So a while back, I was sent a link to a pretty amazing operations management dashboard on Esri's arcgis online site. This dashboard provides a fantastic demo of a spatial view into water facilities operations for the city of Naperville.
The dashboard provides all of the amazing things we've come to expect from Esri maps. It's really easy to show and hide operational layers, to selectively view leaks, spills, projects, and so on. You can easily zoom and pan around the screen, see events superimposed on a street map, satellite imagery, or elevation map. You can hover over any point to see more information about it. You can select areas on the map and extract data from it.
All really great stuff.
But here is what would drive me a little nuts about having a dashboard like this if I were a water operations manager. I'd have a lot of what I need, but not the full meal deal. Why? Because as a manager what I manage most is money, time and resources. The better I can optimize those, the better (and more) services I can provide for the residents of Naperville.
All of the features on the map... the spills, the projects, the workorders... have costs that I analyze in a business intelligence system. I likely have reports with lists and charts for workorder costs by subcontractor & workorder class, to figure out who to call for which job. I might also have charts showing trends over time for material costs versus labor costs versus equipment costs, as well as projections for all of these. And so on.
All of these things are keyed in my data warehouse. They have to have a key in our geodatabase as well.
Hey -- why not bring the two together! That way I could select whatever events I want on the map, and see all of the associated costs and projections at the same time!! I could pick 30 workorders, in different areas, all symbolized based on the cost (or the crew, or the workorder class) and see tabulations/charts showing who did them, how much was budget, how much was actually spent. It would be easy to identify outliers and the problem areas from a cost perspective -- totally visually on my map.
So I go to my IT director and ask about the possiblility of integrating this dashboard with my IBM Cognos BI system. Well, Esri has great APIs he says, and so does IBM Cognos he believes, so we can do that. But it'll take some work. There will have to be a requirements gathering phase, then some design and a prototype built. How much budget do we have for this, he asks.
Then he asks: "How is this BI system secured and authenticated?" I'm not totally sure, but I think I heard about an LDAP server somewhere, with reports secured by role, group, and user.
"OK" he says. "That'll take a little investigation as well. I'll get someone on it and we should have an accurate idea of the security integration effort inside a week or two."
The questions keep coming... what's the underlying database system? Is it real time data? Is it on one physical server or deployed across many? Is it relational or OLAP? And so on. By the time we're finished, we could be looking at a 4 to 6 month project, with 2 or 3 dedicated resources and business input/oversight.
Well, that's a little more than I wanted to spend, actually.
Then my trusty assistant reminds me about SpotOn Vantage Maps -- a solution that integrates Esri geodata with IBM Cognos BI data. She's seen some demos that lead her to believe we could integrate our current Esri map services with in-place BI in a matter of a few weeks -- possibly even faster. And the best part, she says, is that it uses in-place resources: the map services we already have, and our IBM Cognos BI report authors. No API coding whatsoever.
That's more like it. Using in-place resources, I get a solution targetted to solve exactly the business problem I am interested in solving. And out of it, I get the best of both worlds. I see and analyze my business costs, budgets, trends, and projections in the context of the physical world in which they happen.
That's why geospatial BI dashboards trump standalone GIS or BI dashboards.
- dk