Towards Creating an Adaptable and Resilient Digital Enterprise
An organization that strives to achieve significant improvements in their engineering practice via the implementation of a Digital Engineering Strategy (DES) must also recognize that not only are there many tools to bring to the fight, but that those tools themselves will adapt and be acted upon by other emerging technologies and processes. In fact, this is a desirable situation for an effective digital engineering enterprise because to only emphasize the primacy of models, and/or to just specify authoritative models/data without “engineering innovation” into the enterprise threatens to create a stovepipe all of its own.
Considering this aspect even further: modern technological activities/projects are producing enormous amounts of data, most of it assumed to be of value. Unfortunately, many organizations, in their attempt to produce the best insights that are made available to those who need it in the most efficient manner possible, are struggling with the effective processing of this data. Hence, we observe increased interest and scrutiny of enabling technologies such as high power computing, artificial intelligence, machine learning, digital twins and a host of other examples (see image above) to help deal with the issue.
From a life-cycle perspective, the organization’s goal is to leverage the digital domain in the early stages to rapidly and accurately build up the system concept and mission application, and then in the later stages (build, test, deploy, sustain) to efficiently acquire new system data to be fed back into the digital model(s) to evolve the design, obtain new insights, and communicate impacts to all stakeholders.
Let’s imagine a digitally-designed combat ship, composed of a myriad of systems, sub-systems, components, etc. As a whole, the ship is to fulfill a number of assigned missions: maritime defence/control, interdiction, counter-drug, search and rescue, etc. Digital mission profiles can be created for each of these scenarios and used as decision-supporting data for capability development and acquisition (build or buy).
Upon deployment, information regarding the system performance during mission execution can be fed back to the existing digital mission models to determine how closely they align with reality. The key point here is not so much that the existing system’s model is improved (e.g., improvements in tactics, techniques and procedures (TTPs), is one example, but insight into system maintenance and sustainability is another ), but that future capabilities/needs are informed to support the organization’s overall mission (think mid-life upgrades or next generation ship design).
How Technology Can Enhance Digital Engineering
Let’s take a quick look at some of these emerging technologies and try to understand their possible impact on Digital Engineering:
High Power Computing – According to InsideHPC: High Performance Computing most generally refers to the practice of aggregating computing power in a way that delivers much higher performance than one could get out of a typical desktop computer or workstation in order to solve large problems in science, engineering, or business.
Impact on DE: More and more, the nexus of sensor input, processing algorithms, and technological scalability has become the focal point for tackling complex engineering problems in a time and/or resource efficient manner. The value placed on the insights to be gained from the processing of this data is pushing more organizations to not just consider researching high power computing architectures, but to actually plan for them at the outset.
Artificial Intelligence/Machine Learning – For those sufficiently schooled in the terms of artificial intelligence (AI), deep learning (DL) and machine learning (ML), it is well known that AI has roots in the 1950s and that concepts such as DL and ML became technological sub-sets of AI – but with key differences.
AI, in a broad sense, is the application of computer algorithms to create a computer response similar to that which might be expected from a human being. ‘Alexa’ and ‘Siri’ are good examples of AI in which these entities attempt to respond to human requests in near-human ways.
Machine Learning takes a different path and instead involves the provision of great amounts of data and means to learn from it. It is very much patterned on the way that humans experience and learn things, and thus the computer ‘learns’ by itself over time.
Impact on DE: Since the development of complex systems involves a great deal of resources, inputs, and analyses, the responsive challenge is to create systems and/or processes that can evaluate data and suggest courses of action, or to identify patterns that could indicate system problems far into the lifecycle. In most, if not all, cases, the intent is not to fully replace the human-in-the-loop, but rather to provide intelligent assistance to arrive at the best design decisions sooner.
Digital Twins – the IBM UK Technical Consultancy Group defines Digital Twins as the following: “A digital twin is a dynamic virtual representation of a physical object or system, usually across multiple stages of its lifecycle. It uses real-world data, simulation or machine learning models, combined with data analysis, to enable understanding, learning, and reasoning. Digital twins can be used to answer what-if questions and should be able to present the insights in an intuitive way.”
Impact on DE: Beyond the fact that the terminology above aligns very closely with the goals and objectives of digital engineering, digital twin technology is a powerful way for an organization to ‘learn in real-time‘ about its complex systems as they feed data back into the virtual model/process. Increased and advanced awareness of physical system status, especially as it experiences conditions that had been modeled for its design, validates data/model assumptions, from which future decisions can be made.
Aspiring to incorporate the benefits of technology in support of the digital engineering practice is akin to an organization becoming ‘a life-long learner.’ As the need to represent, evaluate and analyze complex systems increases, so too must the application of leading-edge approaches to working with the data, models, output, etc to continue to accelerate their development and deployment, while striving to keep the risk level low. These enablers are like ‘oars working in unison’ to propel the system design forward.
Up to now, we have been considering organizational options for those that are at least willing to ‘talk the talk‘ regarding implementing a Digital Engineering Strategy (DES). This has involved specifying their intent, objective and vision of what a DES could look like and potential outcomes. Next, we will delve into ‘walking the walk‘, i.e., what kind of collaborative, comprehensive yet secure infrastructure is needed for DE success.