In the last post I mentioned that you probably had lots of data that you might not even know about all over the place just waiting to be collected and put into your historian. I suggested that if you collected that data you could put the data to work and would probably find out that the data was actually very valuable and quite worth it to have collected. In this post, I’d like to suggest another idea to get more out of your historian.
You probably have a data warehouse of some kind somewhere. It’s probably part of a business intelligence initiative and it’s almost certainly owned by the IT department. And, you may not even know it’s there and might never have even used it. And, it probably has no manufacturing data in it and almost certainly has no data from the historian in it.
But, it’s probably there. And, there’s probably a lot of people that use it. And, if they knew about the data you have in the historian, they would want some of that data in the data warehouse like right now.
But, this is a little more complicated that just running a cable, because the data warehouse can’t handle the data the way a historian can; at least not the data volume anyway. So, one thing you’re probably going to have to do is summarize the data so there’s less total volume of data before you put it into the data warehouse.
And, you need to make sure you get the right data. The amps and rpm’s off motor XYZ-1234 are probably pretty important to you, but might not be as important to some other people. So, sifting through the data to find the right data is very important. What you’re looking for is data that correlates into the business KPI’s – things like productivity, yields, throughput, and so on.
You probably have that data, and if everyone knew you had it, they would want it. You just have to sort it out and get it usable. And, that’s usable for them, not for you. (It’s already usable for you since it’s in the historian.)
And, you’re going to need to make sure the data is keyed to the right dimensions in the data warehouse. That might even be a little tougher but still doable. The dimensions in the data warehouse are just keys or access paths to the data and you have to make sure the data you have is lined up correctly with the dimensions. (You know, calling it data from machine X-477-877-AB probably won’t help since no one knows what that means.)
But, once you get the right data summarized and keyed correctly, it can fit nicely in the data warehouse. And, that then gives everybody access to a lot more data than they’ve ever had before. And, it gives everyone a new capability that they’ve never had before.
It may take a little work, but getting the data together, correlating the data together, makes all the data far more powerful. The historian has lots of data. The data warehouse has lots of data. Getting the two together only makes sense. And, that gives you and your colleagues a lot of new information that they didn’t have before. Information that should be very valuable to running the business and making it even that much better.