Digitalization and end-to-end plant lifecycle management are spoken of as top priorities by companies across the chemical, oil and gas, process and power industries.
The goal? A complete digital footprint of the plant engineering helps optimize operations, plant maintenance, build better products and provide better services.
But how can traditional Engineer to Order (ETO) businesses, who often lag behind other sectors in terms of information technology, product and data management, even dream of a plant digital twin?
Complex one-off products and low volumes
Products are complex and produced in low volumes and in high variety.
Plants and Facilities are almost always the result of a project. Engineering, Procurement and Construction companies (EPCs) design and build these plants to meet single Owner-Operator (OO) requirements, which often lead to Engineer-To-Order solutions. The order specification process is often lengthy and complex, a labour-intensive process that requires deep technical knowledge.
It typically takes several years from initial customer inquiry to the closing of the deal—and yet the deal is still not done. The detailed design of the plant or facility is not carried out until an order has been signed. Each new order often involves new product development to match customer specifications. Products usually end up being highly customized.
This is entirely different than traditional “PLM-ish” industries, where the product is designed before it’s sold.
The main challenges: departmental silos and interoperability
The final plant is the result of an intense collaboration between and among functions and disciplines. The civil department, mechanical equipment, structural building, piping… everyone has their own set of discipline-specific tools. Applications and core systems are usually not well integrated with each other. And information ends up imprisoned in documents and discipline- specific authoring applications.
The result? A “messy” application landscape, departmental silos, lack of transparency and a lot of duplicated data. Each discipline gets locked into their own engineering tools and standards. Information handovers are sluggish and hinder the sorely needed cross-discipline collaboration.
Plugging into the power of the connected plant engineering
Connectivity within plant businesses is not new. Yet today’s advanced technologies and cheap computing, storage, and bandwidth costs make it possible to move beyond single plant and equipment automation toward more complex, connected product and process networks. The integration of data from several installed products—the so-called product instances—enables a holistic view of product performance and generates actionable insights to improve next-generation designs.
All this improves equipment performance throughout, leverages data-driven insights, increases the quality of the end product and prevents breakdowns before they occur. Identifying and responding to small problems before they become big, improves plant engineering and performance, and lowers maintenance costs.
For example, using cloud-based technologies and connected data, a plant digital twin enables an expert sitting in a central control room to monitor a product installed in several plants located around the world, troubleshoot problems and extract performance insights to help product development teams design the product’s next generation. It enables a field service engineer to scan an equipment, load its digital twin, follow maintenance instructions, check out historical reports and use advisory tools to improve equipment performance. And it equips analysts with a wealth of connected data to answer the questions: “How often?” “What?” “When?” “Where?” “Who?” and “Why is this a problem?”
The possibilities are many, and realizing them begins with setting the foundation for a connected model that allows for digitalization at the plant level. Product Lifecycle Management, enabled by the latest technological capabilities, provides this foundation and allows plant businesses to capitalize on the convergence of the digital and physical worlds and plug into the power of the connected plant.
Setting meaningful goals for your plant digital twin strategy
The decision to embark on a plant digital twin initiative should align with the organization’s specific needs. Understanding how the company intends to compete and aligning this strategic “true north” with the digital investments is crucial to success.
For example, some EPCs could decide to focus on their service portfolio and to invest in fast, reliable information to offer better services. Others may choose to invest in product configuration to speed up quotations and close more deals. Operators might want to monitor plant performance and invest in maintenance to extend asset lifecycles.
It’s important to prioritize strategic initiatives for investment based on the company’s specific objectives. What follows is a summary of the six overarching trends that seem to be accelerating the drive toward plant digital twins.
Leveraging the service side of business
Plants are products with long lifecycles. And EPCs are undergoing a fundamental business transformation to monetize these long lifecycles. It’s a shift from project to lifecycle business.
Rather than the “one and done” project approach of the past, the business model for most Plant and Facility solution providers is evolving from delivering projects to providing services for the entire plant’s lifecycle.
A portfolio of services ranging from spare parts delivery to plant maintenance or process monitoring can greatly boost the revenue from after-sales while building longer, closer relationships with customers. Better service means higher revenues, greater profitability and, over time, a powerful competitive advantage.
