A recent new twitter follower had an interesting post on his blog back in 2010 – The 4Ds “Detect, Derive, Decide and Do. I liked the description of this pattern and as it is a common one for Decision Management Systems I thought I would make a couple of quick comments:
- The Derive piece of this pattern might be driven entirely by the analysis of events as Jeff describes but it is also a potential place for Decision Management, especially when analytics can play a role. This step can involve determinations that could get quite complex and that could use predictive analytics or rules derived using data mining. Deriving patterns from data is something at which analytic models excel.
- The key role for Decision Management and Decision Management Systems, of course, is in the Decide step. Not only are business rules management systems common here to manage the logic of a decision but the use of predictive analytics to inform and improve that decision is also a powerful use case. The key is to focus on this as a decision and to model and manage it as such, rather than being immediately distracted by the rules or IF…THEN logic that drives it. I wrote about the differences between decisions, decision services and rules before.
- The overall pattern is one that is driving a number of Business Rules Management Systems to support event-handling rules. This allows the Derive and Decide steps to be managed together in a common environment.
Event-based systems are an increasingly common context for Decision Management Systems. If you want to know more about the technologies involved, why not download or read online our report on Decision Management Systems platform technologies.