Analyze the hourly variation of cases belonging to different groups

[Ley]

Hi,

I want to analize hourly increase of cases (tickets) belonging to different clusters or groups. I need some ideas about what DAX function could I use in order to calculate hourly or daily variation of tickets for different clusters.

Also, It would be helpful any idea about how could show this data to easily detect when the number of cases belonging to a cluster increase drastically. 

Thanks

[v-juanli-msft]

Hi @Ley 

Is this problem sloved? 

If it is sloved, could you kindly accept it as a solution to close this case and help the other members find it more quickly?

If not, please feel free to let me know.

Best Regards

Maggie

[v-juanli-msft]

Hi @Ley 

With this information I need to calculate a function to determine the porcentaje of increase or decrease in the number of tickets for each group. I would need somthing similar to this "for the cluster 7 the increase of tickets number is 20%".

Please provide day or time window period so i can give a proper answer.

"for the cluster 7 the increase of tickets number is 20%"-> 

as i know, this can be achieved by Q&A visual, please complete this question with a certain date or time filter,

for example, for the cluster 7 the increase of tickets  number in 2020/1/11  is ****.

Best Regards

Maggie

[v-juanli-msft]

Hi @Ley 

Assume you have tables as below:

1.

Add a new table

datetime table = CROSSJOIN(CALENDARAUTO(),ADDCOLUMNS(GENERATESERIES(0,23,1),"time",[Value]&":00:00"))

add columns in this table

date/time = [Date]&" "&[time]

convert this column into "Date/Time" type.

2. create columns in your table

date_hour = [date/time].[Date]&" "&HOUR([date/time])&":00:00"

convert this column into "date/time" type

3. create a relationship between two tables above

4. then you can create a measure and use line chart or waterfall chart

DISCOUNT = DISTINCTCOUNT('Table 3'[ticket id])

Best Regards
Maggie
Community Support Team _ Maggie Li
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[Ley]

Hi Maggie,

Thanks for your reply. Is there any way to compare all clusters at once? I mean, your solution is useful if I have a few clusters, but If I would have more than 100 clusters and I can not filter one by one, is there an alternative (table) to show variation for all?.

[v-juanli-msft]

Hi @Ley 

I can count the tickets per day and per day&&hour for every cluster as displayed in a matrix below:

calculate the incremental pecentage per day for every cluster

Best Regards
Maggie
Community Support Team _ Maggie Li
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[amitchandak]

@Ley ,

Can you share sample data and sample output in a table format?

You can create an hourly bucket and use that in measure find the last hourly bucket, Depend on data

[Ley]

Hi 

I attach more information:

1. I have created and linked a calendar table

2. An example of ticket information is the following:

Ticket	cluster	date
A022	1	01/01/2010 13:30
A023	1	01/01/2010 13:30
A024	4	01/01/2010 13:30
A025	4	04/01/2010 13:30
A026	6	05/01/2010 13:30
A027	7	06/01/2010 13:30
A028	8	01/01/2010 13:30
A029	9	08/01/2010 13:30
A030	2	01/01/2010 13:30
A031	6	01/01/2010 13:30
A032	7	11/01/2010 13:30
A033	7	01/01/2010 13:30
A034	7	01/01/2010 13:30
A035	7	01/01/2010 13:30

With this information I need to calculate a function to determine the porcentaje of increase or decrease in the number of tickets for each group. I would need somthing similar to this "for the cluster 7 the increase of tickets number is 20%".

I need hourly variation as weel as variation between days.