thethe outputoutput（針對(duì)產(chǎn)出的行動(dòng)）（針對(duì)產(chǎn)出

4、的行動(dòng)） 隔離、分檢、重工、修理、報(bào)廢隔離、分檢、重工、修理、報(bào)廢 過程不穩(wěn)定、不夠能力時(shí)，不得已還是要做。過程不穩(wěn)定、不夠能力時(shí)，不得已還是要做。制程控制系統(tǒng)制程控制系統(tǒng) 有反饋的過程控制系統(tǒng)模型有反饋的過程控制系統(tǒng)模型 過程的呼聲 人 設(shè)備 材料 方法 產(chǎn)品或 環(huán)境 服務(wù) 輸入 過程/系統(tǒng) 輸出 顧客的呼聲我們工作的方式/資源的融合統(tǒng)計(jì)方法顧客識(shí)別不斷變化的需求和期望抽樣推論樣本樣本統(tǒng)計(jì)量母體參數(shù)母體單峰對(duì)稱平均數(shù)的意義 所有觀察值的總合，除以所有觀察值的個(gè)數(shù)。平均數(shù)在數(shù)線上代表資料的平衡點(diǎn)。平均數(shù)求下列數(shù)據(jù)的平均數(shù)平均數(shù)(62.53+62.47+62.68+62.55+62.38+62.

5、79+62.41)中位數(shù)將觀察值由大而小排列，位于中央的數(shù)值即為中位數(shù)，若觀察值為偶數(shù)，中位數(shù)為中央兩個(gè)數(shù)值的平均。求下列數(shù)據(jù)的中位數(shù)中位數(shù)為眾數(shù)量測(cè)值中出現(xiàn)次數(shù)最多的數(shù)值。求下列數(shù)值的眾數(shù)眾數(shù)為極差將量測(cè)值中的最大值減去量測(cè)值中的最小值，所得到的數(shù)值即為極差。極差 樣本標(biāo)準(zhǔn)偏差公式為距離重心所在位置的距離，平方之後的平均數(shù)樣本與母體公式不一樣求下列數(shù)值的樣本標(biāo)準(zhǔn)偏差平均數(shù)樣本標(biāo)準(zhǔn)偏差求下列數(shù)據(jù)的平均數(shù)、中位數(shù)、眾數(shù)、極差與標(biāo)準(zhǔn)偏差。平均數(shù)中位數(shù) 首先將數(shù)據(jù)排序中位數(shù) 眾數(shù)極差標(biāo)準(zhǔn)偏差DistributionDistribution分配分配 Pieces vary from each othe

6、r. But they form a pattern that, if stable, can be described as a distribution.有大有小才有分配 Distribution can differ in location, spread and shape.位置、分布寬度、形狀 Figure upper side 每件產(chǎn)品的尺寸與別的都不同 范圍 范圍 范圍 范圍但它們形成一個(gè)模型，若穩(wěn)定，可以描述為一個(gè)分布 范圍 范圍 范圍分布可以通過以下因素來加以區(qū)分 位置 分布寬度 形狀 或這些因素的組合態(tài)分配曲線以鐘形呈現(xiàn)，在正中央部份為其頂峰，該點(diǎn)是機(jī)率分配的平均數(shù)，中位數(shù)

7、及眾數(shù)所在。態(tài)分配以其平均數(shù)為中心，左右對(duì)稱且向左右延伸，漸趨向X軸靠近。首先將區(qū)間上下限予以標(biāo)準(zhǔn)化，設(shè)若要求(a , b )區(qū)間內(nèi)的機(jī)率，則先將數(shù)值a與b予以標(biāo)準(zhǔn)化。所謂標(biāo)準(zhǔn)化，即是將欲求的數(shù)值換算為與平均值的距離為幾個(gè)標(biāo)準(zhǔn)偏差已知平均值為，標(biāo)準(zhǔn)偏差為，請(qǐng)將區(qū)間( 1112 )化成正態(tài)分配的標(biāo)準(zhǔn)化數(shù)值表示。首先計(jì)算226克于正態(tài)分配中的位置成品總重小于226公克的比例為95.25%計(jì)算225克于正態(tài)分配中的位置成品總重介于225公克與226公克之間的比例為34.61%首先計(jì)算于正態(tài)分配中的位置成品總長度小于的比例為11.31%計(jì)算于正態(tài)分配中的位置總重介于與之間的比例為85.25%首先計(jì)算規(guī)

