Distributing Demand through the week

  • Context:
  • Thread starter Thread starter Bushy
  • Start date Start date
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
3 replies · 2K views
Bushy
Messages
40
Reaction score
0
Hi all,

If weekly demand for a product is over five work days and assuming a bank holiday on a Thursday does not affect the weekly demand of 370 how should the total be distributed through the week ignoring Thursday?

Example

A normal week

[TABLE="class: MsoNormalTable, width: 150"]
[TR]
[TD="width: 64, bgcolor: transparent"][/TD]
[TD="width: 64, bgcolor: transparent"]Demand
[/TD]
[TD="width: 73, bgcolor: transparent"]Distribution factor
[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Mon
[/TD]
[TD="width: 64, bgcolor: transparent"]
100


[/TD]
[TD="width: 73, bgcolor: transparent"]
1.351351


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Tue
[/TD]
[TD="width: 64, bgcolor: transparent"]
80


[/TD]
[TD="width: 73, bgcolor: transparent"]
1.081081


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Wed
[/TD]
[TD="width: 64, bgcolor: transparent"]
80


[/TD]
[TD="width: 73, bgcolor: transparent"]
1.081081


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Thu
[/TD]
[TD="width: 64, bgcolor: transparent"]
60


[/TD]
[TD="width: 73, bgcolor: transparent"]
0.810811


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Fri
[/TD]
[TD="width: 64, bgcolor: transparent"]
50


[/TD]
[TD="width: 73, bgcolor: transparent"]
0.675676


[/TD]
[/TR]
[/TABLE]With no Thursday demand should the new distribution factors (yet to be decided) add to 4 given we have removed a day or can it be kept at 5, either way will give the total demand required but which is mathematically correct?
 
Physics news on Phys.org
Bushy said:
Hi all,

If weekly demand for a product is over five work days and assuming a bank holiday on a Thursday does not affect the weekly demand of 370 how should the total be distributed through the week ignoring Thursday?

Example

A normal week

[TABLE="class: MsoNormalTable, width: 150"]
[TR]
[TD="width: 64, bgcolor: transparent"][/TD]
[TD="width: 64, bgcolor: transparent"]Demand[/TD]
[TD="width: 73, bgcolor: transparent"]Distribution factor[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Mon[/TD]
[TD="width: 64, bgcolor: transparent"]
100


[/TD]
[TD="width: 73, bgcolor: transparent"]
1.351351


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Tue[/TD]
[TD="width: 64, bgcolor: transparent"]
80


[/TD]
[TD="width: 73, bgcolor: transparent"]
1.081081


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Wed[/TD]
[TD="width: 64, bgcolor: transparent"]
80


[/TD]
[TD="width: 73, bgcolor: transparent"]
1.081081


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Thu[/TD]
[TD="width: 64, bgcolor: transparent"]
60


[/TD]
[TD="width: 73, bgcolor: transparent"]
0.810811


[/TD]
[/TR]
[TR]
[TD="width: 64, bgcolor: transparent"]Fri[/TD]
[TD="width: 64, bgcolor: transparent"]
50


[/TD]
[TD="width: 73, bgcolor: transparent"]
0.675676


[/TD]
[/TR]
[/TABLE]With no Thursday demand should the new distribution factors (yet to be decided) add to 4 given we have removed a day or can it be kept at 5, either way will give the total demand required but which is mathematically correct?


Hi Bushy! :)

Both methods will give the same answer.
They are mathematically equivalent.
So both are mathematically correct.
 
Would you favour a particular method?
 
Bushy said:
Would you favour a particular method?

For myself I would prefer a third method.

Leave the distribution factors as they are, except for Thursday for which the factor will become zero.
Then calculate the new demand for each day with:

$\text{Demand} = \dfrac{\text{Distribution factor}}{\text{Sum of all distribution factors}} \times 370$

As you can see, there is no 4 or 5 involved.

Originally, the distribution factors were calculated from the actual demand.
Now that you want to create a forecast for a week with a bank holiday, that forecast is based on the previously calculated distribution factors.
I feel that they should not be changed to find the new demand - they are your input.