Ruminating on DMN - a new modeling notation for business rules

We all know how the BPMN standard helped in interoperability of business process definitions across lines of business and also across organizations. But there was one element missing in the BPMN standard - i.e. the ability to model business rules/decisions in a standard way so that they can be interoperable.

Most of the current business rule engines (aka BRMS) follow a proprietary standard for modeling rules and migration from one BRMS suite to another was usually painful. Hence, we are pretty excited about DMN (Decision Model and Notation), a standard from OMG that can be considered complimentary to BPMN.

Many Rule Modeling tools such as IBM Decision Composer, OpenRules and Drools already support the DMN standard. Part of the DMN standard is also a Friendly Enough Expression Language (FEEL). FEEL defines a syntax for embedding expressions in the rule.
An excellent tutorial about DMN using decision tables can be found here - https://camunda.org/dmn/tutorial/. A good video tutorial on IBM decision composer is here - https://www.youtube.com/watch?v=hG2rGDhowcU

It is important to understand the difference between a decision modeling tool and a decision execution engine. A decision modeling tool would give a GUI to define the decision model using the DMN standard. This decision model can be exported as a *.dmn file (which is in XML format - example here). The decision execution engine actually is the runtime to execute the model. Most of the existing rule engines have extended their support to run rules defined in DMN model. For example, the new IBM Decision Composer is a web-based rule modeling tool, but the modeled rules can be exported to run in the existing ODM engines.

So in theory, the DMN model created by one tool can be used to execute the model in another tool . Some folks have tested this and noted down the challenges in this whitepaper - The Effectiveness of DMN Portability

Ruminating on Decision Trees

Decision trees are tree-like structures that can be used for decision making, classification of data, etc.
The following simple example (on the IBM SPSS Modeler Infocenter Site) shows a decision tree for making a car purchase.

Another example of a decision tree that can be used for classification is shown below. These diagrams are taken from the article available at - www.cse.msu.edu/~cse802/DecisionTrees.pdf‎

Any tree with a branching factor of 2 (only 2 leafs) is called as a "binary decision tree". Any tree with a variety of branching factors can be represented in an equivalent binary tree. For e.g. the below binary tree will evaluate to the same result as the first tree.

It is easy to see that such decision tree models can help us in segmentation. For e.g. segmentation of patients into high-risk and low-risk categories; high-risk credit vs. low risk credit; etc.
An excellent whitepaper on Decision Trees by SAS is available here.

Decision trees can also be used in predictive modeling - this is known as Decision Tree Learning of Decision Tree Induction. Other names for such tree models are classification trees or regression trees; aka Classification And Regression Tree (CART).
Essentially "Decision Tree Learning" is a data mining technique using which a decision tree is constructed by slicing and dicing the data using statistical algorithms. Decision trees are produced by algorithms that identify various ways of splitting a data set into branch-like segments.
For e.g. On Wikipedia, there is a good example of a decision tree that was constructed by looking at the historic data of titanic survivors.

Decision Tree constructed through Data Mining of Titanic passengers.

Once such a decision tree model has been created, it can be exported as a standard PMML file. This PMML file can then be used in a real time scoring engine such as JPMML.

There is another open source project called as 'OpenScoring' that uses JPMML behind the scenes and provides us with a REST API to score data against our model. A simple example (with probability prediction mode) for identifying a flower based on attributes is illustrated here: https://github.com/jpmml/openscoring 

Decision Trees can also be modeled in Rule Engines. IBM iLog BRMS suite (WODM) supports the modeling of rules as a Decision Tree. 

When to use a Rule Engine?

The following article on JessRules.com is a good read before we jump on using a RuleEngine for each and every problem.

http://www.jessrules.com/guidelines.shtml

Earlier, I had also written another blog-post that lists down the simple steps one should take to understand what kind of data (logic) should be put in a rule engine. 

What rules to put in a BRE?

Most organizations today understand the value of externalizing rules from application code by using a Business Rules Engine. The agility and flexibility derived from BREs allows for quicker roll out of new business rules and better business IT alignment.

But how do we determine what business rules should go into the BRE? Not all business logic needs to go into a BRE. The following methodology can be used to arrive at a good set of externalized business rules.

• Prepare a set of variable data/rules for the application. Ask questions such as: What changes? Frequency of change? Who/What triggers the change?

• Assign importance of agility to each variable data. What changes are time-critical? Business Impact if a business rule is not changed in time? Business benefit if we can change the variable data quickly.

• Check if business users can make changes to the variable data. For e.g. if-else conditions can be handled by a business analyst, but change in complex statistical formulas may not be easy.

If we go through the above thought process, we can arrive at a logical set of variable data and rules that should be externalized by the system.