To prevent spam users, you can only post on this forum after registration, which is by invitation. If you want to post on the forum, please send me a mail (h DOT m DOT w DOT verbeek AT tue DOT nl) and I'll send you an invitation in return for an account.
How put 2 sets of data together?
Hi,
my name is Richard and I am currently writing my bachelor thesis about process mining. I've gathered some data about processes from a partner company which is quite nice, but I am not sure how make best use of this.
The data is about both the order administration and manufacturing processes. The tricky part is that orders from customers can have more than one ordered items. Sometimes the same customer order gets splitted apart when it comes to manufacturing the ordered items.
for example Order 001:
Order 001 consits of Item1, Item2 and Item3 and arrived at 12 am
After order registration there will be 3 seperate "sub orders" for manufacturing:
Order 0011 for producing item 1
Order 0012 for producing item 2
Order 0013 for producing item 3
All 3 suborders would then be seperatly processed, each with their own timestamps. Since I've got data for both the customer and manufacturing orders, the question is now how can I put together those sets of order data?
Option 1:
I can map the manufacturing sub orders to the original customer orders. E.g. all 3 Order 0011, Order 0012 and Order 0013 would turn into Order 001. This means I can evaluate the complete process from the order from the incomming order from the customer to the succesfull shipment and billing but I lose the information from the context of the suborders.
Option 2:
I could create for every manufacturing sub order a corresponding higher order. E. g. for Order 0011 there would be Customer Order 0011 with Order arrival at 12 pm, Customer Order 0012 with Order arrival at 12 pm and also Customer Order 0013 with Order arrival at 12 pm. This doesnt seem like the optimal scientific approach and would tripple the already not so small data size.
Does anybody know if there is a nice way to merge those data sets together? Optimally there would be a possibility to start with the single customer order and lateron split into the seperate sub orders, kind of like the 3 way split seen at the grafik.
Currently I am working with option 1 but honestly this kind of feels not right.
Feel free to ask anything about the problem, I realy struggled while trying to explain myself but hope I got my problem accross
Help or suggestions about to approach this problem would really be appreciated )
Have a nice day
Richard
my name is Richard and I am currently writing my bachelor thesis about process mining. I've gathered some data about processes from a partner company which is quite nice, but I am not sure how make best use of this.
The data is about both the order administration and manufacturing processes. The tricky part is that orders from customers can have more than one ordered items. Sometimes the same customer order gets splitted apart when it comes to manufacturing the ordered items.
for example Order 001:
Order 001 consits of Item1, Item2 and Item3 and arrived at 12 am
After order registration there will be 3 seperate "sub orders" for manufacturing:
Order 0011 for producing item 1
Order 0012 for producing item 2
Order 0013 for producing item 3
All 3 suborders would then be seperatly processed, each with their own timestamps. Since I've got data for both the customer and manufacturing orders, the question is now how can I put together those sets of order data?
Option 1:
I can map the manufacturing sub orders to the original customer orders. E.g. all 3 Order 0011, Order 0012 and Order 0013 would turn into Order 001. This means I can evaluate the complete process from the order from the incomming order from the customer to the succesfull shipment and billing but I lose the information from the context of the suborders.
Option 2:
I could create for every manufacturing sub order a corresponding higher order. E. g. for Order 0011 there would be Customer Order 0011 with Order arrival at 12 pm, Customer Order 0012 with Order arrival at 12 pm and also Customer Order 0013 with Order arrival at 12 pm. This doesnt seem like the optimal scientific approach and would tripple the already not so small data size.
Does anybody know if there is a nice way to merge those data sets together? Optimally there would be a possibility to start with the single customer order and lateron split into the seperate sub orders, kind of like the 3 way split seen at the grafik.
Currently I am working with option 1 but honestly this kind of feels not right.
Feel free to ask anything about the problem, I realy struggled while trying to explain myself but hope I got my problem accross
Help or suggestions about to approach this problem would really be appreciated )
Have a nice day
Richard
Answers

Hi Richard,How do you see the aggregation of two different orders? In your example, you have an order 001 with three items. Suppose we would also have a second order 002 with only two items. How should the three items of order 001 be mapped onto the two items of order 002? Can this be done? You mention a threeway split, what about this split now?Kind regards,Eric.

Hi Eric,
thank you for your answer! Though I am not sure if I understand correctly what you mean with the mapping of the items.
I've double checked my customer order data set and realised that I can actually match my customer order positions with the manufacturing orders which means that I can trace my orders from the start to the end.
Still thank you very much
Greetings
Richard
Howdy, Stranger!
Categories
 1.6K All Categories
 45 Announcements / News
 224 Process Mining
 6  BPI Challenge 2020
 9  BPI Challenge 2019
 24  BPI Challenge 2018
 27  BPI Challenge 2017
 8  BPI Challenge 2016
 67 Research
 1K ProM 6
 390  Usage
 287  Development
 9 RapidProM
 1  Usage
 7  Development
 54 ProM5
 19  Usage
 186 Event Logs
 31  ProMimport
 75  XESame