Digitally Designing the Factory of the Future

Factory design and simulation is at an advantageous crossroads as more digital tools are available to build efficiency and economy into their design. This includes creating the plant layout, optimal operations and other facets of the workflow of today’s modern manufacturing facilities. 

Modeling and simulation, including an avid use of digital twins, put precision and accuracy into plant design that has not always been available in the past. 

Stephen Hooper is the vice president of design and manufacturing of Autodesk. Autodesk recently acquired FlexSim, a factory simulation technology company focused on the design and modeling of production lines and facilities. 

“There are quite a few individual technologies that today, and increasingly in the future, have the power to disrupt the design-and-make industry, putting factory simulation at the fingertips of any operator,” says Hooper. “The drastic positive disruption these technologies bring to manufacturing, however, comes when they’re combined. 

“More specifically, any one of several modern technologies—such as affordable access to point cloud scanning from mobile devices, immersive [virtual reality/augmented reality (AR/VR)] multimedia experiences and powerful Internet of Things solutions being retrofit into existing factories and their machines—will increase the efficiency of manufacturing,” Hooper says. “But plugging several of these technologies together is where you get the real factory simulation payoff.”

According to Hooper, the ultimate Holy Grail of factory simulation (which Autodesk is working toward) is artificial intelligence (AI)-powered generative design of factory layout. Optimizing a factory using AI and machine learning (ML), much like optimization of a product design, is a multivariant puzzle that requires balancing numerous trade-offs, Hooper explains. 

“Are you aiming to minimize waste and energy consumption for sustainability or maximize machine usage for greatest return on resource investment, or perhaps quickest throughput is your priority, or highest possible quantity of factory output?” he asks. “Optimizing for just one [parameter], of course, may throw the others way outside acceptable parameters. Balancing all of them, and more, is what you really want, but doing so can be very complex. However, offering solutions that balance a multitude of variables is exactly what generative design does best.”

Value of Digital Twins 

Ray Wong is the senior director of business development at Twinify Technologies. The company specializes in building digital twins for industry to help consummate the digital transformation so many companies are seeking. Digital twins incorporate all the comprehensive inputs into the production line, including labor, fuel, environmental goals and supply chain inputs. 

“There is a lot of focus on what happens at the assembly/production line, but what’s often overlooked and is equally important are the assets that support the manufacturing operations,” says Wong. “Power generating assets such as combustion turbines and solar panels, which power the facility are an example of this.” 

Wong explains how recently manufacturers have faced traditional challenges such as supply chain disruptions, quality control and labor shortages, as well as energy cost hikes and sustainability demands. 

“There are growing pressures for manufacturers to adopt sustainable practices, especially with consumers being more eco-conscious and preferring to do business with companies who share their values,” Wong says. “Additionally, transitioning to greener processes and the greater use of renewable energy sources lowers total costs of goods for products that are being made.”

He also cites an increase in environmental regulations, which impacts compliance related to emissions from plant manufacturing and energy production processes.

Wong explains that there are three types or functional focus of digital twins: design, operational and process. 

Twinify Technologies builds operational and process digital twins of energy production assets and that these simulations provide a strategic advantage to manufacturers by providing them with a past, present, and future view of their power generating assets. 

“Digital twins capture crucial insights and turn those insights into outputs that experts can easily understand and implement,” says Wong. “Advanced simulation and analytics allow plant managers to understand if they have enough power production to meet requirements of the manufacturing line, operate more efficiently, and gain insights to quickly adapt their approach. Additionally, these simulations can help by providing predictive analytics such as the ability to inform an asset manager that their equipment needs an overhaul sooner than expected.”

Colm Gavin is a portfolio development manager with Siemens Digital Industries Software. Gavin says that for production machines a digital twin is used to validate and optimize the machine before it is physically constructed by validating its automation and electrical performance against a kinematic model. 

“This ‘virtual commissioning’ of the automation logic means the various disciplines in your development department can work in parallel (instead of sequentially) and test almost all functions using the digital twin,” says Gavin. “The disruptive aspect of virtual commissioning forces these different disciplines to work closer together, requiring a more accurate mechanical model for the exact placement of sensors and the correct nomenclature used in the naming of the I/O signals used in the 3D model, electrical schematics and PLC [programmable logic controller] program.” 

Gavin explains that in some use cases it can expose weak business processes in a company, for example, the hand-off between the mechanical, electrical and automation disciplines. 

“However, even though more upfront effort is needed in the creation of these digital twin models, by working in parallel errors are identified and rectified at an early stage, reducing the overall costs, especially if rework is required for machines already in production,” Gavin says.

A Bright Future 

Kuruvilla Mathew is the chief innovation architect with UST. UST specializes in helping companies with their digital transformations in different capacities.

He says that simulation technologies are already playing a role in improving yield, running several “what-if” scenarios. 

“Factories can review the various manufacturing lines, the packaging lines, and the raw material staging areas, including other complementary functions in the factory,” says Mathew. 

Simulating factory functions digitally is key to improving effectiveness on a factory floor, according to Mathew. 

“Expanding a specialized set of scenarios like standard operating procedures (SOP) for a manufacturing a part or product, with [the] right types of 3D simulations, help understand gaps earlier in the process,” Mathew explains. “Extending this into mixed reality with first-, second- and third-person view helps improve the deeper understanding of the equipment, the process that overall helps create a safe environment for factory workers and address equipment damage. 

“With the progression of AI, the reduced cost and availability of GPU hardware to run the AI models, the improvements in 3D simulations, XR and sensor technologies have accelerated the 'What If' simulations to a completely new level,” Mathew adds. 

He notes that companies that do not leverage these simulations may not realize the gaps earlier in the process, noting how 3D simulations help reduce the gap.

“Simulation technology will play a key role for companies in their path to efficiency and productivity,” says Dheeraj Vemula, who works in the digital twin business development for Altair. “The demand for manufacturing lines is more than ever and there is little room for mistakes, which can be expensive and time-consuming. The only way to ensure that these lines are performing at their best is to have well validated simulations far ahead of the building process. This can help companies minimize the expensive costs associated with changes in production. In addition, this will also help them reduce time to market which provides a competitive advantage.”

He adds that technologies like digital twins are helping companies take simulation into production.

“By creating a virtual model of the real-world process or product, companies can get real-time insights about the operations on their factory floor and take the necessary corrective actions,” he says. “This helps manufacturers reduce material wastage, and improve up-time and efficiency, which is not only beneficial for their bottom line but also enables a sustainable approach to production processes.”

Vemula notes that one of their customers’ digital twins helped reduce production waste by more than 15%, reduced simulation time from hours to seconds, and enabled the company to monitor its forming process in a real-time environment.

The Road Ahead for the Factory of the Future

Suffice it to say that digital technologies that incorporate many new capabilities are part of a confluence of maturing technologies that should have a significant input on the finished look and results of manufacturers that commission world class design of such facilities and production lines. 

“Digital thread, AI/ML and digital twins are technologies that will be disruptive for factory simulation and design,” says Altair’s Vemula. “With applications such as predictive maintenance, virtual sensors and what-if studies there is a tremendous amount of value to be gained from these technologies. The manufacturing and operations industry has invested heavily in technologies like Industry 4.0 and IoT however, the complete value of such technologies is not yet realized. By using the aforementioned technologies, companies can reap additional value like avoiding unplanned downtime and reduced maintenance costs.”

More Altair Coverage

More Autodesk Coverage

More Siemens Digital Industries Software Coverage