We created this data source “Aviation Statistics” in SharePoint, and to begin my Power View report I just need to click on that link. What that will do is to launch the Silverlight based Power View product. This is a browsing, exploration and design environment all rolled into one.
I can use this as an ad-hoc query tool, and I can use it to put together visuals that I’m going to deliver to others.
The design paradigm is that I have a canvas (surface) that I can drop things onto and organize, and it’s all very easy to use with a “point-and-click” interface. There’s not much to configure as we go through this exercise.
Our data model is a database of all airline flights from 1988 until late 2011. I’m going to work with one measure from the Flight table. I have many measures but I’m only going to work with the Flight Count. We’ll look at how many flights we’ve had. Over the 1988-2011 time period we’ve had about 140 million or so.
I’m going to add a title on top, which will be Originating Flights by Major US Airlines. This report can be resized to fill the entire screen. I know it’s going to be a matrix, because I’m going to a cross-tab, which is one of the most interesting ways to look at things.
My setup panel changed a little bit when I chose cross-tab, so I can see my column groups and row groups. That allows me to structure my cross-tab.
In my title I wrote “by Major US Airlines”. In the data model I have a dimension for Carrier, so I’m going to take advantage of something Power View has, which is a graphic column type. As I drop that in, I’ll get a list of airlines, but in an Icon view. This is setup in the data model to display that this field is a graphic.
I did say I really just wanted Major airlines, and what I’ll do to accomplish that is to click on the Filters area, and add a filter on the “Major” airline column. That limits the view to only the set of Major airlines.
The next thing I want to add is the originating flight airport. I’ll just drag the airport code and drop it on the column group. So now across the columns are all the flights by each airline over a broad period of time.
Actually I’m not that interested in every hub, so let me give the user the ability to choose whether they want only large hubs.
You notice that as I drag data elements into the view the connections between views are being made automatically. I don’t have to “connect the dots” and tell Power View how elements are related to one another. I just drag items in and the connections are automatically made.
That limits my airports to on the large hubs (as classified by the FAA).
To make this a little more interesting I’m going to bring in the aircraft model that this airline was using. Currently I have AirTrans selected, so I see AirTrans flights by Major Hub.
I scroll down and find the Aircraft model that flew on that flight and drag that to rows I get an interesting insight: which airplane was AirTran flying for each of those flights? I’d like to see the Airplane that flew the most flights on the top, so if I just click this column header, I’ll get a sort by that column. I want to go with the most number of flights on the top.
What that tells me is that AirTran, over the period of 1988-2011 flew more flights in the 717 than any other airplane. Maybe I really don’t want to look at so much history. If I don’t want to I can filter the report by the date. So I’ll find the Year. The Year gives me this interesting slider, so I can slide to select between two other years, like 2008 – 2012–four years of history.
I’ve put together a pretty interesting view in about four minutes!
The text is a little small at this resolution, but I can improve this by going to the Home ribbon and clicking the “Full Screen” button, which will make the view as large as possible.
Now we can use this as a consumer and play around with it. How about Southwest airlines? they have a small number of different airplane types. We can see that an airline that flies a lot of different airplanes, though, is American. What happens if we click on that Airline? We see a lot of different aircraft models.
I can do the same for other air carriers as well, and look at how many different aircraft types are they flying from various terminals over time.
So that’s a very basic Power View report that we implemented in a short period of time, and really while we were exploring the data and looking at it we ended up with a really nice report that we can reuse and distribute to others.
That’s part of the power of this tool versus some others. We’ll look at more details and specific techniques as we go forward with other lessons.