It’s been about two years since my last Back to Basics blog post, where I detailed what makes a rules engine and Business Rules Management System (BRMS). Since that time, InRule has evolved its offerings, going beyond a BRMS to delivering a solution that helps enterprises both automate and understand decisions – the InRule Decision Platform.
Naturally, this leads me to another Back to Basics post explaining what makes a decision platform.
Let’s start with a general definition for a decision platform: a decision platform is an end-to-end solution that allows users to centrally manage, automate and provide insights into enterprise decisions. So where does the BRMS fit in? A BRMS is one part of a decision platform.
At InRule, we believe that a decision platform must provide five main capabilities:
- Decision Authoring
- Lifecycle Management
- Business Process and Execution
- Monitoring, Analytics and Improvement
In this post, I’ll examine each of these five capabilities and address how the InRule Decision Platform provides these capabilities to our users. Some of the functionality outlined here is available now, while some is visionary and illustrates where we are headed.
At InRule, everything starts with decision authoring. With our author-first approach, we have always been – and remain – committed to making it as easy as possible for all users, technical and non-technical, to create and maintain decision logic. Therefore, a decision platform should give its users multiple ways to author decisions.
As I mentioned in my previous post, InRule provides multiple ways to author decisions. A benefit of using InRule for decision authoring is the ability to establish custom vocabulary that allows users to create rules in language unique to their business. The use of custom vocabulary makes the decision authoring process easier for subject matter experts, simplifying complex expressions by using terminology they understand.
In addition to allowing a user to author a new decision, a decision platform should allow organizations to leverage logic that resides within existing applications, including Excel sheets, XSLT, COBOL, stored procedures, VB, Java or other programming languages.
InRule makes it possible to automate the migration of existing logic, preventing manual re-writing of rules that power a business and extending the shelf life of existing decisions. By centralizing and reusing existing decision logic across applications, InRule’s decision platform becomes an enterprise’s single source of truth. Additionally, by automating rule harvesting, a decision platform provides a path for digital transformation.
Users are encouraged to explore Decision Modeling Notation (DMN) when authoring decisions to gauge whether it meets the needs of their enterprise architecture practice. We believe that DMN is important because it provides architects a way to model data and link requirements to operational decisions. DMN makes decisions more discoverable and sharable, and therefore more collaborative.
Finally, I should mention the role of machine learning in this process. By pairing machine learning with declarative rules, a decision platform allows you to operationalize machine learning by adding a commonsense layer to the process.
Once logic has been migrated or created, and vocabulary and actions have been established, we arrive at the foundation of the decision platform, the Business Rules Management System (BRMS). As previously mentioned, a BRMS is a solution for automating decisions and includes components for authoring, testing, storage and execution.
InRule offers multiple ways to express decision logic and flow. Users can write rules in a business language editor, using the previously mentioned custom vocabulary. Additionally, InRule users can author decisions with an Excel-like syntax editor, decision tables, or via graphical decision mapping.
The decision flow guides users through the creation of a decision and helps them visualize the steps in the process of that decision. Additionally, the decision map provides an understanding of the lifecycle of each decision.
One of the key benefits of the decision map is that it minimizes the learning curve for new users by guiding them through the authoring process and breaking down barriers for getting started. Most importantly, it allows users to focus on overcoming business problems – not learning a new tool.
Once a decision has been written, users must be able to test their logic and verify that it works as expected and achieves the desired outcome. By using the authoring and testing components of the InRule Decision Platform, users can ensure their decisions are working as intended, and if necessary, revise them before pushing to production.
Decision Lifecycle Management
A decision platform needs to provide complete decision lifecycle management, tracking the logic and related modifications. It should also offer controls around who can author, view and execute decisions. An approval workflow makes these automated processes consistent, more efficient, and allows organizations to easily and automatically route tasks and decisions to specific individuals for sign-off.
The storage component of the decision platform acts as a central repository for everything related to an organization’s decisions by housing all rules, vocabulary, entity models and other metadata. Users can check logic out and in and track changes, allowing all team members see what was changed, ensuring the integrity of the decision logic.
