Need Help with Case Study Questions

[Bev D]

its a trick question. you cannot do OQ without DoE. you cannot do cost effective and timely OQ without DoE. fractional factorials - otherwise known as robustness testing. the FDA standard on process validation is all over the use of DoE for OQ.

 

[LOUPEN]

Any chance you can look over my responses to the this case study and see if it's on target? It's a unfamiliar subject so I am replying on google and lots of reading to help me through it.

 

[Bev D]

my advice: answer to the best of your ability. when in doubt about what they are asking (for example with industry specific terms) be honest and ask for clarification. bring out your strengths, including your willingness to learn and take on new challenges.

most of the time these interviews aren't about getting the exact right answer, but more about your thought process and how you deal with them.

 

[Steve Prevette]

I think the assumption here would be that a defect slipped through the cracks even after control limits, procedures and inspection points are established throughout the process. Am I left to assume human error? What are the steps?

The case study gives you lattitude to make hypothetical assumptions. So, why make it artificially harder than it has to be? I'd run through scenarios of what could cause the receipt inspection to be ineffective, by what means the problem could slip through. The case study already points out a difficulty with needing destructive testing to find the problem. What could be better alternatives?

Bev's done a good job adding detail on the Qualification studies. In the end though, this is your job interview, and you need to answer naturally and not like you are overprepared, especially with someone elses' words. And they probably will throw a ringer or two at you during the running of the case study.

 

[LOUPEN]

I truly appreciate everyone's guidance with this. Thank you so much!

 

[TWA - not the airline]

I'd go with Steve's initial assessment: First of all concentrate on the problem-solving process. Start out with containment actions and corrections, implement an immediate corrective action that will allow you to start manufacturing again (like e.g. additional inspection steps (if applicable)), do the root cause analysis and the final corrective and preventive actions, verify/validate the actions and then set up the effectiveness control.
Show the people that you address such issues in a structured and analytical manner. For the root causes analysis think of gathering and assessing all available data first (Is/Is Not, review of existing documentation like batch records, DHF, initial qualification data, process risk management) then do Ishikawa, 5xWhy, Cause-Effect or FMEA or even brainstorming. Then set up your hypotheses and do the testing to prove/disprove them.

 

[Jim Wynne]

This is a case study given to me by a potential employer. Since I have been working primarily in manufacturing that's NON-Medical. So, I need some guidance on how I can answer these questions. Particularly, The last two examples are two areas I'm not too familiar with and I have "googled" various references but, they do not give specific examples on how to solve the "dilemma" exhibited on the case study. Can someone guide me? I'm all ears! Thanks guys!

For the third situation, note that there is a single important word in this:

...classic performance qualification (including DOE and Cpk studies) for each combination of machine/materials/tool can be cost- and time-prohibitive.

My emphasis.

The fact that all possible combinations can't be subjected to experiments doesn't mean that none of them can. The idea would be to consider what all of the different variables might have in common and arrive at a reasonable sampling of them, and be able to support the strategy with data. This is still a daunting task, but much less so than trying to test all of the possible combinations.

A word to the wise--if you attempt to present the impression that you know things you don't know, things might be very difficult for you if you were asked to discuss and elaborate on your answers to the questions.