Home › Forums › Operations › Manufacturing › Scrap Rate Weighted by Output Per Operator

This topic contains 5 replies, has 4 voices, and was last updated by Chris Seider 1 week ago.

Viewing 6 posts - 1 through 6 (of 6 total)

- AuthorPosts

JRHi guys, hoping someone can help me. Is there some way to calculate a scrap rate for different parts, weighted on the number of parts produced by each operator for that specific part. For example, lets say I have 3 operators (A, B, and C) each making 4 different, but similar, parts (1, 2, 3, 4…..10).

Operator A produced

5 part 1’s and scrapped 80%

65 part 2’s and scrapped 40%

60 part 5’s and scrapped 25%

35 part 7’s and scrapped 15%.Operator B produced

75 part 3’s and scrapped 45%

35 part 5’s and scrapped 40%

40 part 6’s and scrapped 75%

30 part 8’s and scrapped 20%Operator C produced

80 part 2’s and scrapped 40%

90 part 3’s and scrapped 15%

45 part 5’s and scrapped 2%

40 part 9’s and scrapped 5%I’m trying to determine if one operator is performing significantly better than their peers so we can capitalize on any best practices such as loading techniques, etc. In this example it appears that Operator C is performing best, but say for instance, we know part 6 generally has a higher scrap rate and Operator C didn’t run any. Also, Operator A had one of the highest scrap rates with part 1, but only ran 5 pieces. I hope this makes sense, and any help will be appreciated.

Put your data into statistical analysis software such as Minitab, or at minimum into a 3D matrix in a spreadsheet and graph it. If you have questions about what it’s telling you, ask again and someone here will probably help.

JR so when it says Operator A produced 5 part #1’s and scrapped 80% then he basically scrapped 4 out of 5 parts. Am I reading that correctly?

Without any analysis, there is a part number difference. If they are similar then there is a real issue. With scrap rates this high I can’t imagine you are going to learn more from analysis than you will from standing on the line watch people build parts.

I would agree with Mike here.

I would also want to look at Op C. Their reject rates for parts 5 & 9 are significantly lower that the other 2 It’s. So, is that because they have a better process – and one to learn from? Or is it that they have the same problems as everyone else, but can’t recognise the bad parts as well?

Chris ButterworthThose are high scrap rates!

I would use a contingency table to assess statistical significance between operators and also between parts.

Consider some graphical analysis with your 2 X’s shown–operator and product

- AuthorPosts

Viewing 6 posts - 1 through 6 (of 6 total)