Guest Post: How to Integrate Process Models with Decision Intelligence
I recently attended Quantellia’s Getting Started with Decision Intelligence training, where we learned to differentiate between a Causal Decision Diagram (CDD)—a key component of Decision Intelligence (DI)—and a process map (PM). The following is a brief recap of what I learned from Dr. Pratt and an idea regarding how we might combine these two disciplines to manage strategic decisions across the organization.
Process mapping is a likely familiar practice of using flow chart diagramming to map out the steps of a business process. It creates a reference artifact to standardize best practices for a business process. Such a process might be “how to apply for travel funding”, “how to process an order for a designer suit”, or “how to ask for project funding”. A Process Map uses simple graphic elements to represent the tasks and routing necessary to complete a process. Here’s an example:
In contrast, Decision Intelligence uses CDDs that map “actions to outcomes”. A CDD is a diagram of the impacts of a decision – the ripple effects diagrammed in a chain of consequences. This chain of consequences in a CDD can include both direct and indirect implications of the decision, thereby providing a powerful check on unintended consequences. Also quite powerful is the ability of a CDD to represent effects within your organization and beyond, thus enabling formal consideration of social, regulatory, or environmental impacts. Here is an example of a CDD:
In my example from above, a PM might walk you through the steps to take and forms to fill out to request funding to fly to a conference for work. A CDD would walk you through the impacts, risks, and outcomes of that decision – such as the negative climate impact, positive career impact, and attainment of organization goals. Both PM and DI analysis are highly valuable, yet they’re not usually combined well, much less strategically integrated.
How Process Maps (PMs) and Causal Decision Diagrams (CDDs) can work together
The combination of Decision Intelligence (DI) and Process Mapping (PM) across an organization would enable systematic coordination and optimization of decision making. Process mapping would provide a view of business practices and the location of key decision-making points. These decision-making points in the process map can then be considered “portals” to the DI process and Causal Decision Diagram (CDD) for that decision. Decision Intelligence would then optimize those decisions.
Not all processes or decisions would benefit from this level of management, as I illustrate in the image below. Simple processes with simple decisions may only need a checklist. Processes with simple decisions may only need a process map. Complex decisions that are standalone or part of simple processes may only need a DI project, including a CDD. However, in cases where the decision and processes, or network of decisions, is complex and of high value, then the combination of DI and process mapping may best optimize the organization’s effectiveness.
This combination of DI and PM may form the basis for a “Decision Intelligence Ops” (DIOps). DIOps would be the collaborative design and management of decisions and processes focused on optimizing decision making and coordinating decision processes among decision makers, stakeholders, and decision support resources (e.g., AI, ML, data science) across an organization.
A new kind of diagram, for “DIOps”
We could map the interplay of the process steps and CDD results in a “DIOps diagram”. A DIOps diagram would include both DI and PM diagrams that are cross-linked to support optimizing business processes and decisions. For example, this diagram could integrate the use of DI to help inform business process design in the planning stage. Once the business process is implemented, a DIOps diagram could further coordinate the integration of ongoing CDD simulation to direct the business process flow at significant decision points. It can also help illustrate the proximal and distal impacts of a decision across a business process and across upstream and downstream decisions. Here’s an example:
There is a growing appreciation for the value of data, analytics, and artificial intelligence. With Decision Intelligence, we can see that these technologies are just a part of a greater value realized through advanced decision practices. DI is bringing about a revolution in leveraging data, AI, and human insight to answer our most complex and challenging questions. Other agile disciplines (e.g., DevOps, DataOps) have realized significant advantages in strategically integrating and scaling high-value practices across the organization. It is not too soon to see that developing a DIOps strategy for bringing DI capabilities to scale will be paramount for organizations.
William O'Shea, Ph.D., has two decades of experience working across organizations to solve problems and leverage insights. He is currently the Director of an analytics department at a private university in Oregon. In addition to higher education, his background includes experience in financial services, utilities, and online retail. William has written about and given invited presentations on data visualization, analytics development, and artificial intelligence.