As users are working on decisions, they will need to publish changes to different environments, such as development, staging or production. A decision platform allows for this capability in order to make teams more productive and minimize risk.
Finally, users may find themselves working on different workstreams of the same set of decisions. The platform needs to allow these branches to be merged back together at any given point to create a singular set of decisions.
Organizations often deploy a best-of-breed technology stack, implementing multiple technologies to solve a range of problems. A common issue with this approach is that decision logic can become replicated across these disparate systems – and a change to one decision could lead to different outcomes for different systems. Decision logic and outcomes need to be consistent and available across all applications and throughout the organization.
Most organizations run their business using CRM, ERP, BPM and other technologies. InRule acts as the “digital mind” of an organization by integrating with these technologies allowing them to better execute the intent of the business through automated and augmented decisions.
Business Process and Execution
InRule makes it possible to publish one or more processes to execute your decisions. Process inputs and outputs can be defined upfront and fed into the rules engine for execution. Depending on how the decision is defined, the decision process can execute any number of logic steps and actions to arrive at a decision.
The InRule Decision Platform provides the widest range of models for rule execution for on-premises, self-hosted SaaS environments, or vendor-hosted SaaS environments. We support many self-hosted cloud environments such as Amazon Web Services (AWS), Microsoft Azure, Heroku and more, so that organizations can take full advantage of the cloud’s elasticity.
By supporting a range of deployment options, users can leverage the option that meets their business’ needs and budget. The InRule rules engine is enterprise-proven and scalable, allowing millions of decisions to be processed in seconds.
Monitoring, Analytics and Improvement
Last, but not least, a decision platform should provide analytics that deliver visibility and insights into business decisions. InRule accomplishes this in three ways. The first is by providing metrics into automated decisions. While authoring decisions, users can select fields as key performance indicators (KPIs) that are important to their business. After execution, users can visualize the outcomes in a dashboard. For example, an online loan originator could easily view which region of the country is yielding the most applicants – and which regions could benefit from additional marketing investment.
The second way the InRule Decision Platform enables monitoring and improvement is by displaying the outcomes of decisions. These decision insights empower users to optimize rules and decision logic. For example, users can explore various aspects of decisions such as: how frequently certain rules are firing, which rules are being used most (or least), and the impact of each rule. In a loan application, InRule allows a user to easily see which logic is disqualifying the largest percentage of individuals from being approved for a loan. Then, if desired, a user can modify the logic to increase the pool of eligible applicants.
The third analytical component of the InRule Decision Platform is decision simulation. Decision simulation allows users to run large-scale what-if scenarios with decision logic. Users can adjust decision logic to see how the outcomes would change if that logic were to be deployed in a production environment. For example, a user could simulate broadening loan approval parameters and see how many additional loans would be approved with modified business logic.
By including analytics with the InRule Decision Platform, users can see the impact of their automated decisions, share those insights and advance the ability to make data-driven decisions across the enterprise.
Industries and Use Cases
A decision platform can add value to organizations across a wide range of industries and applications. Decision platforms are often used in insurance for underwriting, quoting, rating and claims adjudication. Additionally, we see a large use of decision platforms in the financial services for loan underwriting and approval, fraud detection and credit scoring. Public sector agencies also rely on decision platforms for eligibility and case management. Finally, we see decision platforms in the healthcare space for claims adjudication, benefit verification and cost estimation, and clinical testing and trials.
Wrapping it Up
I hope I have helped you gain a better understanding of how InRule defines a decision platform. As technologies emerge, we are confident that the platform, and the benefits it will bring to enterprises, will evolve.
One final thing to note: this blog post summarizes InRule’s thoughts on a decision platform as of Q4 2019. As time goes on and customer needs and technologies evolve, so might our thinking, designs and roadmap.
What do you think? Do you agree with our vision for a decision platform? Is there something we’re missing? Please share your thoughts in the comment section below. Thanks for reading!