Getting the Where There

By Hal Reid

Hyperion Solutions, a Business Intelligence (BI) company, hosted a WebEx presentation last Thursday (May 26th).The presenters included Dan Everett with Hyperion Solutions, Michael Gonzales of HandsOn-BI and Peter Rossiter of Integeo, a map intelligence company headquartered in Australia.

The presentation focused on three areas.
  1. Hyperion "dashboards"
  2. Adding the concept of where to BI's standard list of who, what and when, to create pervasive BI
  3. The concept of integrating map intelligence with BI for a bi-directional data flow from BI to maps and back again.
This presentation was important as it offered a view into the future of management of 21st century companies that use advanced BI tools to seamlessly bring the spatial element into the the process of managing a business visually.

BI has included the use of "dashboards" (sometimes called "executive information systems") for a long time.Their purpose is to graphically show what is happening within a business in terms of sales mix, performance, inventory management, etc.What has not been shown is where it is all happening.

During the presentation, Hyperion showed several of its customers' dashboards.An example of incorporating mapping into a BI dashboard is provided below.

This image shows the relationship between sales and number of units and which part of the world those sales levels are occurring.Source: Hyperion.Used with permission.(Click for larger image)

The people at Hyperion see the potential with maps, which is why this presentation included Gonzales and Rossiter, both of whom bring in knowledge of spatial data and analysis.

Adding the Where
Gonzales is a champion of including "the where" (e.g.Business GIS functionality) into BI.He believes that in order to be really useful and pervasive in an organization, BI has to include location intelligence.His logic is that everything happens "somewhere," and that "somewhere" has to be well understood.

He gave the example that knowing that you have in inventory stored in Memphis just isn't good enough.You needed to know which warehouse, where in that warehouse and the exact quantities, and see it all on a map.Not only do people at headquarters need to know where it is, the people at the warehouse need to know too.The obvious cost savings to knowing "the where" in this case, was not having to look for lost inventory.

Gonzalez's objective is to create rich databases for BI.He argues that the richer the database the more pervasive it can be.Database richness requires location intelligence, which he describes as:
  1. Spatial - expanding "the who" to enable the analysis of "the where"
  2. Visualization - identifying complex data interrelationships of items, or in other words, the visualization of spatial relationships.
Both of these points support his view that humans needed to see logic visually.Gonzalez said, "When 'the where' is added, all downstream data analysis is enhanced." Without it, he feels you have impaired intelligence, like knowing what products were purchased, but not knowing where the customers who bought those products live.

Source: Mike Gonzalez.Used with permission.(Click for larger image)

The information enhancement image above shows the processing of adding "the where" to "the who," e.g., geocoding a customer's address, storing the geocode in the database as a spatial object, showing those customer(s) on a map, and then graphing the concentrations by individual sales on the BI dashboard.

Map Intelligence
Rossiter expanded on the dashboard theme by showing an integration of the Integeo products with the Hyperion dashboards.The key feature was the bi-directional data flow between the BI system and the mapping system.This was useful because as data moved into a map, they could be filtered by selection sets (buffers, thematics, etc.) and then brought back into the traditional dashboard to enhance it using the dashboards graphical representations.Below is an example.The first image shows the dashboard, while the second shows that data on a map.Using the map, the data is filtered and moved back to the dashboard.

Dashboard Data

Source: Peter Rossiter and Integeo.Used with permission.(Click for larger image)

Dashboard Data Driving a Map

This data was focused by region, then by point data per collision down to the individual events.Using those selection sets the data was moved back to the dashboard.Source: Peter Rossiter and Integeo.Used with permission.(Click for larger image)

Map Data Driving the Dashboard

Source: Peter Rossiter and Integeo.Used with permission.(Click for larger image)

Each step in the analysis could be bookmarked for a historical record of how the analysis was done, which is a useful feature.Each bookmark could be seen as a live "mini-map" as the analysis steps continued. Rossiter explained that this process can be independent of the map engine used.However, in this case there is a custom interface to do the analysis and create the bookmarks.

Interesting tidbits
When the audience attending the presentation was asked how many of them are now using mapping software, 55% reported they were not.This is a clear indicator of why "the where isn't there yet."

Two questions from the audience did address some interesting spatial and temporal issues.The first was about drive times and ring studies and calculating them on the fly.This functionality is not supported in the current releases of Hyperion dashboards or Integeo's Map Intelligence product.The Hyperion and Integeo representatives said they were aware of the value of these features and would add them to future releases.

The second question was about data animation, or the display of data over time.The reply: also not currently supported.Yet, this potential user had grasped the need to not only include "the where," but to allow "the when" to be visualized not only in the past and present, but modeled for the future.

Also missing is a 3D workspace.3D would allow for data relationships to be seen and "the where" extended to virtual geography.Certainly the data animation issue is key to understanding what happened when.

The important thing is that location intelligence is a part of BI and fortunately, there are those who understand that neither BI nor LI can stand-alone and yet achieve Gonzalez's goal of pervasive business intelligence.

Published Tuesday, June 7th, 2005

Written by Hal Reid

If you liked this article subscribe to our bimonthly newsletter...stay informed on the latest geospatial technology

Sign up

© 2017 Directions Media. All Rights Reserved.