As industrial enterprises dig themselves out of the rubble of the latest disruption, long-accepted views of product development and other product lifecycle phases, such as production and service, are increasingly recognized as inadequately defined and supported. Complexities, risks and opportunities that have been downplayed or ignored are being tackled on the factory floor, in the field, in the executive suite and everywhere in between.
The result is a massive upheaval sweeping through the enterprise—one that impacts how the enterprise operates, prospers, and views itself, its competencies and its products. I've noticed that with increasing urgency, leadership teams and corporate management are demanding speed-ups in digital transformations to sharpen their competitive edges.
Part of this upheaval is the rapid acceleration of the extraction of information that is trapped in paper documents, or tangled in data formats that often mimic paper. Data access and quality issues burden almost all product development, production and support teams. Liberating this information requires transforming an unfathomable amount, and variety, of data into the model-based enterprise (MBE) and its many model-based structures.
Disruptions of supply chains, component shortages, workforce skill mismatches and the after-effects of the ever-evolving COVID-19 pandemic are turning digital transformation’s long journey into a race. I don't think slow-moving organizations are likely to prosper; they will be blindsided by changes that are multiplying in every marketplace.
You do not have to read them in any specific order, but this is the first of four self-contained articles written by CIMdata for engineering.com. The next article to release was Every Business Needs Data, But Do They Understand the Big Picture?. The next article released covered how, and why, Digital Threads in PLM are really webs of data. The remaining article will explain what digital transformation requires of the workforce.
The Big Challenges in Digital Transformation
Complex market requirements, and the resulting increased product and process complexities, demand faster and more effective business approaches that span all product-oriented lifecycle disciplines and leverage all MBE enabling technologies. In MBE environments, it is crucial to understand all the potential benefits of digital transformation; in particular, moving from a document-centric environment to one that is data-centric. Until the benefits are completely understood, I fear investments in technologies and applications that enable true end-to-end product lifecycle management (PLM) may dry up. Fortunately, help is at hand.
PLM solution providers and standards bodies are expanding their connectivity offerings to enable various forms of digital transformation, digital threads and digital twins CIMdata and similar organizations track these developments and help clients choose and implement the most appropriate solutions. Configuration management and data governance help align and regularly update enterprise business models and organization value propositions for better decisions. Both configuration management and data governance are fundamental to digital transformation.
What we think of as the “big picture,” commonly seen in the early stages of the lifecycle (during product development), has never told us everything we need to know to compile and evaluate a new product’s complete lifecycle requirements.
I like to label these sweeping changes as forcing factors. They include:
- Large and growing electronics content and software in physical products.
- New manufacturing techniques and support processes.
- The introduction of new materials that are lighter, stronger and “greener.”
- Mass customization, with all the latest technological features in each new offering.
- Shorter product lifecycles and nonstop innovation.
- The Internet of Things (IoT) with its enablement of continuous marketplace feedback.
What this means is that complex and continually changing marketplace requirements demand up-front, cross-domain connectivity—all with the appropriate and easily accessible model-based data constructs.
This complexity can only be tackled with cross-functional activities and investigations to root out the unknowns and then nail down wise solutions that are optimized from a complete system of systems perspective. Why this has been so difficult and where to concentrate productive efforts is the focus of my digital transformation commentary.
The Top 12 Trends and Enablers of Digital Transformation
To show why this journey has often been so difficult, I want to present CIMdata’s Critical Dozen. They are 12 familiar, evolving trends and key enablers of digital transformation that you cannot, or should not, live without.
We will dig into each, but briefly they are:
- End-to-end connectivity
- Data and process management
- Configuration management (CM)
- Bills of information (BoI)
- Model-based structures
- Digital thread
- Digital twin
- IoT and product lifecycle management (PLM)
- Changing views of the “product”
- Big data & analytics
- Data governance (DG)
- Digital skills transformation
Aspects of each of the Critical Dozen have proven to be disruptive. Still, a company’s resistance to change has often impeded its ability to digitally transform fully and successfully. The real difficulties are often up-front in the lifecycle (e.g., weak project/product/process planning) and failing to accept that true and lasting transformation requires change among people, organizational structures, processes, data and in all the associated enabling technologies.
1. End-to-end Connectivity as a Digital Transformation Enabler
Over the life of a product—from the early stages of requirements definition to design, to production, to field service and finally to disposal or recycling—connectivity is about linking people, processes, data and technology.
