When applying this to my situation, what should I do first?
I am confused with how do this this as I am able to apply INDEX() and MAX() statements to get identify how things rank in my list, but I cannot find a way to filter / sort them.
Are you able to help me further Bharat?
Is it something like this? If yes, then you just need to put the index into filter and select 1. I attached the workbook for your reference.
Hope this helps
Kind of. This is the stage I have been able to get to by myself, but this only seems to be able to sort by 1 aspect? So this is why I am stuck as I don't know if there is a workaround...
I need to do it in such a way that it matches the spreadsheet (at least in terms of the sorting aspect) so that D-UNTH registers as the worst performing Registration. D-UNTF and D-UNTB (7th and 8th place) would be level on 247 occurrences for the Count of Reg column, but D-UNTF would come first with 19192 total minutes as opposed to 16140 minutes for D-UNTB.
Here is a screenshot of the Excel Sorting Wizard which displays how I set up the spreadsheet called MatrixIssue. This is the order I would like to replicate in Tableau.
I made something using rank -a calculation to pick the Reg with max count of delays and another to pick the Reg with more total delay where there are multiple reg with same count ( Ex: for rank 7)
A final calculation will pick up which of these to use. I haven't had time to test it.
Attached workbook. Might be useful as a starting point.
I was trying to highlight the top reg with max number of delays
What you have created is pretty much what I am wanting to identify/highlight Jyothisree Rayagiri, however the way I would like to visualise it is just a single cell saying the aircraft name and then a shape object (which will be a picture of an aircraft underneath). I think I will be able to do that myself by recreating your visualisation first and then changing the layout. The main thing I was needing was the sorting in the right order etc.
Relating to the two calculated fields that you had: MaxReg_ByCount and MaxReg_ByTotalDelay... The data source I am using (in my confidential database rather than the spreadsheet I posted before) for this is an Access Database and it is giving me an error when I try using the same formula saying Level of Detail expressions are not supported by the federated data source.
How do I go about resolving this?
Is there a way to change the expression so that I don't have to extract the data? I'd like to know about both methods depending on my needs etc.
For example I have seen certain people put up posts saying that they are a stuck because they need to use LOD expressions, but also need to connect to a live data source.
Is there a way I can tweak these LODs?
After looking into the visualisation a bit more last night Jyothisree Rayagiri, I saw that D-UNTB and D-UNTF were the "wrong" way round as the REGs were being sorted by Count of REG rather than Count of REG and then Sum of Total Minutes.
The bars were not sorted first by count and then by total delay like we can in excel.
They are sorted by count . I have written different calculations to pick up the REG with higher total delay in case we have more than one REG with same rank for count.
Need to combine both the calculations to pick up a single REG as the final one we need to highlight.
Since we do not have 2 REG at top rank , I wasn't able to test it. Will try to update later today.
One more way I can think of will require duplicating the dataset - if it is feasible with the volume of your data.