Actually, I think I over simplified my question which is why I am having such a hard time getting my arms around this problem. In my case, the "event type" is actually a state that can change at any given time. (i.e. machine x changed from state 'a' to state 'b')
So I just realized that part of my problem is that I am deduping which eliminates all events except for the final state change. This in combination of the idea that I am not keeping a running tally for each day, I don't see an easy way to track trends.
In the end, what I was hoping to accomplish was show trends in change in state over a period of time. I just need to rethink this a little.
Thanks for the response. Greatly appreciated.
... View more