Easily overlooked are semi-isolated workgroups and processes at the beginning of a new project and in the project’s final stages/end of life. Among these are competitive analysis, tooling, requirements planning, field service and their processes and data—which developers of software and strategies for lifecycle management and product innovation tackle every day.
Today’s industrial enterprises are almost totally dependent on data and how it is processed into usable information for making decisions. This data—0s and 1s placed in context—is a valuable asset, not just digital clutter. Incorrect or incomplete information underlies nearly all enterprise anguish.
2. How Data and Process Management Enables Digital Transformation
Information must be freed from silos. This includes the stashes of departmental and business-unit data and obsolete (“legacy”) data-handling systems. While these once satisfied specific business needs, they now hamstring the ready access, reuse and analysis of vital data in all its forms. Without end-to-end connectivity, basic information far down in a file’s layers and tiers—specifications, for example—is often unreachable. Newer data management solutions are replacing silos and legacies, but this is moving too slowly.
The two images below compare and contrast information flows. Users of many complementary solutions and task-focused “apps” still struggle; as do industry standards committees and solution providers. Most apps and solutions fall short of linking interrelated data flows and the processes by which they are created and/or used, let alone span the enterprise end to end, top to bottom.
My organization's product innovation platform model lays out the six key considerations in multi-disciplinary lifecycle optimization—the Functional Domains that operate in just about every industrial organization.
The domains’ establishment and evolution are orchestrated in a system of systems approach. When done at the enterprise level, every decision made during a product’s lifecycle can be supported and optimized to drive the best return on investment. Piecemeal, departmental approaches to innovation platforms inevitably miss two or three key functions and never work particularly well.
3. Configuration Management Ensures Enterprise Intentions
Fully implemented configuration management (CM) tells the enterprise that all its products, facilities, IT systems, services, processes and more are aligned with what is intended. CM “knows” where everything is, who has it and whether it is properly implemented and supported. “Everything” here includes product ideas and requirements, and it runs right through to removal from service. “Everything” also includes design, production, installation, warranties, operation and maintenance, plus the testing and validation within all of these.
CM also ensures that all changes to the information are properly evaluated, authorized and implemented. Data and information that defines and manages product-related processes are kept current, accurate and structured to promptly meet all users’ needs.
4. Bills of Information Structure Data for Digital Transformation
The Bill of Information (BoI) is a comprehensive and structured approach to managing product data. This data construct mimics Bill of Materials (BoMs) that tally every component, instance, source and cost of a product's parts. A BoI will also include supporting data like requirements, process definitions, options and variants, as well as decisions and their associated rationales. Fundamentally, the BoI tracks everything going into a product or service and their virtual counterparts (as in digital twins, see below).
Compiling everything representing a physical product (or service) into one logical BoI structure can be a challenge. And PLM can be a big help. A BoI must link each and every component, sub-component and every variant with unique views, serial numbers, suppliers, analyses, test data, materials, tooling and much more. Variants and components often have their own BoMs, sometimes dozens of them, with no two alike.
Nevertheless, compared to a stack of BoMs, the BoI simplifies and speeds up handling product-related data.
5. Model-based Structures Define the Enterprise’s Processes, Products and Services
Model-based structures are best understood as the moving parts of the model-based enterprise (MBE). In MBEs, information is fed into multi-dimensional computer models of every component, tool, system, product, service, function and process in the enterprise.
Model-based disciplines include model-based engineering (also using MBE as an acronym), model-based systems engineering (MBSE), model-based definition (MBD) and model-based design (also MBD). Because MBSE, the MBDs and the MBEs overlap, peering into the model-based enterprise can get murky. Getting information that enables each appropriate data structure requires sorting out the overlapping and confusing “model-based” abbreviations and acronyms.
Like so much else in digital transformation, sorting out MBE components is not for the faint of heart—especially when essential information is in data silos.
In hopes of sidestepping an acronym war, many organizations roll it all together as MBx, which boils down to model-based anything and everything.
6. Digital Threads Ensures an Asset's Data is Accurate
I view the digital thread as the bidirectional communication framework that connects data flows related to products. These flows produce an integrated and holistic view of an asset’s data from both physical and virtual systems (i.e., its digital twin) and throughout its lifecycle across traditional functional perspectives. This definition is also sourced from the U.S. Air Force ManTech program documents.
