Intelligent Enterprise

When James Taylor and I wrote Smart Enough Systems: How to Deliver Competitive Advantage by Automating Hidden Decisions we knew it was going to be a lonely outpost for a while. It isn't easy for two guys to insert such a big idea into the collective consciousness. We didn't really have a natural sponsor for the whole picture to walk with us. The business rules community more or less understood it as a Business Rules Management System, which is a key component of the architecture. The predictive analytics crowd was interested as PA was prominent in the architecture, but in reality, very few really grasped EDM in its entirety. It was certainly easy to describe:

If you want to put EDM together, you need to assemble tools from some pretty divergent areas. First, you need decent information management at both ends – at the beginning to investigate things and to create test data for your data mining exercises, and at the far the far end to continuously evaluate the results of your models. Unless your models are perfectly understood and driven by policy or regulation, you need predictive models to help you understand how to implement your decision models and, as I mentioned, BRMS to drive the process. The "process" can't exist on its own, it typically needs to be part of an orchestrated process, there are all of those externalities, too, such as ESB, SOA, etc.

No wonder people didn't just ask us "how much?" It's a lot of work and we assumed it would take quite a while to catch on while we wrote, consulted, taught and speeched (OK, not a word). So I was surprised, even delighted to see that Ventana Research has recently come out with a couple of papers on exactly this topic, EDM.

"Extending BI to Support Operational Decision Management"

"Improving Operational Performance through Decision Management"

Thanks, Ventana.

E-MAIL | FOLLOW US