Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
Hi,
I'm trying to implement a Rule based change to Control limits.
Although, the rules are working individually, the rule violation is being triggered based on overall trend whereas it needs to reset the Control limit based on the first Rule violation and the next trigger should be reset from the date (Month) of first violation occurance.
Essentially, the value of changed Control limit should remain constant and next rule should trigger a response only after it crosses the threshold both in terms of time (after first occurance) and the actual rule violation (based on the first rule being violated).
Example: Rule2 = Two out of 3 consecutive values above or below 2 STDEV. So, if the first violation of this rule occured in Aug'22, the next set of rules should be checked from Aug'22 onwards which is at least 3 months after the first violation (since it's Monthly data and the next trigger can only happen when 2 out of 3 values cross the threshold after Aug'22).
Sample data based on current state (in black) and required output in Green (in screenshot below):
Rules:
Unfortunately, I don't have a sample raw data file to share because of data sensitivity.
Unfortunately, I don't have a sample raw data file to share because of data sensitivity.
How about creating a fake data set that demonstrates the issue? This I believe you can do? With today's tools everything is possible and very often very simple and quick. You just have to put in some effort... right?
User | Count |
---|---|
47 | |
26 | |
21 | |
17 | |
15 |
User | Count |
---|---|
53 | |
34 | |
17 | |
17 | |
15 |