Calculate date when trendline of resolved issues equals current created issues

Hello, I used LinregPoint to create a resolved trendline for issues. What I’d like to do is determine the crossover date of when the current trendline for resolved issues hits the current cumulative issues created. So if we had no new created issues I could see when the current open issues would be done based on the trendline.

If you can see the chart I would like to put a vertical line where the trendline equals cumulative issues created

Thanks

Take a look at this example: https://eazybi.com/accounts/1000/cubes/Issues/reports/139130-open-issues-trend

There is an “Open Issues until today” measure that you may have to modify to meet your requirement. You then click ton the legend line and change the type to “vertical line”.

Thanks. In your example you have Milestone End Date calculated off of ‘report end date’. What I’d like to do instead is calculate ‘report end date’ based on the date in the future when ‘cumulative resolved issues’ is >= to ‘cumulative created issues’ for future projections. i.e use the slope of LinRegPoint and calculate the date in future when we have as many resolved issues as created issues.

Thanks and Regards,

Matt

Any thoughts on this? Thanks

@matnu1

try something like this:
image

Then change the type to “Vertical Line”

Hi,
I guess @fabian.lim gives the right direction for the solution. A minor remark is that exact equality might not be possible (linear regression gives decimal values), and the CASE condition might never be true. I tried a following version of the same idea:

CASE WHEN
  [Measures].[Trend Cum issues resolved]>
  [Measures].[Cumulative issues resolved]
AND
  ([Measures].[Trend Cum issues resolved],
   [Time.Weekly].CurrentHierarchyMember.PrevMember)<
    [Measures].[Cumulative issues resolved]
THEN
  "End date"
END

Kindly,
Janis, eazyBI support

1 Like

Thanks. I’ll give it a try.