SPC进阶培训.ppt

* B. STATISTICS, Participant Workbook, page 3 Each of these points could represent an average of several pin heights or circuit widths. Question: If they were averages, what type of control chart would this be? Answer: average or average chart. Question: If the points were ranges, what type of control chart would this be? Answer: Range or R chart. Question: If the points were the individual measurement values, what type of control chart would this be? Answer: X or individuals chart. There’s quite a bit we can learn about our processes from data like this. When we put the data on a control chart, the points form an ever-changing picture. It’s like having a movie of the process. There’s one more statistic I’d like to introduce now. It’s another way to get a picture of the data from a process. * B. STATISTICS, Participant Workbook, page 4 The third type of statistic you need to know about is called distribution. A distribution tells us where the data points are concentrated across the range. For our height example, we could establish different height categories in inches (or centimeters) and see how many people in the room would be in each category. Note: Give examples of categories of height (people less than 5 feet tall, 5’ to 5’6”, 5’7” to 6’, etc.). In our example, we’ll just use adjectives -- short, tall, etc. Or, have the participants form a human histogram by lining up in predetermined height categories. Once we establish various height categories, we count to see how many people are in each category. That tells us how your heights are distributed across the full range of heights of people in the room. Are there lots of short people or only a few? PRESS DOWN ARROW KEY TO ANIMATE THE SLIDE. Question: What questions do you have at this point? Note: Respond to questions. Question: What if these were pin heights on a backplane or the depth of drilled holes on a circuit board? Would it be helpful to know how the pin heights were dis