Forum Discussion

Rajesh1's avatar
Rajesh1
Icon for Helper I rankHelper I
7 years ago

Max and Min date/times based on condition

hello all,

 

I have the following table:

 

v.ID.LocDateStatus
V1P01-Feb-2019 13:00:00Slow
V1P01-Feb-2019 12:30:00Slow
V1P01-Feb-2019 12:15:00Slow
V1P31-Jan-2019 23:00:00Slow
V1R31-Jan-2019 20:30:00Slow
V1R31-Jan-2019 18:00:00Slow
V1R30-Jan-2019 16:00:00Slow
V1R30-Jan-2019 14:00:00Slow
V1R30-Jan-2019 12:00:00Slow
V1R30-Jan-2019 11:00:00Normal
V1R30-Jan-2019 10:00:00Normal
V1R30-Jan-2019 09:00:00Normal
V1R30-Jan-2019 07:00:00Normal
V1R30-Jan-2019 06:30:00Slow
V1R30-Jan-2019 01:00:00Slow
V1R29-Jan-2019 23:00:00Slow
V1R29-Jan-2019 18:00:00Slow
V1R29-Jan-2019 15:00:00Slow

 

There will be more rows for different vehicle Ids (v. ID) but for I've limited it to one vehicle id.

The expected output is to identify the latest date/time and earliest date/time for vid+loc+status combination. Then, compute the difference in days for this combination. The expected output for the above should be as below:

 

 

v. ID.LocStatusFrom DateTo dateDays
V1PSlow31-Jan-2019 23:00:0001-Feb-2019 13:00:000.583333
V1RSlow30-Jan-2019 12:00:0031-Jan-2019 20:30:001.354167
V1RNormal30-Jan-2019 07:00:0030-Jan-2019 11:00:000.166667
V1RSlow29-Jan-2019 15:00:0030-Jan-2019 06:30:000.645833

 

I attempted various options like transpose, summarizecolumns with filter etc. but failed miserably.  Please can someone assist on how I am to go about achieving this.

 

thanks in advance.

Rajesh

Need Help

2 Replies

  • v-juanli-msft's avatar
    v-juanli-msft
    Icon for Community Support rankCommunity Support
    7 years ago

    Hi Rajesh1 

    Is there a column to define differnt vehicle running record as below?

    v.ID. Loc Date Status column
    V1 P 2/1/2019 13:00 Slow 1
    V1 P 2/1/2019 12:30 Slow 1
    V1 P 2/1/2019 12:15 Slow 1
    V1 P 1/31/2019 23:00 Slow 1
    V1 R 1/31/2019 20:30 Slow 2
    V1 R 1/31/2019 18:00 Slow 2
    V1 R 1/30/2019 16:00 Slow 2
    V1 R 1/30/2019 14:00 Slow 2
    V1 R 1/30/2019 12:00 Slow 2
    V1 R 1/30/2019 11:00 Normal 3
    V1 R 1/30/2019 10:00 Normal 3
    V1 R 1/30/2019 9:00 Normal 3
    V1 R 1/30/2019 7:00 Normal 3
    V1 R 1/30/2019 6:30 Slow 4
    V1 R 1/30/2019 1:00 Slow 4
    V1 R 1/29/2019 23:00 Slow 4
    V1 R 1/29/2019 18:00 Slow 4
    V1 R 1/29/2019 15:00 Slow 4

    From your screenshot and provided information, it is hard to differ the row 1 and row4.

    v. ID. Loc Status From Date To date Days  
    V1 P Slow ######## ######## 0.583333 1
    V1 R Slow ######## ######## 1.354167 2
    V1 R Normal ######## ######## 0.166667 3
    V1 R Slow ######## ######## 0.645833 4

     

    Best Regards

    Maggie

    • Rajesh1's avatar
      Rajesh1
      Icon for Helper I rankHelper I
      7 years ago

      hi v-juanli-msft ,

       

      Thanks for your reply.

       

      I have been unable to work out a way to create such a column. 
      Tried Current_Column_Value = EARLIER(Column Value) in a calculate + filter.  I am yet to find a way to get this 'key' column created. If I am able to create this column, I can then do a summarizecolumn and get the job done.

      This is where I am stuck at.

      Looking forward to some ideas to achieve this.

       

      Thanks again,

      Rajesh