I am modeling our warehouse operations and it has a bunch of line items and "categories" as follows:
The line items I have are Operation1, Op2, and so on.
For each line item, they are either a function that is "Inbound"/"Outbound" and belong to "Direct Labor" or "Indirect Labor". These categories go into tracking KPIs eventually.
Each line item also has associated measures like Volume, # of headcount on that operation, throughput and so on.
I have modeled the above as follows:
Module with line items as above.
List dimension that has measures like: Volume, headcount, etc; and "Applies To" all line items
Any ideas on how to model the "categories" and also, how to import the data into the module (I ignored categories for now, and am stuck on how to import it as the columns are three in number for each Date/Time dimension as seen in the data table I have posted)
Please guide on how to model this, and how to import.
Solved! Go to Solution.
A couple of questions...Is Operation essentially your transaction ID where you could have 1,000's of them? Also, can Operation 1 only be Inbound and direct or can there be another Operation 1 with the values of Inbound and Inbound? If the answer is yes to the last part, create a list having a code of the code of Operation 1 and concatenate that with the I/O and D/I codes. So, the code could look like this:
Codes for IO and DI:
Inbound = 100
Direct = 200
Indirect = 400
***Make sure you have a delimiter to separate the different pieces of information in your transactional list ***
Transactional data would look like this:
Create a module based off the above list (without Time) with two line items (I/O and D/I). Since you already know what the values are based off the code, you can formulaically figure out the value using Finditem().
For the actual transactional amounts, create a module based on the list above that you just created with 3 line items: Volume, Headcount, Throughput which is dimensionalized by Time.
Load the data in.
Hope this helps,
Thanks for answering my questions on this forum, again!
To answer your question: No, the no. of operations are not too large -- around 20 of them and I do not expect them to change a lot over time. These are not transactions but describe a process -- hypotehtically, say it refers to some operation on the assembly line. For example: Operation1 could be fitting the wheels in a car, Operation 2 could be fitting the body and so on.
Every operation is either part of an "Inbound" or an "Outbound" (cannot be both) process. Every operation is also belonging to "Direct Labor" or "Indirect Labor" and cannot be both.
So the I and D codes are unique to each operation. You cannot have duplicate Operation with different codes.
Under such circumstances, do you still recommend going with your solution?
Note that I have modeled various "Operations" as line items as they will all roll up to various categories in other modules and ultimately into "Direct Inbound Labor Cost" or "Indirect Inbound Labor Cost" or "Direct Outbound Labor" and so on.
If you still recommend the same solution, can you throw more light on how you construct "Operation_100_200" ?? Do you create a list with Codes for IO and DI and then create line items which concatenate these codes? How do you do that?
Actually, I would not do ALL of what I said earlier with the concatenation of the codes of I/O and D/O. Can you just have a list of Operations with the metadata about them (I/O, D/O) in a module and then in another module (dimensionalized by Time and Operation), you have the three line items I had above.
Your second post/response makes sense to me. I am going to try implement that and see if it works for my purpose and update/accept this solution.
Correct, simply is always better than complex and by making the formulas easier to read as well as maintain, you are helping the system. So yes, break them out.