Calculating weeks on hand based on forecast of subsequent weeks

Hi,

 

I have the following dataset:

 

Item

Date

Projected Inventory(PE)

Forecast(FC)

A

Jan 28, 2019

604

50

A

Feb 04, 2019

630

52

A

Feb 11, 2019

1303

49

A

Feb 18, 2019

1257

46

A

Feb 25, 2019

1176

51

A

Mar 04, 2019

1136

55

A

Mar 11, 2019

1083

53

A

Mar 18, 2019

1030

53

A

Mar 25, 2019

980

50

A

Apr 01, 2019

900

65

A

Apr 08, 2019

853

62

A

Apr 15, 2019

944

59

A

Apr 22, 2019

1303

49

 

The Desired O/ P is:

Item

Date

Weeks On Hand

A

Jan 28, 2019

11

A

Feb 04, 2019

12

A.

Feb 11, 2019

9

A

Feb 18, 2019

8

A

Feb 24, 2019

0 (there should be at least 10 weeks of Forecast for an item , if not the WOH = 0)

A

...

0

 

The logic is as follows:

I need to calculate the number of weeks on hand for each item for that particular week based on forecast for several weeks going forward.

So  for week of Jan 28, 2019 , the Projected Inventory is 604.

In order to calculate the Weeks On hand of Item A for week of Jan 28, 2019 the calculation is as follows:

 

FOR WEEK OF Jan 28, 2019:

 

Iteration 1:

 

604 – 50(FC for week 1) = 554

Weeks on Hand(WOH)  = 0 + 1 = 1

 

Iteration 2:

554 – 52(FC for week 2) = 502

WOH = 1 + 1 = 2

 

Iteration 3:

502 – 49 (FC for week 3) = 453

WOH = 2 + 1 = 3

 And so on…

 

Finally we get WOH = 11  for Item A for week of Jan 28, 2019.

 

FOR WEEK OF Feb,04 2019:

 

Iteration 1:

 

650 – 52(FC for week 2) = 598

Weeks on Hand(WOH)  = 0 + 1 = 1

 

Iteration 2:

598 – 49(FC for week 2) = 549

WOH = 1 + 1 = 2

 

Iteration 3:

549 – 46 (FC for week 3) = 503

WOH = 2 + 1 = 3

 And so on…

 

Finally we get WOH = 12  for Item A for week of Feb 04, 2019.

 

 

Similarly I need to calculate WOH for subsequent weeks and for different items

NOTE: Also there should be at least 10 weeks of Forecast for an item , if not the WOH = 0.

 

How to achieve this result in My SQL?

I am thinking Stored Procedures using cursors is the way to go. However I am not really familiar with Stored Procedure.

Is there an alternate method ? Any help will be appreciated.

 

Thanks in advance.

Prajwal

Comments

  • Do you have access to Redshift?  Or does it need to be a MySQL dataflow?


    “There is a superhero in all of us, we just need the courage to put on the cape.” -Superman
  • I have access to Redshift, yes. It does not necessarily have to be MySQL dataflow.

  • I'll try to work on this tomorrow.  Can I get an idea of how many Items are in your data set?  Are you also forecasting for multiple locations?


    “There is a superhero in all of us, we just need the courage to put on the cape.” -Superman
  • There are a total of 1172 Items , but all Locations aggregated for an item. So basically just 1 location.

    For each item there are upto 72 weeks of data so the dataset has a total of 84K+ rows.

    Thanks.

  • Could someone help me out with this or provide me with some direction?

     

    Thanks in advance.