Leveraging the service side of business requires easy access to equipment-relevant information and smooth data transfer between project and service organizations. Service-relevant master data such as spare parts, spare-part kits or maintenance-relevant information must be defined and available to streamline service quotations and delivery.
Deep technical knowledge is needed to sell and deliver the final plant. And this knowledge often resides in the heads of a handful of experienced employees. Typically, procedures and rules aren’t written down and shared with others. A technically educated sales team is needed to fit the product to the customers’ needs.
Standardizing their methods and creating reusable knowledge allows plant businesses to take on more projects, scale and grow. Documenting processes and “decoding” product configuration and sizing rules from the experts’ heads is crucial to retaining knowledge and democratizing the selling process.
Gaining efficiency through reuse
Reuse (and the re-users) is the crux of the matter when it comes to efficiency. Reuse is a major cost-saver. Not only does it reduce time spent in delivering the final product, but it also allows development costs to be amortized over many projects.
Modularity is a design strategy for building and organizing complex products effectively. A product is modular if it has standardized interfaces and the components perform one or very few functions.
Providing less variety along products and adding commonality among components simplifies complexity and paves the way towards more advanced product configuration tools. You can still allow a certain degree of product customization by combining standard and custom components wisely and involving the customer in the product specification process.
Exploiting reuse requires a good dose of change management. By making designs reusable and storing them wisely so they can be found, your company prepares the ground to get reusers on board and fight the “not invented here” attitude.
Speeding up quotations to boost sales
Quotation speed and delivery time are order-winners. A clear and concise product description, using modular designs that limit the options available to customers, is key to speeding up sales quotations and closing more deals.
Salespeople can build a reliable, high-quality proposal faster and at a better price than others by using product configurators, templates and standard documents that have been fine-tuned and proven to be successful.
That’s not to say that the sales process operates on autopilot. Think of product configuration as a toolkit that salespeople have at hand to put together a compelling proposal while applying their skills to close the deal.
Traceability and compliance
Manufacturing and the supply of a large proportion of the plant’s equipment is outsourced to third parties. Orchestrating these equipment deliveries with suppliers around the world is a complex process that requires close attention to quality, scheduling and cost.
Adding to the complexity, orders and specifications might change during the project. This often leads to product design and documentation adjustments that need to be tracked and controlled. End-to-end traceability along the project and supply chain are critical to ensure compliance and quality.
A well-defined Plant Digital Thread is a keystone to demonstrating to regulatory authorities the integrity of plant-maintained information.
New digital revenue streams
The plant digital twin generates reams of data that can be monetized through the development of new products and services. Advisory tools to provide corrective advice to optimize process performance augmented reality to help field services maintain and repair equipment, remote process monitoring services analytics to reveal asset performance issues … the options are many, and by listening to customer challenges it’s easy to be inspired, develop new solutions and dive into new revenue streams that capitalize on access to fast, connected information.
Getting started: taking small steps towards the plant digital twin
The journey towards the plant digital twin requires navigating complexity and a keen focus on acquiring needed capabilities. While an overly simplistic plant model may not yield the value a digital twin promises, taking an overly fast and broad approach can almost guarantee that you’ll get lost in the complexity. The good news is that by thinking big and defining an ambitious true north, starting small with agile bimodal projects,and scaling fast based on the learnings, it’s possible to see the shoreline beyond the rough waters of the plant digital twin.
1. Start with a plant structure
Start by describing the plant’s structure. As my peer Bjorn Fidjeland explains in his Plant PLM eCourse , several plant information structures are required to describe plant engineering.
To get going, start with a functional structure. A plant’s engineering and functional structure is discipline-independent, hierarchical and breaks down the plant from the functional perspective. A plant structure includes plant processes, systems and equipment.
Once you’ve agreed on the plant functional structure “on paper”, it’s time to move it to a system. And where does the functional structure belong? If you’re already using a PLM system, that’s probably the place. If you still don’t have a system, you could experiment with Aras PLM, or look into Siemens, Dassault or Aveva comprehensive plant solutions.
2. Identify and share relevant information
Plants require intense collaboration between and among different disciplines, using a variety of different tools. The information-sharing across disciplines and functions is a major challenge and affects the traceability of information throughout the plant’s lifecycle. All stakeholders struggle to access information effectively due to its lack of visibility, a problem that is compounded when information is split among several discipline-specific applications.