These data flows help bring together the virtual and the physical by linking the alphabet soup of enterprise systems (including ALM, EDA/MDA, ERP, PDM, CAE, MES, MOM, MRO, M&S, testing, QA, V&V and TDPs). The digital thread’s continuous updates also track values throughout a product’s digital life. They are the key to permanently unlocking siloed data. Digital thread connectivity requires interoperable and bidirectional data flows.
7. Adding Digital Twins to Digital Transformations Cuts Development, Construction, Service Issues and Failure
Products and processes need more representation than 3D CAD-like images, hence the digital twin. Here are two experts’ summaries:
“The ultimate vision for the digital twin is to create, test and build our equipment in a virtual environment,” according to John Vickers, longtime manager of the National Center for Advanced Manufacturing at the NASA Michoud Assembly Facility in New Orleans. “Only when we get it to where it performs to our requirements do we physically manufacture it. We then want that physical build to tie back to its digital twin through sensors so that the digital twin contains all the information that we could have by inspecting the physical build.” (Vickers is now a senior leader for Advanced and In-Space Manufacturing Capabilities in the Materials and Processes Laboratory at NASA’s Marshall Space Flight Center.)
“The real advantage of the digital twin…materializes when all aspects, from design to real-time data feed, are brought together to optimize over the lifetime of the asset,” said Michael Grieves, writer for GE Lookahead and Executive Director of the Digital Twin Institute. “An accurate digital description of a physical asset, for example, does not just cut prototyping or construction costs, it also enables to predict failure more easily once real-time data is fed into the model, thus reducing both maintenance costs and downtime.”
Six key capabilities are required to define and manage a digital twin, which can only be done effectively through PLM enablement.
- End-to-end data and process connectivity that helps ensure the definitions of the virtual product and the virtual process are always clear, concise, valid and in sync with their physical counterparts.
- Managing the evolving data configuration of a product or service through its entire lifecycle.
- Bidirectional traceability via associated digital threads.
- Optimization of all the relevant lifecycles across all relevant systems.
- Full associativity between the virtual and physical (i.e., connecting the left and right sides of the Systems Engineering Vee).
- Furthering the economic enablement and expansion of the digital twin with the IoT, Big Data and other technologies and associated initiatives.
Defining the digital twin is relatively easy, but as with the digital thread, end-to-end bidirectional connectivity and interoperability are necessary. In a digital transformation, there are several potential stumbling blocks:
- Weak associativity of data/information across differing file formats and solutions.
- Poor traceability of data/information through multiple locations (silos included).
- Incomplete or confusing definitions of each virtual product and virtual process.
- Finding “downstream” feedback loops to track changes and decisions.
- Indecipherable references buried in specifications.
- Obsolete data systems.
- Persistence of paper documents (particularly in inspection).
- Resistance to change.
8. The IoT and PLM Make Digital Transformation Possible
IoT and PLM have critical roles in the digitally transformed enterprise: managing and verifying data inflows (tsunami or otherwise) from countless devices and sensors. Both the IoT and PLM must be understood—and used—in an extended enterprise-wide systems of systems approach.
Such an approach connects the factory floor to the product while finding and closing data feedback loops. Feedback loops communicate changes and decisions about products, production and processes “upstream” via digital threads, for instance, and tracks them. Feedback loops are ever-changing but their information can be extremely valuable to decision-makers.
Unlike anything else in tech or data management, the IoT can be used to constantly deliver information about a product’s performance while it is in the customers’ hands. The impact of this feedback can be pervasive, making the IoT—and its more focused subset, the Industrial Internet of Things (IIoT)—crucial to end-to-end digital transformations throughout an organization’s production facilities.
Connecting an IoT strategy with PLM begins with understanding how the IoT can be used to interconnect systems of systems, and that the IoT is far more complex than even a big data tsunami. A systems of systems approach usually requires implementing and adopting an enterprise-class product innovation platform.
As MBE enablers, product innovation platforms start with managing and defining all the information supporting the decisions (both tactical and strategic) in developing a new product and/or service. These platforms work throughout the product’s complete lifecycle, including project planning, data discovery, applications, toolsets, functions and feedback loops.
Successful IoT strategies require continuous and decisive action on the data—not just collection and sporadic analysis. The “why” is obvious, but the “what” and “how” are not.
Out of necessity, digital transformation usually embraces PLM and product innovation platforms that enable it to maximize collaboration among users, tools and databases. To ensure collaboration, an innovation platform must be able to access analyses of the impacts of decisions and ultimately tie those analyses to the business models and value propositions that guide enterprise leadership.