8、格上下限于正態(tài)分配中的位置上限下限不良率為1.34%Ppk=1 Ppk=2 中心值距離上下規(guī)格線，1*3 標(biāo)準(zhǔn)偏差時(shí)Ppk=1 中心值距離上下規(guī)格線，2*3 標(biāo)準(zhǔn)偏差時(shí)Ppk=2 中心值偏向哪一邊？距離多少？Ppk=? 中心值距離上下規(guī)格線，各多少？其中小值是多少？Ppk=? Common causes refer to the many causes of variation that consistently acting on the process. common causes within a process produce a stable(穩(wěn)定） and repeatable（可

9、重復(fù)） distribution over time. If only common causes of variation are present and do not change, the output of a process is predictable（可預(yù)測(cè)）. Special causes (often called assignable causes) refer to any factors causing variation that only affect some of the process output. They are often intermittent（間

10、歇） and unpredictable（不可預(yù)測(cè)）. If special causes of variation are present, the process output will not be stable over time（不穩(wěn)定）. 特殊原因并不只是特殊差，可以特殊好。 Local actions局部行動(dòng) Are usually required to eliminate special causes of variation Can usually be taken by people close to the process Can correct typically a

11、bout 15% of process problems Actions on the system系統(tǒng)行動(dòng) Are usually required to reduce the variation due to common causes Almost always require management action for correction(People directly connected with the operation will be in a better position to identify them and pass them to management for a

12、ction.) Are need to correct about 85% of process problems Connections between actions and causes Confusion about the type of action to take Close teamwork between management and those persons directly connected with the operation Corrections or reaction plan Corrective action or preventive action De

13、tected potential 過動(dòng)兒（過嚴(yán)控制、干預(yù)或是過度調(diào)整） 植物人 The decisions about actions affecting the process require balancing the risk of taking action when action is not necessary (over control or tempering) versus failing to take action when action is necessary (under control). These risks should be handled in the

14、context of the two sources of variationspecial causes and common causes. Mistake 1. Ascribe a variation or a mistake to a special cause, when in fact the cause belong to the system (common causes) Mistake 2. Ascribe a variation or a mistake to a system(common cause), when in fact the cause was speci

15、al. Over adjustmenttempering is a common exle of mistake No. 1. Never doing anything to try to find a special cause is a common exle of mistake No. 2. 過程必須保證始終產(chǎn)生準(zhǔn)確和精密的結(jié)果不精密 精密準(zhǔn)確不準(zhǔn)確Statistical controlIn controlOut of controlCapabilityAcceptableCase 1Case 3Unacceptable Case 2Case 4Case 3 processThe cu

16、stomer is insensitive to variation within specifications.The economics involved in acting upon the special cause exceed the benefit to any and all customers. Economically allowable special causes may include tool wear, tool regrind, cyclical (seasonal) variations, etc.The process is mature. That mea

17、ns the process has undergone several cycles of continual improvement. The special cause to be allowed has been shown to act in a consistent manner over a known period of time.A process control plan is in effect which assure conformance to specification of all process output and protection from other

18、 special causes or inconsistency in the allowed special cases.過程控制 受控 （消除了特殊原因） 時(shí)間 范圍 不受控 （存在特殊原因） 過程能力 受控且有能力符合規(guī)范 （普通原因造成的變差已減少） 規(guī)范下限 規(guī)范上限 時(shí)間 范圍 受控但沒有能力符合規(guī)范 （普通原因造成的變差太大）Process capability能力 is determined by the variation that comes from common causes共同原因. It generally represents the best performan

19、ce of the process itself.最佳性能Process performance性能 is determined by the overall output of the process全部產(chǎn)出 . The data used to calculate process indices shall be gathered from processes that are in a state of statistical control.控制下 過程改進(jìn)循環(huán)1、分析過程 2、維護(hù)過程 本過程應(yīng)做什么？ 監(jiān)控過程性能 會(huì)出現(xiàn)什么錯(cuò)誤？ 查找變差的特殊原因并 本過程正在做什么？ 采取措施。 達(dá)到統(tǒng)計(jì)控制狀態(tài)？ 確定能力 計(jì)劃 實(shí)施 計(jì)劃 實(shí)施 措施 研究 措施 研究 計(jì)劃 實(shí)施 3、改進(jìn)過程 措施 研究 改進(jìn)過程從而更好地理解 共同原因變差 減少共同原因變差減少樣本數(shù)或是抽樣頻率