The world of business rules and business rule management has grown up. It has evolved into the world of business decisions – a much more compelling discipline by which companies can master their business logic.
The idea of business rule management, both alone and within the greater context of enterprise decision management, has been a common mantra within organizations for many years now. In theory, business rule management is the means by which business rules (or logic) is identified and analyzed within current business practices, perhaps mined from existing IT systems, then articulated, challenged, automated and placed under the governance of business SMEs as an independent corporate asset. Although this technique presents a significant improvement on the methods that preceded it, separating a vital element of business intellectual property from its IT systems, unfortunately, in practice, it has been lacking in precision and scalability. Many business rule management methods are available [1], but are either a lacking in formal rigor, are subjective, are proprietary and often not supported by adequate tools.
Recent developments have changed this. And this is good news.
The Good News
The advent of The Decision Model (TDM)[2] and supporting tools [3] introduce an era of new rigor and effectiveness for true enterprise decision management. TDM, by design, deliberately corrects many known flaws of business rule management. It leads to the more mature practice of managing business decisions, not individual business rules.
This column reviews the drawbacks of business rules [4] and explains how business decisions overcome the drawbacks. It then presents the synergy between The Decision Model and the Business Analyst. It ends with the value of today’s decision modeling tools and the important software features to consider.
Drawbacks of Business Rules
Many companies appreciate the value of separating business rules (or business logic) from the source code of systems, from definitions of business processes, and from (potentially stale) requirements documents[5] . For these companies, business rules and business rule management are an important fact of life.
Unfortunately, for all of its well-intentioned goals, business rule management has its drawbacks. As a result, some companies have had bad experiences with business rule management. In fact, the following are five common difficulties with BRMS[6] experienced by business rule management adopters.
Growing Pains: There is difficulty in safely growing rule repositories in excess of 2,000 rules within BRMS products, without descending into chaos. Although BRMS products physically support large rule repositories, they lack a higher level business-oriented structure to coordinate the natural relationships among rules.
Lack of Coordinated Validation: Although many BRMS products can validate individual rules; they frequently cannot validate the mutual consistency of large rule sets used collectively to reach a coordinated conclusion. To fill this need, organizations are often obliged to develop their own mechanism.
IT Focus: Many BRMS products, despite best intentions, were originally developed for the technical user rather than the business user. These products are based on a technical rule language or require that development tools be the means for maintaining business terms, business rules, and rule flow. This technical origination prolongs the business’s dependency on IT for business rule management. Furthermore, the tools frequently divorce rules from their business rationale and process context, making it harder to judge their on-going effectiveness.
Lack of Agility: Partially as a result of the above, BRMS products don’t empower business SMEs and analysts to own the business vocabulary. The business vocabulary is at the very heart of business decisions because it is the set of business words with which business people want to express their rules. As a result, the control remains with IT and this means that IT is involved in every significant change. This is an important barrier to agility and undesirable for many other reasons.
Lack of Business Testing Facilities: A vital part of developing business rules is testing them by executing them. This is an activity in which business SMEs and analysts should have center stage. This refers, not to systems integration testing (which should be controlled by IT). Instead it is the initial unit and regression testing of decisions based on a pure business perspective. Business SMEs should be actively involved in such testing of their decisions. Because testing in many BRMS products is very hard to configure, it requires extensive IT support.
Overcoming These Difficulties with Business Decisions
The Decision Model and tools that support it offer four essential ingredients that ‘change the game’ depicted in Table 1.
| Game-Changing Ingredient |
Description |
| Rigorous Principles |
A set of structural and semantic principles which establish and protect the integrity of a business decision, allowing it to retain coherence and conformance regardless of size |
| Business Governance |
A primary focus on business stewardship of decisions, from specification all the way through testing |
| Connections to Other Artifacts |
A defined and distinct ‘place’ for decision models separated from, but connected to, business processes, organizations, events and data |
| Industry Standard |
A new emerging standard from the OMG called Decision Model and Notation (DMN) to define a common ground by which interested software vendors can provide decision modeling functionality. |
Table 1: Essential Ingredients Offered by The Decision Model and Supporting Tools
These ingredients together provide the following advantages:
Rigorous Structure that Binds Business Rules into a Cohesive Whole. Just as databases provide referential integrity to enforce consistency, decision modeling supplies inferential integrity to keep decisions totally consistent.
Enterprise Scalability. The built-in cohesiveness in decision models radically decreases the entropy associated with large rule sets and facilitates practical management of much larger repositories.
Traceability Between Business Decisions and KPIs[7] .Business decisions are not defined in a vacuum. There is a clear and documented association between business decisions, business processes, business motivation, business organizational structures, business events and business data. Managing connections from business decisions to related policies or regulations allows for easy cross-reference and estimations of future changes. Managing connections from business decisions to expected business impact provides a basis for measuring the effectiveness of the logic behind those business decisions.
Full Life-Cycle Business Governance of Decision Models: TDM prescribes, by definition, true business vocabulary, decision model creation, validation and testing. The rigor of TDM enables automatic or manual creation of test data. Together these facilities offer unprecedented independence on the part of business governance over decision logic with minimal or delayed dependence on IT.
Business Focus Devoid of Technicalities. TDM’s focus on only business-driven vocabulary (and not technical artifacts) sets the stage for sharing decision models across organizational boundaries. This removes the need for specialized technical training and allows more staff to benefit from the explicit statement of company policy as it appears in decision model format.
