Our published articles relating to Business Rules
Patent: Driving an interactive decision service from a forward-chaining rule engine
Raj Rao
November 2013
In several areas, such as a self-service kiosk or an interactive voice response system, a computer application needs to involve a human user in an interactive dialog before arriving at a decision. Such computer applications may be termed interactive decision services. These computer applications need to determine which question or set of questions to ask next, given a partial set of input data.
The patent is for a novel technique for generating questions by a regular forward chaining rule engine without the need for explicit question generation rules, and offering rich conflict resolution strategies. This technique generates the next question to ask without sacrificing the natural representation of domain specific business rules.
Using Hosted Transparent Decision Service Interface (HTDS) in IBM Operational Decision Manager
Raj Rao and Matt Voss
IBM DeveloperWorks, December 2013
Operational Decision Manager (ODM), IBM's business rule management system (BRMS) product, offers an easy way to expose rulesets as web services through Hosted Transparent Decision Services (HTDS). This article briefly introduces HTDS and discusses the approaches to building an HTDS client and using the RESTful interface that is new to IBM ODM V8.5. The primary focus of the article is potential issues and pitfalls in using HTDS, mostly related to controlling its service interface, and some practical, hands-on approaches to resolve these issues.
Using dynamic domains for flexible rule authoring in WebSphere Operational Decision Management
Raj Rao and Catherine Rivi
IBM DeveloperWorks, Feb 2013
Ensuring rule accuracy is an important part of rule development in WebSphere® Operational Decision Management. This article demonstrates how a dynamic domain populated from a Microsoft® Excel® file can be used to reduce inaccuracies when authoring rules. Using domains also eliminates dependencies on IT support for code changes because business users can extend the set of values in the domain by simple edits to the spreadsheet in the Decision Center. Further, because adding a domain value does not result in changes to service contracts, redeployment of rule services is not necessary.
The rogue business object model
Raj Rao
IBM DeveloperWorks, Feb 2012
When is a BOM a time bomb? This article describes pitfalls and anti-patterns to avoid when creating a BOM.
The business object model (BOM) is a fundamental component that provides the business vocabulary used by business policy managers to define rules built using WebSphere® ILOG® JRules or WebSphere Operational Decision Manager V7.5. This article describes the frequently observed anti-patterns and missteps in the creation of a BOM and recommends techniques for avoiding them.
Accessing external data in a rules application
Raj Rao
IBM DeveloperWorks, Feb 2013
Although it is generally recommended that external application data not be accessed from a rule application deployed to a WebSphere® rule execution component, there are certain situations that warrant it. This article describes these situations, and compares options for handling them. An example scenario offers practical, hands-on techniques.
Using business rules in the cloud to solve Sudoku, Part 1: Implementing the rules application
Raj Rao
IBM DeveloperWorks, June 2012
This series presents a hybrid approach that combines business rules and heuristic depth-first search to solve Sudoku puzzles. This approach emulates the methodology humans use to solve Sudoku. Rules built using WebSphere® Operational Decision Management apply forward-chaining logic; when this fails to find the solution, the heuristic search with backtracking emulates the trial-and-error approach.
Using business rules in the cloud to solve Sudoku, Part 2: Deploying the rules application to the cloud
Raj Rao
IBM DeveloperWorks, Sept 2012
In Part 1, I presented a hybrid approach to solving Sudoku puzzles - an approach that combines business rules built using WebSphere Operational Decision Management (ODM) with a heuristic depth-first search algorithm. In this column, I'll focus on executing the rule application on the cloud. Can a web application leveraging a rule engine be deployed to a public cloud? If so, which rule integration pattern would we use? Which ODM components would we deploy? What level of execution performance can we expect? What challenges await us? These are the questions that this column seeks to answer for two widely available cloud offerings: Google App Engine™ and Amazon Web Services™.
Reusing rules within a ruleset to avoid creating redundant business rule variations
Raj Rao
IBM DeveloperWorks, Sept 2011
One of the advantages of using BRMS is that rulesets can be reused across different client applications – but this article is not about that. Instead, this article focuses on those situations where a naïve practitioner might clone business rules within a ruleset to apply the same business policy to disparate business entities or different contexts. This article presents such a scenario and outlines different techniques for reusing the rules across different contexts, thereby avoiding duplication and improving maintainability and performance.
Develop decision services, Part 1: A smarter city case study
Raj Rao and Sandeep Desai
IBM DeveloperWorks, June 2011
In today's ever-changing business environment, there is a need for systems to be smarter to reduce the overall cost and complexity and to help businesses react quickly to changing environments. To do this, a smarter system needs an effective decision subsystem. In this three-part article series, we use a smarter city use case to guide architects and developers in selecting and implementing a decision subsystem. We walk you through a complete application development lifecycle of a business rule management system (BRMS). We discuss the BRMS architecture, rules discovery process, developing the rule application, and rule maintenance. The smarter city use case involves implementing a decision subsystem for intelligent coordination among multiple city departments during severe weather emergencies. The decision subsystem is based on IBM WebSphere® ILOG® JRules to automate the decision-making process.
Develop decision services, Part 2: Rules development process
Raj Rao and Sandeep Desai
IBM DeveloperWorks, June 2011
Part 1 introduced you to the smarter city scenario and the use case requirements. We described the role of the decision subsystem and the rationale for selecting IBM WebSphere® ILOG® JRules as a business rule management system (BRMS) to implement the decision subsystem. Part 2 walks you through the how-to of the rules development process.
Develop decision services, Part 3: Deployment and beyond
Raj Rao and Sandeep Desai
IBM DeveloperWorks, Aug 2011
Part 1 of this series introduced you to the smarter city scenario and the use case requirements. We described the role of the decision subsystem and the rationale for selecting IBM WebSphere® ILOG® JRules as a business rule management system (BRMS) to implement the decision subsystem. Part 2 covered the process of rules development. Now, learn how to deploy the initial rule artifacts and enable non-technical business users to continue rule development and testing.
Creating intermediate facts in WebSphere ILOG JRules using synthetic objects
Raj Rao
IBM DeveloperWorks, Nov 2010
This article uses a scenario to introduce the intermediate fact pattern in business rule processing and describes a technique in IBM® WebSphere® ILOG® JRules to create intermediate facts using synthetic objects in the business object model.