This topic contains 10 replies, has 4 voices, and was last updated by Monica 1 week ago.
Hi, does anyone know how to calculate DPMO for XmR control chart?
I’ve seen that the formula of DPMO is (Total Defects / Total Opportunities) * 1,000,000
But i still don’t get it, do we still call it Total Defects even if the data is variable?
We can use DPMO for both discrete and variable data, right?
I would be very thankful if anyone can give me a simple example.
Your answers mean a lot. Thank you!
I’m obviously feeling benevolent today but still think you need to suggest the answer and why. My two penn’orth is this extract from the dictionary that may turn on a light for you. Read it and then suggest the answer.
Continuous data is information that can be measured on a continuum or scale. Continuous data can have almost any numeric value and can be meaningfully subdivided into finer and finer increments, depending upon the precision of the measurement system.
As opposed to discrete data like good or bad, off or on, etc., continuous data can be recorded at many different points (length, size, width, time, temperature, cost, etc.).
Continuous data is data that can be measured and broken down into smaller parts and still have meaning. Money, temperature and time are continous.Volume (like volume of water or air) and size are continuous data.
Let’s say you are measuring the size of a marble. To be within specification, the marble must be at least 25mm but no bigger than 27mm. If you measure and simply count the number of marbles that are out of spec (good vs bad) you are collecting attribute data. However, if you are actually measuring each marble and recording the size (i.e 25.2mm, 26.1mm, 27.5mm, etc) that’s continuous data, and you actually get more information about what you’re measuring from continuous data than from attribute data.
Data can be continuous in the geometry or continuous in the range of values. The range of values for a particular data item has a minimum and a maximum value. Continuous data can be any value in between.
It is the data that can be measured on a scale.
Thank you Mr. Parr for answering.
Yes, it is a continuous data and it can be measure on a scale.
So actually i’m working on my thesis about minimizing cooking oil waste in instant noodle industry by implementing six sigma. The waste is measured on kg basis. But i’m confused how to measure the DPMO while my data is continuous, because the most literature i found in the internet is measuring discrete data (defects). Could anyone help me? Thanks!
Monica – at what point does your waste become a defect? Is it when the amount is outside control limits? is it when there is any waste at all?
Your waste is on a kg basis but kg per what in terms of output? You need to decide what is acceptable and what constitutes a defect. What are you measuring and for what purpose?
If you cannot determine when you have a defect maybe you just need to think in terms of reduction in waste rather than DPMO.
As someone might say, just my opinion.
Monica – use whatever unit of measure makes most sense. Kg, liters, etc. Just make sure that you use that unit of measure for all values. Hope this helps.
Thanks for kindly answering me. Well i’m working to analyze the usage level of the cooking oil, because based on the report this industry seemed using too much oil that make it wasteful. Daily, the oil usage is measured in kg basis, but in the report its stated in percentage of waste. Let’s say that they have specification limit of oil waste level, which is 0.3%, and i found that for 3 months period, the level of oil waste were 0.6%, 0.8%, and 0.4%.
If I want to measure the sigma level of the oil waste, what do i suppose to do? Because based on the literature, i need to calculate the DPMO first before calculating the sigma level.
Your answers mean a lot. Thank you.
It’s not as easy of question. Is there a technological waste as part of the waste %? Would you care to analyze waste above that technological waste or all of the waste?
@MBBinWI will have even more ideas for you.
Monica – What is your operational definition for oil waste? Is it how much is added to bring the amount up to regular level? Some of that would be absorbed by the items cooked. Different types and sizes will have different levels of absorption.
To get back to your original question, variable data is no problem in determining DPMO. You should take data over a period of time. Determine the mean and std dev for the data. You can identify how many std dev’s you have between the mean and the spec level. What is outside of that level is your defects. Find a Z-table and you can establish the number of defects. Multiply by 1M for DPMO. There are many versions of Z-tables, so you’ll need to identify how yours is established to get the defect level.
Actually the oil waste refers to
((actual oil usage – standard oil usage) / standard oil usage)*100%
So what i mean is somehow this industry just used too much oil more than it’s allowed. They do not wasting the oil away, but actually they just haven’t found the optimum processing parameter which make the process yet totally controlled. So, yes, this oil waste refers to a condition when they use certain amount of oil which exceed the regular or allowed standard.
@MBBinWI I guess i need to calculate the DPMO by the formula
DPMO USL = P[z≥(USL-x) / s] × 1000.000
Is this correct?
Thank you so much.
That looks correct.
From your definition, you have two potential sources of error – 1) the “standard usage” amount, and 2) the measurement of the replacement oil used. As you dig into this deeper, keep these in mind.
Ah i see. Thank you so much for your answers! It really helps me.