【正文】 d training course 控制圖 控制圖是 SPC的主要部份及靈魂，用以 進(jìn)行中的過(guò)程控制 / 監(jiān)察， 確定問(wèn)題 (變差來(lái)源 )， 在問(wèn)題未發(fā)生前找出問(wèn)題的趨勢(shì)，及 判斷何時(shí)採(cǎi)取適當(dāng)?shù)男袆?dòng) 為過(guò)程提供標(biāo)準(zhǔn)值 確定過(guò)程能力 在繪圖控制前，必須進(jìn)行 MSA 以確保量具的能力。 SPCadvanced training course 控制圖的種類(lèi) 特征值 控制圖的種類(lèi) 計(jì)量型數(shù)據(jù) X R chart (平均值及極差 ) X s chart (平均值及標(biāo)準(zhǔn)差 ) X MR chart (單值及移動(dòng)極差 ) Precontrol chart (當(dāng)員工未有接受統(tǒng)計(jì)概念訓(xùn)練， 這是最合適的使用方法 ) 計(jì)數(shù)型數(shù)據(jù) np chart (不良產(chǎn)品數(shù) ) p chart (不良率 ) c chart (缺陷數(shù) ) u chart (每個(gè)產(chǎn)品的缺陷數(shù) ) _ _ SPCadvanced training course ? There are 4 types of Attribute Charts 43 u p c np Constant Lot/Unit Size Variable Lot/Unit Size Defects (Poisson Distribution) Percent/Fraction Defective (Binomial Distribution) SPCadvanced training course 控制圖的發(fā)展及使用 推行控制圖的 4個(gè)階段 1. 抽樣計(jì)劃 計(jì)劃組的容量、抽樣頻率、子組的數(shù)目 . 2. 收集數(shù)據(jù) 從新的、有代表性樣本的過(guò)程收集數(shù)據(jù) 3. 過(guò)程控制及解釋 – 184。213。165。206。控制極限；特殊原因的對(duì)策 4. 能力的確定及改善 – 過(guò)程能力；普遍成因的對(duì)策 SPCadvanced training course 樣本量 0 . 00 . 10 . 20 . 30 . 40 . 50 . 60 . 70 . 80 . 91 . 00 0 . 5 1 1 . 5 2 2 . 5 3Sh i ft, i n Sta n d a r d D e v i a ti o n sProbability of DetectionProbability of Detecting a Shift in Mean in the Next Subgroup for Various Control Chart Subgroups of Size n n=20 n=15 n=10 n=8 n=5 n=4 n=3 n=2 ? Sample Size = 8 90% confidence of detecting SPCadvanced training course 建立控制圖 X R ? Choose an appropriate subgroup size – 5 is a reasonable choice for an initial investigative chart ? Take at least 25 subgroups ? Enter the data in column form (Minitab) – Response is one column – The subgroup number or designation can be put in another column, but is not necessary for equal sizes ? In Minitab: Stat Control Charts Xbar R – Enter the Response Column – Enter either the subgroup size or the subgroup column SPCadvanced training course 時(shí)間 nxsCL??339。 ??R?339。 ??? RsCL置信區(qū)間 SPCadvanced training course XR 控制圖 SPCadvanced training course 過(guò)程控制 a) 計(jì)算及繪畫(huà)控制限 樣本容量低於 7時(shí)，極差並沒(méi)有 LCL SPCadvanced training course 解釋控制圖 U CL LCL Avg U CL LCL Avg Change in level Change in variability SPCadvanced training course 解釋控制圖 確定是否有不受控的情況 檢查過(guò)程記錄，找出並列明任何特殊原因。 再計(jì)算控制限。 控制上限 (UCL) 中線 控制下限 (LCL) 控制上限 (UCL) 中線 控制下限 (LCL) 受控過(guò)程 不受擯過(guò)程 SPCadvanced training course UCL LCL A A B C C B UCL LCL A A B C C B UCL LCL A A B C C B UCL LCL A A B C C B Test 1. One Point Beyond Zone A Test 2. Nine Points in a Row on One Side of the Center Line Test 3. Six Points in a Row Steadily Increasing or Decreasing Test 4. Fourteen Points in a Row Alternating Up and Down x x x x x x 控制圖失控檢驗(yàn) SPCadvanced training course UCL LCL A A B C C B UCL LCL A A B C C B UCL LCL A A B C C B UCL LCL A A B C C B Test 5. Two Out of Three Points in a Row in Zone A or Beyond Test 6. Four Out of Five Points in a Row in Zone B and Beyond Test 7. Fifteen Points in a Row in Zone C (Above and Below Centerline) Test 8. Eight Points in a Row on Both Sides of Centerline with None in Zone C x x x x x x x 控制圖的失控檢驗(yàn) SPCadvanced training course UCL LCL A A B C C B 控制圖失控檢驗(yàn) One point above UCL Two out of three successive point here or above Four out of five successive point here or above Seven out of eight successive point here or above One point above LCL Two out of three successive point here or bellow Four out of five successive point here or bellow Seven out of eight successive point here or bellow 3? 2? 1? 3? 2? 1? 0 % % % % % % % % SPCadvanced training course 2 0 1 0 0 O b s e r v a t i o n N u m b e r UCL LCL 2 0 1 0 0 O b s e r v a t i o n N u m b e r UCL LCL Seasonal Trends: These could be caused by systematic changes like: temperature, operator fatigue, voltage or pressure fluctuations, etc. Mixtures: A clear indication is when most of the points are close to the control limits, and almost no points near the center line. These are usually due to underestimating the process variability or “overcontrol” , where operators intervene with the process frequently trying to respond to inherent variability instead of assignable causes. This also occurs when controlling two processes on the same chart. 控制圖的趨勢(shì)判斷規(guī)則 SPCadvanced training course 2 0 1 0 0 O b s e r v a t i o n N u m b e r UCL LCL 1 5 1 0 5 0 O b s e r v a t i o n N u m b e r UCL LCL Changes in process means: These are usually due to the introduction of new operators, machines, procedures, etc. Can also happen because of process improvements. Upward/Downward Trends: Continuous trending on the same direction. These are usually caused by ponent wear among others. 控制圖的趨勢(shì)判斷規(guī)則 SPCadvanced training course 2 0 1 0 0 O b s e r v a t i o n N u m b e r UCL LCL “ White Space”: Points tend to cluster around the mean. Usually caused by: (1) incorrect control limits (overestimating process variability), (2) continuous improvements payingoff. Limits should be recalculated when this situation is found. 控制圖的趨勢(shì)判斷規(guī)則 SPCadvanced training course 製 程 控 制 對(duì)特殊原因採(cǎi)取的措施 ，以便將不符合規(guī)格的產(chǎn)品減至最少；並取得新的數(shù)據(jù)作調(diào)查分析。 ，我們可能會(huì)在任何單值子組中得到一個(gè)特殊原因的錯(cuò)誤信息。 SPCadvanced training course Example of OCAP in W7P