Now that we have a common plant structure, the key is to get disciplines to start sharing their data. They need to understand why others need the information they produce and how easier work will be downstream if they spend a little more time on sharing it.
It’s OK to start sharing only documents. The goal at this point is for people in the organization to become familiar with the structure. Afterwards, we’ll start sharing models and metadata.
3. Define a product family modular library
A product family library comprises a predefined set of modules that can be combined in different ways to form a product. Modular designs offer not only an opportunity to speed up product deliveries. They also have an enormous potential to unlock service opportunities and increase customer loyalty.
It’s time to audit your product portfolio, seek commonalities and analyze how to reduce complexity and foster reuse. Defining a product family library to illustrate what modules are available and how they can be combined is a major breakthrough but also a huge task for ETO companies.
The variety and complexity of the product portfolio is high, and while companies usually don’t start from zero, it’s a big effort to come up with a comprehensive modular product library.
My advice? To win early support, begin with products that have the potential to maximize the return on investment. Find the most profitable product families and focus first on the common and profitable products and modules. If a product is seldom sold, the cost of defining a library may exceed the benefit.
4. Connect plant specification to product realization
The next step is to connect the plant’s functional structure, where we specify equipment, to product realization, the asset itself. At this step, it’s very useful to integrate the plant structure to the most common authoring applications. Plant disciplines operate in different CAD environments, and information often gets locked up in department-specific applications. Purchasers can’t access MCAD – they don’t even know how to use it or have a license. And the service colleagues spend more than half of their working time hunting for technical information locked in hidden drawings to order spare parts.
Integrating discipline specific authoring tools to the plant structure enhances visibility and provides easy access to plant’s technical data and product bills of materials. A reliable information flow between CAD and PLM speeds the development process, enables efficient information handovers between disciplines and upholds consistency in product quality.
In the example, we are connecting a centrifugal pump specification to the asset that’s fulfilling the specification. The difference between specification and realization is very important within the plant model. Imagine that this pump gets broken, and we need to change it. We’ll then bring in a new asset, but the specification stays the same.
5. Modeling the plant lifecycle
Most plants and facilities move through a lifecycle with approximately eight distinct phases, starting with the specification of the plant objectives and ending with decommissioning. The plant structure and the connected products will evolve through the lifecycle. At this point, we can start defining a model to maintain the plant’s “digital thread” all the way through the process.
6. Define a plant catalogue
Equipment in the plant is specified using attributes. Some of these attributes are measured by field instruments, creating a digital feedback loop or, in some cases, a remotely accessible live view of the plant. Classifying processes and equipment in the plant and defining reference data are needed to automatically create a plant tag and effectively connect all data—including the data from sensors in the field.
A major challenge when dealing with field data consists of the systematic and efficient information tagging of sensor data so that we don’t end up comparing apples with oranges. If you want to know more about reference data and catalogues, head to my peer Bjorn Fidjeland’s blog. He has several excellent articles where he discusses these topics.
7. Integrate suppliers
Different interactions with suppliers are required during each stage of the plant lifecycle to share and store information. Efficiently collaborating with suppliers is key for plant businesses, where a big portion of the equipment deliveries come from third parties. Integrating suppliers into Plant PLM facilitates document exchange, product model collaboration and change-tracking, among other things.
8. Harness information for action using end-user apps!
Now that we have the data, we can harness it for action using end-user apps! End-user applications are easy to configure and drive meaningful value, whether it be through powering better decision-making or enhancing consumer-facing applications.
Imagine an application for salespeople who need to constantly monitor after-sales opportunities for their products. Through an easy-to-use application, they could track installed equipment, gain performance insights and tap into new service and modernization opportunities.
I’m a big fan of end-user applications because they facilitate daily work and generate additional insights for specific use cases.
A journey of a thousand miles begins with a single step
Jos Voskuil concludes in his “PLM for Owner / Operators” blog that digital continuity requires a “new way of thinking” for plant owners/operators, who are “struggling to grasp a modern digital enterprise concept as their current environment is not model-based but document-driven.”
All in all, true success in achieving early milestones on a digital twin journey will likely rely on an ability to grow and sustain the digital twin initiative in a fashion that can demonstrate increasing value over time.
It can be an overwhelming task to get there, but as a wise person once said: “A journey of a thousand miles begins with a single step.”