Digital Twins + Agility = Faster, Better Outcomes

by Michael Idengren , Principal

In Digital Engineering, engineers spend significant resources and time Modeling, Simulating and Analyzing (MS&A) Digital Twins of systems (especially hardware-inclusive), so they can apply Model-Based System Engineering (MBSE) to simulate the effects of changes, for rapid, cost-effective experimentation - before building an expensive physical prototype.

… but what about the people, processes and technology needed to design, build, test and deploy those systems? Shouldn’t some resources be allocated to helping Digital Engineering experts do more Digital Engineering… or helping teachers do more teaching… or really, any professional delivering faster value?

The Need

For faster research, better decisions and more frictionless business & operations (“bizops”), it’s important to uncover and improve ineffective business/operational processes, in context of the people, process and technology - and to help people more effectively do the job they were hired (and interested in) doing. This is nothing new – the #1 priority of the 2022 Federal Workforce Priorities Report is to Leverage Technology and Modernize IT Processes. Doing this well, and in an agile-principled way, can help “give time back” to the experts we depend on to build these complex systems.

For example, in a US DoD research lab example, there are many talented scientists & engineers (S&E’s) that have crucial institutional knowledge needed to design, build, test and deliver technology into the hands of the warfighter. Every day, these S&E’s can be frustrated by a poor User Experience (UX) to do their jobs – such as outdated, ineffective technology, slow Authority-To-Operate (ATO) technology approval processes, and difficulty staffing (e.g., just getting contractors “in the seat” sometimes takes too long).

So – whether it’s research, engineering, or any kind of business domain - how can we “give more time back” to the experts we depend on?

Asking the Right Questions

To make improvements, it’s first important to understand where the problems are, and which ones to solve first (everything cannot be a priority). Leaders will typically use Objectives and Key Results (OKRs) for this. To make this more ‘actionable’ and collect metrics that matter, we have worked with leadership using a Goals, Questions, Metrics (GQM) approach, decomposing goals into “questions” that get answered with “metrics”.

When the goals and important metrics are agreed on, it’s time to Model, Simulate and Analyze (MS&A) possible solutions, considering improvements across people, process and technology. It’s important to note that these improvements must be modeled together - they cannot be done in isolation, especially because of the nature of trade-offs inherent in process improvement decisions.

For example, many organizations will make the mistake of buying technology, but failing to invest enough time in training and process improvement to effectively utilize the technology. Or, maybe “excessive” security controls are causing a poor User Experience.

Finding the Right Answers: [ Engineering vs. Enterprise ] Digital Twins

Many people are familiar with Model Based System Engineering (MBSE), for example, creating an Engineering Digital Twin of a rocket, or an airplane engine, collecting lots of operational/test data and running simulations to determine optimal designs.

An Enterprise Digital Twin is similar to the concept of an Engineering Digital Twin, except it seeks to improve the organization’s mix of people, process and technology, to help people deliver value faster. Modeling, simulating and analyzing an Enterprise Digital Twin is an effective way to consider all the variables together, and to simulate possible changes to the model. For example, if we reduce the burden of an IT security control (probability * impact), maybe there is “a little bit” more security risk - but it might be worth it, if the subject-matter expert (SME) is able to get 30% of their time back!

Faster Outcomes & Less Blockage with SAFe principles

Powerful tools & techniques, such as Artificial Intelligence (AI)-enabled process intelligence mining tools (e.g., SoftwareAG ARIS) can help leverage whatever data is available and make “smart” suggestions - such as giving a “head start” on modeling by automatically generating a process from a bunch of text, and running thousands of simulations, adjusting inputs to consider possible improvement outcomes.

But all these tools are just that: tools analyzing data and providing options to make trade-offs. Leveraging Scaled Agile (SAFe) principles, it’s helpful to think about making trade-offs and improvements in “smaller chunks”, considering which outcomes are most important to deliver now, and which ones can wait until later. For example – do we increase operational budget to hire more people? Shorter term, do we accept more security risk to improve the UX? Longer term, do we fund capital IT investments to improve both security and UX?

In the end, a human (“in the hot seat”) has to make the hard calls - but by leveraging an Enterprise Digital Twins, AI tooling and agility principles, those humans can make better decisions, helping people deliver faster, higher-quality outcomes.