Business automation is supported by a wide range of standards and platforms.

• Business Process Management (BPM) has Business Process Model and Notation (BPMN) and a wide range of process automation platforms.
• Decision management has Decision Model and Notation (DMN) as well as Business Rules Management Systems and Decision Intelligence platforms.
• Data is modeled using a range of techniques – entity relationship diagrams being the most common – and the range of platform types for storing and managing data is both extensive and continually growing.
• In recent years, a myriad of machine learning models and MLOps platforms as well as AI tools and platforms have become increasingly common.

Each of these elements – process, data, decision and analytics – have a role in effective automation.

Exploring and discovering automation is the most critical step before technology and techniques come into focus.  In many cases, DMN can be used to inventory and explore automation opportunities. To make smarter, more effective and efficient decisions, the decision-making itself should be modeled, centralized and automated. A focus on decision automation and modeling finds ML possibilities, generative AI touchpoints and captures the guardrails of deterministic logic. It also results in more streamlined and simplified processes and clarifies data needs, enriching existing data and focusing analytic investments.

From there, a project team establishes a North Star for what they want and it guides their technology selections later—process tooling, ML algorithms, LLM integration, data availability and governance capabilities.  DMN in many respects is destination first in its orientation and everything else aligns around it. We recommend modeling decisions before modeling the processes of which they are part. We likewise model decisions before selecting data for use in those decisions and before framing and training machine learning models.

When considering decision automation:

• Processes exist to get data ready to make decisions and to act on the decisions made.
• Machine learning models make predictions that inform decisions but don’t make decisions themselves.
• Data flows through the process and must be available and suitable to support decision making.

Automating decisions is fundamental to successful business automation.