The Decision Model and the Business Analyst
Teaching business analysts how to harness the Decision Model (TDM) is straightforward conceptually. Business analysts quickly see that TDM renders business logic easier to comprehend, validate, communicate and manage. They understand that it represents logic in its most minimal form – which is why most changes are made to one and only one place.
Business analysts appreciate that TDM, with its structural and declarative principles, introduces a much needed rigor and business-focus into business logic management. Indeed such is the enthusiasm of many business analysts, after their initial ‘ah-ha’ moment, that the technique virtually sells itself.
However, the greatest barrier to decision modeling has been that The Decision Model was ahead of its own software market. In other words, TDM was introduced to the public in 2009 as an innovative intellectual technique which aimed specifically to overcome the drawbacks of other approaches. But, being new, there was no commercial software to support it.
The Value of Today’s Decision ModellingTools
Today, there are three different kinds of options for creating and managing decision models, with more expected in the near future.
Option 1: MS/Office
The simplest (and initially, the least costly) option is to use MS/Visio and MS/Excel templates for creating and updating decision models. This option offers more when these templates are coupled with OpenRules as an execution and testing platform.
However, MS/Visio and MS/Excel have limitations, such as the lack of cohesion and traceability. With great diligence, it can work. In fact, it has worked with large decision models at major corporations and is still superior to previous business rules approaches because of its discipline and focus on the business-oriented audience. However, using MS/Office products as the basis of a decision model repository is manually intensive, error prone, and can easily become undisciplined. There is the additional overhead if an organization deploys to a specific BRMS, requiring translation of the decision models into rules that will execute in the target BRMS platform. So many moving parts and manual procedures stifle agility.
Option 2: Sapiens’ DECISION
A second option is Sapiens’ DECISION product which represents a sophisticated solution. Sapiens DECISION supports decision modeling within the structure of enterprise governance. Enterprise governance includes a federated approach to community and glossary management and customizable governance workflows. It separates glossary governance from decision model governance. It also supports automated validation, test case generation and execution, as well as simplified mechanisms to deploy to specific and multiple BRMS targets.
Option 3: BiZZDesign’s Decision Modeler
The middle option is BiZZDesign's Decision Modeler. It supports decision modeling, glossary, connection to business process models, some validation, test case execution, and automatic deployment to OpenRules. The support of connections between decisions and other architectural artifacts (BPMN processes, UML objects and ArchiMate business motivation models) offers, for the first time, a very significant potential for building decision models firmly rooted within a business and architectural context.
Important Software Features
When considering decision modeling software, the most important features are:
-
Ability to create relationships from decision models to artifacts of other architectural models.
-
Business-driven vocabulary management
-
Business-driven testing and execution of decision models with minimal IT support.
-
Direct deployment to specific target BRMS platforms.
These four features together make a product more than just a competent TDM modeling environment. It opens the door to business SMEs modeling, developing and testing their own, fully integrated, decision models without habitual IT involvement hampering their agility.
Conclusion
The need to cope with ever increasing rates of change in business policies will never go away; in fact it will only increase. Particularly where regulators or legal compliance are concerned, organizations need to involve business SMEs directly in the capture and stewardship of decisions to improve agility and accountability.
The four game-changing ingredients in Table 1allow organizations to provide greatly enhanced safe agility- faster time to market for new decisions and less latency for existing ones. This is strengthened further by allowing businesses to quickly determine which decisions are impacted by an incipient change of business policy – so it becomes possible to estimate accurately and quickly the impact of any proposed changes. Decision modeling tools allow business SME teams to achieve much more on their own. Specifically, business SMEs may create decision models from initial requirements and build a ‘play area’ for testing the logic, including development of a complex network of decisions and all without the constant need for IT support. The process of business and IT cooperation is greatly streamlined as a result. Lastly, decision modeling tools serve as a vehicle for communicating business decisions throughout the enterprise.
An Interesting Point in Time
This is an exciting time to be in decision management. With the combination of the rigor of decision modeling together with current and evolving software support, there is finally a practical way for an organization to control its business decisions and never again lose them to program code. The benefits are more accessible than ever before.
Expect more advances in the products mentioned here as well as appearance of new ones. The competition has only just begun.
Author: Jan Purchase, Founder, Lux Magi
Dr Jan Purchase has been working in investment banking for 17 years, the last 12 of which he has focused exclusively on the use of business decisions, business rules, business rule management systems (BRMS) and business processes. He is a founder of Lux Magi, a company specializing in mentoring the application of BRMS, The Decision Model and BDMS to investment banking. He hosts a blog on these subjects.
[1] Some business rule (and early business decision) management methods include: ABRD which is IBM’s Business Rule Development Methodology, Enterprise Decision Management, and Ross’s Building Business Solutions.
[2] Simply stated, TDM is a rigorous method and framework for expressing business decisions.
[3] Some tools supporting TDM are BiZZDesign's Decision Modeler, Sapiens' DECISION and OpenRules
[4] If you are new to business rules, want to understand their many benefits, or want to know the difference between business rules and business decisions, please follow these links before continuing.
[5] There are many reasons why you should separate business rules from other aspects of processes and systems.
[6] BRMS stands for business rule management system which is specialized software for defining, deploying, executing, and maintaining the decision logic (business rules) used by operational automated systems.
[7] KPI stands for Key Performance Indicator used to evaluate an organization’s success