How does Control Chart calculates which Issues should be displayed

[Vitalii Kryzhalko]

The question is - how does Control Chart works in details.

Preconditions - for instance we have Kanban dashboard with 3 columns - Open, In Progress, In Testing.
we have following configuration for Control Chart

Timeframe: June 1 till June 7
Columns: Open
Filters: NONE

Let's say we have following issues (tickets):
#1 Created on May 1st (appeared in Open column), moved to "In progress" on June 2nd, resolved on June 6

#2 Created on May 1st (appeared in Open column), moved to "In progress" on June 2nd, resolved on June 10

#3 Created on June 2nd (appeared in Open column), moved to "In progress" on June 2nd, resolved on June 6

#4 Created on June 2nd (appeared in Open column), moved to "In progress" on June 2nd, resolved on June 10

Which of this Issues will be displayed on Control Chart?
Which of this Issues will be included in calculation for items above the chart: average, median, min time, max time, issues ?
What changes if I add one more column to the Chart configuration (Columns: Open, In progress) ?

[Danut M _StonikByte_]

Hi @Vitalii Kryzhalko ,

You could try our app Great Gadgets for Jira Server or Great Gadgets for Jira Cloud.

One of the 10 dashboard gadgets included is the Control Chart gadget, which is based on a Jira filter and is fully customizable. Optionally, it can also display a data table showing where the data comes from.

See https://bitbucket.org/StonikByte/great-gadgets-add-on/wiki/Home#!control-chart-gadget

Danut

[Alexander Minin]

Hi!

I'm also desperately trying to find an answer to this question.

And it seems to me that the statement "The timeframe is for resolution date." is not correct.

My Control chart, for instance, I have two major concerns about:

1. It contains issues which are not resolved yet.
2. It does not contain issues which are supposed to be there.
   Example.
   Timeframe: 3 months
   Columns: New, In progress, Done.
   Filters: NONE
   Issue has been created 4 months ago but moved to 'In progress' 1 month ago. It is not shown on my chart.

Any ideas?

[Randy]

If I understand you correctly, it's working as intended.  The control chart is a lagging indicator and the datapoints represent resolved issues. 

If your issue is still in progress, it wont show in the chart.

[Dan Driscoll]

Does it include unresolved issues that transitioned into and back out of your final column status Done?  (i.e. re-opened)

I'm wrestling with understanding how this all works as well.  Clearly the columns included in your overall board configuration determine the entirety of the journey.    And when you refine the reports (control chart, CFD) to a subset of columns, you get more issues that travelled through any of those filtered out statuses.  

Also, if you create a board that is a subset of your overall board (e.g. a front end board for refinement alone - Open->Scoping->Refinement->To Do) it results in a completely different number of items (since To Do is its final status, determining its lagging indicator) vs any larger board's charts with different states and final lagging indicator state.

[Randy]

it actually doesn't have much to do with status.  It key's off the resolution field and includes only issues with a resolution.

If you are setting a resolution when the issue moves to status Done but do not clear the resolution when the issue reopens and moves back to an earlier status in your workflow, it will still be included in the control chart.

[Dan Driscoll]

[editing a previous reply]

I was inaccurately under the impression that the Control Chart as a lagging indicator was populated by default with issues with final Resolutions, and was wondering why work-in-progress was included.   After drilling into the Tips and Examples on the documentation page, I found that one must configure quick-filters to only include issues that are in the desired final Done/Done status/resolutions. 

Thanks for your help!

https://confluence.atlassian.com/jirasoftwarecloud/control-chart-777002660.html

[Randy]

The timeframe is for resolution date.  So only 1, 2, 3 would display and included in the calculation.

You have to select at least one column and the time will be calculated only based on the time in the columns selected for the issues in your dataset.  If you add columns, the times will be based on all included columns.

[Vitalii Kryzhalko]

Tank you for you response.
One clarification - Why #2 is included if resolved on June 10 and Timeframe is June 1 till June 7? Am I correct by saying that only #1 and #3 will be included?

[Randy]

You're right.  It should be ony #1 and #3.