In any worthwhile IoT-enabled strategy, a PLM solution is the cross-functional process and data backbone. As noted in a 2016 Genpact Research Institute study of digital transformation, this backbone must reach throughout the enterprise and its supply chains and must be scalable to meet new needs and opportunities.
In effect, this backbone helps ensure quicker and better executive decisions that translate into timely investments, more competitive products/processes and greater assurance of sustainable profits. This is making the urgency and value of upgrading obsolete PLM solutions more obvious.
Ensuring timely investments, we should add, means tallying PLM benefits to secure ongoing project support—such as faster time to market, improved profit margins on products and services and reduced warranty claims. Don’t be sidetracked by small reductions in headcount or product costs.
9. What is a “Product” These Days?
Perhaps unnoticed amid all these disruptions, the meaning of product is changing. We now must ask if each product is:
- The individual connected device?
- The system of connected devices from one supplier?
- The total system of connected devices?
- Inclusive of data?
- Compliant with industry standards?
In a world where even bolt heads have connectivity (to guard against over-torque and cross-threading), “standalone” is a fading concept while data and connectivity multiply the complexity and value of the physical product. This accounts for the speedier obsolescence of products, processes and systems. Everyone in production, analysis and management are scrambling to keep up.
10. Big Data and Analytics Are Necessary, But Easy Data Access Poses Security Risks
Big Data, delivered by the IoT, and the analytics tools that make sense of it all, go hand in hand. Tapping into Big Data can immediately lead to an exponential surge of incoming data—ten, a hundred, a thousand times or more every day.
Much of this data will be accessed automatically; keyboard data entry may cease altogether. Legacy analytical tools may be overwhelmed so highly scalable, robust replacements will be essential. Robustness is an analytical tool’s ability to handle a wide variety of fast-changing data.
Reaping the benefits of greater connectivity and easier data access also poses enterprise risks that we must not ignore. Namely, it multiplies the potential points of entry for hackers. Ditto for the costs/consequences of a successful hack. Make sure enterprise cybersecurity is fully on-board with any boost in connectivity.
11. Data Governance Defines and Enforces the Policies, Structures and Procedures of Digital Transformations
Data governance (DG) is essential information management with end-to-end data oversight, accountability and stewardship. I define DG as the organization and implementation of policies, procedures, structure, roles and responsibilities that outline and enforce rules of engagement, decision rights and accountabilities to effectively manage information assets.
DG goals include sustainability and change management for data/information, guidance for in-house software development and analytics and measuring improvements. Note that DG and configuration management are complementary, not an either-or proposition.
12. Your Workforce’s Digital Skills Must Be Transformed As Well
The final Critical Dozen for digital transformations is digital skills transformation. Understanding new terminology, implementing new processes, learning and using new digital techniques and updating business paradigms are more important than ever. A major consequence is that on-the-job experience and intuition are being pushed aside.
Digital transformation and the model-based enterprise require equipping the workforce with new digital and analytical skills to meet endless change. This includes identifying which skills will be most needed and a new mix of retraining and hiring.
Several types of know-how are essential in this “reskilling.” They include a grasp of:
- Technology trends.
- Information content.
- Delivery methods.
- Organizational processes and their solutions/structures/platforms.
- Implementation and support of whatever is new.
- Incorporating varying experience levels.
- Skill attainments using a single standard throughout the enterprise.
Workforce transformation will make the enterprise more adept at handling change and growth, new technologies and updated processes. Employee frustration and turnover will be reduced, allowing decision-making and responsibility to be pushed deeper into the organization.
Final Thoughts: Critical Dozen to Digital Transformations
We have now reviewed all the connectivity and access components of digital transformation and PLM’s central role. It’s a long list—bills of information, configuration management, industry standards, commercially offered solutions, the model-based enterprise (no more paper), digital twins, digital threads, the IoT, Big Data and analytics, data governance and the transformation of workforce digital skills.
How these elements fit together will vary by industry and by enterprise, but none should be skipped. As we have noted, incorrect/incomplete information underlies nearly all enterprise grief. Look for future articles that address specific aspects of the Critical Dozen we have introduced here.
NOTES:
2. “2016 Reality Check: Transforming Industrial Businesses with the Internet of Things” by GE Digital and the Industrial Internet Consortium, sponsored by GENPACT in conjunction with IndustryWeek magazine.