Newbie Capability question

Bjourne · Jun 8, 2014

Hello,

Seems to be a very intellectual group of guys here

I'm new here to this forum(I usually just visit and participate in security forums such as Wilderssecurity and MalwareTips) and I just wanted to know more about Capability study. I am not an engineering graduate but have been exposed to semiconductor manufacturing process way back 2007 as I work in molding department for IC encapsulation. I want to learn more about the process side of things in our company but sad to say there isn't any training available or if there is I am darn busy at the production line(most of the time).

I have some notes that I got from some of our engineers. Some notes that they use for Capability studies but I am having a hard time learning it. Can I post it here so someone might check if I did it wrong...?

This is a sort of self-studying just in case I get the slot for promotion or can transfer to another company...just wanna learn more and not be stuck being a technician. I hope you understand.

Thanks for understanding.

Regards,

Bjourne

dgriffith · Jun 8, 2014

Re: Newbie question

Bjourne said:
. . . just wanna learn more and not be stuck being a technician. I hope you understand.

Thanks for understanding.

Regards,

Bjourne

It's good that you want to learn more. Wish I'd learned so much more when I was younger, and I'm still learning.
However, and I'm sure you didn't mean it the way I'm reading it, but I must come to the aide of the technician. I've been stuck being a technician all my life. Good technicians are needed--who the heck else is going to keep engineers straight? Keep them on their toes by catching their errors? Suggest better ways to implement their napkin designs? Turn their plans into working constructs? Test their ideas, calibrate their instruments, machine their parts, and so on. Most importantly, tell them when they're wrong!
Good technicians keep engineers out of trouble, even if they are stuck.

And, please, do post your questions here and many will be happy to look at them. There are also many threads with similar topics you can research.

Wes Bucey · Jun 8, 2014

Welcome to the Cove! :bigwave:

Here's a link to give you a broad outline of a process capability study (https://qualityamerica.com/knowledg...how_to_perform_a_process_capability_study.asp)

For an on-line text (FREE) NIST's Engineering Statistics Handbook
is a great resource.

I took the liberty of starting you off with this link to "process capability" in the Handbook: http://www.itl.nist.gov/div898/handbook/pmc/section1/pmc16.htm

Happy learning!

Bjourne · Jun 10, 2014

Re: Newbie question

dgriffith said:
It's good that you want to learn more. Wish I'd learned so much more when I was younger, and I'm still learning.
However, and I'm sure you didn't mean it the way I'm reading it, but I must come to the aide of the technician. I've been stuck being a technician all my life. Good technicians are needed--who the heck else is going to keep engineers straight? Keep them on their toes by catching their errors? Suggest better ways to implement their napkin designs? Turn their plans into working constructs? Test their ideas, calibrate their instruments, machine their parts, and so on. Most importantly, tell them when they're wrong!
Good technicians keep engineers out of trouble, even if they are stuck.

And, please, do post your questions here and many will be happy to look at them. There are also many threads with similar topics you can research.

I feel the same way when it comes to being a technician. Some times I seem to feel that they are more theoretical rather than actual. I do the dirty work and they the coffee and meetings and meetings and reports. There are times that there are some basic know-how that engineers do not know! Anyway that is the reason that I wanna learn more so if possible when I get out of this job and I have a chance to be at least an associate engineer I'd know what to do

Thanks there

Bjourne · Jun 10, 2014

Wes Bucey said:
Welcome to the Cove!

Here's a link to give you a broad outline of a process capability study

For an on-line text (FREE)
is a great resource.

I took the liberty of starting you off with this link to "process capability" in the Handbook: ....

Happy learning!

Hi!

Thanks for the reply guys I really appreciate it! I am downloading the handbook now as I type

Thanks there!

Bjourne · Jun 10, 2014

May I post a question now?

I was reading on a pdf file about Control Charts and I got interested on defects. We have some or a lot of defects at the production line and I read that you can use npChart it. I am in the process of charting the data I have, but something hit me....if I plotted it as correctly and there are outliers or out-of-control points, how can I help improve that particular process from where the defect originated?

The process is stable when there are no outliers right?

So if I have, say, 2 outliers from a given upper limit / lower limit in the npChart how can I control that outlier? Upper limit / Lower limit or Upper Control limit / Lower Control limit...which is correct?

What can be done to eliminate that outlier?

Might you show me some examples please....

Thanks guys

So sorry I was not able to post an image of the sample data table I got from work. It's here.

Question: I see that when I post with a link I am reminded of a spam warning...how can I post images from imgur? I got 2 warnings from quoting a reply...I dont wanna be banned so I asked

Thanks again

Bev D · Jun 10, 2014

the phrase to use is "control limit" not just "limit"...

in the data you posted there is one subgroup that is beyond the upper control limit.

In order to 'control' that you must find the cause and take action to prevent recurrence.

When you have an out of control point, we say that there is an "assignable" cause. In other words either an input factor ahs varied to a level that is different than in the previous subgroups, or there is a new factor tha is present. in theory, assignable causes are 'easier' to determine because they are 'new' or fresh or come from an obvious cause.

the baseline "stable" variation is known as "common cause" variation becase it si the daily , normal variation in the process and improving it typically takes more involved problem solving methods and tools.

Now these are only approximate truths. I have encountered 'assignable cases that were very difficult to understand and fix and common causes that were really very simpel and easy to determine and fix.

None the less, in order to 'control' the defects one must use problem solving methods to understand the cause and correct it....

Bjourne · Jun 11, 2014

Bev D said:
the phrase to use is "control limit" not just "limit"...

in the data you posted there is one subgroup that is beyond the upper control limit.

In order to 'control' that you must find the cause and take action to prevent recurrence.

When you have an out of control point, we say that there is an "assignable" cause. In other words either an input factor ahs varied to a level that is different than in the previous subgroups, or there is a new factor tha is present. in theory, assignable causes are 'easier' to determine because they are 'new' or fresh or come from an obvious cause.

the baseline "stable" variation is known as "common cause" variation becase it si the daily , normal variation in the process and improving it typically takes more involved problem solving methods and tools.

Now these are only approximate truths. I have encountered 'assignable cases that were very difficult to understand and fix and common causes that were really very simpel and easy to determine and fix.

None the less, in order to 'control' the defects one must use problem solving methods to understand the cause and correct it....

Thanks for the reply there

I think I need to get the findings of what happened first that arrived at that data...Do allow me to get the info from my officemates and then I'll plot the npChart (crudely with Excel) and then post it here. I hope for your understanding guys as really I am a beginner and am really trying to study/learn more

Newbie Capability question

Bjourne

dgriffith

Quite Involved in Discussions

Wes Bucey

Prophet of Profit

Bjourne

Bjourne

Bjourne

Attachments

Bev D

Heretical Statistician

Bjourne

Similar threads