Successfully managing industrial operations is becoming an increasingly complex task for plant managers, process engineers, and operators.
The need to make timely decisions to satisfy the competing interests of efficiency, quality, and cost—while adhering to ever-higher standards of safety and sustainability—puts immense pressure on people and systems. Often, the stringent expectations for today’s industrial operations go beyond the capabilities of many facilities, even those with advanced process control systems.
But there is a viable way forward. In this article we’ll examine the most pressing challenges for today’s industrial operations and how the creation of “digital twins” can address them head-on.
The key challenges for industrial operations today
As plants attempt to navigate an evolving operating landscape, we are seeing the following common challenges in our engagements with clients:
- Asset health and safety remains the number one concern for plant managers. Risks need to be continuously monitored, assessed, and managed. This becomes more challenging as facilities grow, age, or pivot, or the stringency of these requirements increases and the need to consider social and environmental impacts expands.
- Data silos created by the layering of systems without proper integration is leading to inefficient decision-making. As is the case with many plants today, a plethora of in-house and vendor-developed applications have been deployed independently over decades across different process areas to aid in production management, process monitoring, and operational control. Poor integration between these systems creates many data silos that result in a fragmented rather than holistic view of your industry value chain. Opportunities for better communication, analysis, and value capture from such things as integrated optimization of raw material procurement to logistics/inventory management, and production planning are lost. This lack of integration also limits the potential for new technologies such as Machine Learning (ML) and Artificial Intelligence (AI) to achieve the groundbreaking business transformations they’re capable of enabling.
- Skills shortages pose a substantial risk for the future. Due to aging workforces, more and more plants are encountering difficulties in maintaining and transferring process knowledge to the next generation. New hires are typically more technology-oriented and less attuned to “old-fashioned” ways of working. Plant engineers often get tied up fielding daily operating issues and thus little time and resources are available for longer-term investments in training, knowledge transfer, and performance improvement.
- Remote capabilities are becoming more critical. The industry has seen increased demand for remote monitoring, diagnostics, operations, and technical support. Remote capabilities are no longer a nice-to-have, they're an essential part of your business. Not only are they a key driver of things like safety and productivity under normal circumstances, but they're indispensable for responding quickly, safely, and with minimal down time to emergent risks like the global coronavirus pandemic we find ourselves in today.
What is a digital twin and why are so many organizations building them?
Over the past five years, the concept of the fourth industrial revolution or Industry 4.0—which can loosely be understood as the digitalization of manufacturing—has been an attractive subject for articles, conferences, and research studies. But it’s no longer just a buzzword. The convergence of market pressure, technology innovations, and industry-wide acceptance of digitalization is truly driving transformation in all industries.
The growing maturity of digital developments like the Internet of Things (IoT), cloud infrastructure, Extended Reality (XR) environments, and ML/AI tools are leading many organizations across a wide range of industries to create digital twins.
A digital twin is a dynamic digital representation of physical assets or systems. The representation is built from multiple modeling technologies (first-principal, data-driven, 2D, and 3D CAD modeling) together with real-time data and advanced data analysis tools. The result enables deeper, more advanced, and thus more meaningful insights into critical operations to improve human decision-making.
Where did the idea of a digital twin come from? Although the concept of a digital twin has been around for many years, the first clear definition was given by NASA [1] in 2010: “A digital twin is an integrated multi-physics, multi-scale, probabilistic simulation of a vehicle or system that uses the best available physical models, sensor updates, fleet history, etc., to mirror the life of its flying twin.”
Among many emerging disruptive technologies, creating digital twins is one of the first and most pivotal steps organizations—now including many industrial operations—are taking to unlock greater value for their businesses.
Three key pillars of a Hatch digital twin
In our practiced experience, an effective digital twin should consist of the following three key pillars: integration, intelligence, and interaction (see Figure 1).
- Connecting data through IoT integration. Picture a dynamic visualization environment where all sources of data from process operation, equipment maintenance, and product quality together with high-fidelity 2D/3D asset models are combined to create a digital version of what’s actually going on in real life. A real-time data exchange between the digital representation and its physical twin creates a Single Version of Truth (SVOT) to present on-demand, contextualized information and ensure data transparency and availability for all designated users through a secured cloud infrastructure.
- Connecting knowledge with built-in intelligence. Built-in intelligence is the key differentiator between the previous generation of business intelligence dashboards and a digital twin. The latter has the capability to consolidate various types and sources of knowledge in one centralized place, including data generated from other systems. A digital twin provides a fully and meaningfully connected analytical platform to effectively combine first-principle models, big data, and ML/AI technology to generate actionable insights and make intelligent decisions related to the physical twin’s safety, reliability, efficiency, and profitability.
- Connecting value using interaction. A digital twin adds value to business by creating new and more meaningful points of interaction with humans and machines through various services. Functions such as predictive maintenance, scenario analysis, augmented training, and remote expert support are just a few potential value-added services a digital twin can provide. The digital twin offers a more streamlined and more centralized model (see Figure 2) than the existing ISA-95 standard for the integration of enterprise and control systems. The digital twin approach breaks down silos
and promotes new forms of collaboration between operational divisions, functional departments, and external partners like OEMs and service providers.
Starting point on the digital transformation journey
With the advanced development of Industry 4.0, there is no doubt that more and more industries will transform their traditional operations towards ever-greater digital sophistication. Following the vision for shaping digital twins through integration, intelligence, and interaction, we have been making significant progress on their optimal design and deployment. Our experience indicates that the digital twin is an ideal starting point on the journey through digitalization. The improved transparency, connectivity, engagement, and decision-making it enables are the building blocks of a digitally sustainable business—one that can respond and adapt to the new and unexpected changes that are becoming increasingly commonplace for business leaders in a truly globalized society.
Several digital twin applications we’ve developed through collaborative partnerships are already proving the viability of this technology as a powerful disruptive tool. In our next article, we’ll talk about our development approach, some important lessons learned, and how digital twins are helping to address our industry’s most pressing challenges.
How are you using digital innovation to respond to your industry’s biggest challenges? Let us know in the comments below.
REFERENCES
- M. Shafto, M. Conroy, R. Doyle, et al., DRAFT Modeling, Simulation, Information Technology & Processing Roadmap Technology Area 11, NASA, Nov. 2010
