Filtering shared date table / multiple date dimensions

I'm building a new dataset and struggling with which date method to use.

I want to be able to use slicers to filter any of the three dates that will be used in the report, and I have a few date groupings that are specific to the separate date types (manufacture date, Availability date, Expiry Date).
This leads me to think separate date tables may be the simplest method.

However, it occurred to me that this would make it challenging (impossible?) to create a view of a measure over different date types (example below, because my words are not working well)

The only method i can think of to achieve both outcomes is to create the seperate date dimensions and then have a master date table connecting all three - snowflake style. Is there something i'm missing here? is there a simpler/better way?