制程统计管制(SPC)初级培训.ppt

Two kinds of data can be used for measuring purposes: Discrete (attribute) and Continuous (variable). One way to look at the data is to merely count the frequency of occurrence; e.g.... the number of times something happens or fails to happen. Notice that such data is not capable of being meaningfully subdivided into more precise increments and, therefore, is said to be discrete by nature. Go/no go and pass/fail data are also forms of discrete data. Data that characterizes a product or process feature in terms of its weight, volts, size for example, is said to be continuous by nature. In other words, the measurement scale can be meaningfully divided into finer and finer increments of precision. Time is a continuous scale, but meeting or not meeting a “due date” is discrete data. To apply the normal distribution, one must necessarily use continuous data. As an example, the Agent suggested we consider the process of driving a car. To control this process, the driver uses both discrete and continuous data on the dashboard. The speedometer provides continuous data about the speed. Warning lights provide discrete data. A warning light only tells the driver when the coolant fluid is too hot. A temperature gage would tell the driver the actual temperature. Also, by monitoring the gage, the driver may see trends and then may forecast a problem, allowing him time to do something about it. From this analogy, one can readily see that continuous data provides much more information than discrete data. Because continuous data provides much more information than discrete data, it is usually considered to cost less to collect continuous data than discrete, because we require so much less of it (for an equivalent level of knowledge provided) 製程統計管制(SPC)初級班培訓練課程 課程大綱 1.基本品管統計概念 2.製程統計品管(SPC)基本概念. 3.Cpk基本概念 4.管制圖基本概念 基本統計概念 變異 變異乃宇宙現象,它是人類生活中的一部分, 因此 學習如何適應變異乃智慧之一端. 統計學意義 統計學:是處理數量數據的收集.列表.分析.解 說和陳述的一種科學. 統計學分類 記述統計:用來描述和分析一群事物. 推論統計:依據有限的數據(樣本)中求得有關大量 數據(群体)的重要結