A
Andrews
How effective is 100% inspection ? If we have to do 100% inspection for the interim period of taking corrective action , then how can we improve the effectiveness of 100% inspection.
Elsmar Forum Sponsor
It might help if you told us what is being inspected, and what problem the corrective action is for.Andrews said:
Claes Gefvenberg
Admin
Jennifer and JSW05 are both correct imo. There is a world of a difference between an inspection where you can see a defect from a mile away and looking for something that is barely measurable. Depending on the circumstances the result could range between nil and 100%
As Jennifer noted, I think we need more data.
/Claes
If we are talking about eyeballing: Not very, and definitely not as effective as most people would like to think. It is however, a great deal more effective than no inspection at all.Andrews said:
As Jennifer noted, I think we need more data.
/Claes
R
Randy Stewart
I believe there was a study done at John Hopkins on this very topic. What they found was that 100% inspection will catch approximately 85% of the problems.
There were a bunch of variables, fatique, gauges, type of issue etc. but the bottom line was that 15% of all errors were not found.
There were a bunch of variables, fatique, gauges, type of issue etc. but the bottom line was that 15% of all errors were not found.
R
Rob Nix
Someone asked a similar question on the ASQ Discussion Boards about a week ago, to which I replied, similar to Claes' comment...
Rob Nix @ ASQ said:
A
Andrews
100% inspection is for a diameter 15.87/ 15.83. The cause for defect in this characteristic is plating thickness variation on this component.
There's simply no way for anyone here to know more than you do about this. Think of all of the variables that might contribute to failure of a 100% inspection process, ranging from gage error to operator error to the quantity of parts to be inspected to the conditions (light, etc.) under which the inspections take place...Andrews said:
You have to use your own best judgement and knowledge of the conditions in deciding the best way to approach this. There is no universal formula for reliability of 100% inspection that will help you.
Andrews said:
How good is homemade ice cream? That's your answer. In other words, as others have said, it varies greatly -- from awful to darn near perfect.
In your case, you're measuring a tight tolerance (if the units are mm), so if the "bad" parts are "close" and especially if they are not perfectly round, and there are lots of them, expect that you will be well less than perfect. Maybe some testing of the test is in order and you may want to implement a "guard band" in the customer's favor.
I strongly agree with JWS: There is no universal formula for reliability of 100% inspection that will help you.
JSW05 and Mike S are both correct in that there is an slmost infinite range between the possible and what you get. Variables that influence outcomes include, among others:Andrews said:
1. How much time you are willing to devote to the inspection.
2. Inspecting each part once, in one place, to 50 times in 50 places--the detail you are willing to go to.
3. The tools, equipment and environment you use--their sensitivity and accuracy. Hot environments may induce a slight expansion coefficient effect in some materials.
4. The inspectors' training, experience, technique, visual acuity, state of mind, health, fatigue level, consistency.
5. Damage to the part created with this extra handling that may cause you to minimize the inspection process.
6. The parts' accuracy/size before plating.
7. Plating process: is there more material in some areas than others? Crud in the plating or on the parts while going through the process?
So you see why it is impossible to give a direct answer. Even on different days in the same place the results will likely vary.
Similar threads
