I couldn't find a similar question from earlier posts so here it goes.
We have three different products. Let's call them A, B and C.
We know that on a monthly level we produce 50% of A, 40% of B and 10% of C.
The total throughput is ranging somewhere around 80-120 depending on the actual share of products (high variety comes from the fact that C equals 10pcs of A from capacity PoV).
Production orders come into the process unevenly i.e. first week of month we get a relatively steady load of e.g. 5 production orders on Monday, 7 on Tuesday, then 6, 5, and finally 8 on Friday.
But then next week the situation changes and for some reason we get 13 production orders on Monday, 11 on Tuesday, 2 on Wednesday, 3 on Thursday, and finally 2 on Friday.
So even though the weekly amount is the same (it doesn't have to be) the throughput times will be way different between these weeks as the latter will cause longer queues.
We are measuring the throughput time as a KPI and should be able to deliver within 24 hours so obviously the latter week is nightmare from that point of view.
The aim is to be able to compare two months against each other where other month only has steadyly inflowing production orders while the other can have one or two weeks of "rush hours" like in above example.
Any ideas to how to model the source to create those production orders in a way we would need? We couldn't find a distribution to fit our needs but it can be also so that we didn't try a correct one and/or configured it improperly.
Solved! Go to Solution.
one way to implement this would be to use a delivery table in your source. This delivery table then can be filled and configured using methods to suit your needs. Attached is a model where this was done using roughly the information you provided. Would that implementation cover your case? If not, could you provide a model where you implemented the described behavior?