Joining the two datasources would be complicated.
The fields to join are the [date], [from] and [to] --
and the duplication would occur because of
the different granularity of the tables to join.
So the blend approach seems feasible.
The (business) question still remains:
How would you measure the Occupancy?
Answering this question would imply
the way(s) to the solution (if exists).
Hi Yuri, Thanks for Your reply.
I already used KNIME to join the datasources, maybe I should switch to right join, so I will always have the full and correct capacity.
But I still have problems with the correct figures.
If the occupancy is 80%, I would also like to see what part of this 80% is booked via call center, but if the capacity changes as well, I will never have the correct figures.
Thank you for reply.
The original problem is lying in the fact that
the granularity of your datasources is different --
the Capacity is 'coarser' (less detailed) than the Nr of Bookings.
Moreover, joining them together (even with the Full Join) doesn't work either,
because of the range of Dates in the Capacity datasource to be less than
the one in the Nr of Bookings -- when the Sales Channel is taking into account.
Thinking about it further, i came up with the approach
using a specially prepared 'Scaffold' datasource,
and joining both your extracts to it (at their granularity).
Please find the attached.
Hope it could help a bit.
this seems to be a working workaround. I will try this with our dataset, but I am afraid this will cause an increase that is too big to work with.
But thanks for the answer!
Sander, you're welcome.