Future behavior unlocked: Digital twins enable next level automation

Digital twins pave the way for a greater use of collaborative robots, new research concludes. The study has been acknowledged in the academic community and has given the Danish company LINAK A/S a foundation to further automatize their factories. The project was cofinanced by LINAK A/S as part of the research and innovation platform MADE Digital and with University of Southern Denmark as the academic partner.

Normally you would spend many weeks building, tuning and optimizing new production lines which is a more costly and a high-risk approach to designing factories.

A digital twin offers the possibility of building and running in new production facilities virtually. A digital twin is used to simulate and optimize the design, test, and running in of manufacturing processes and operations.

This saves time, materials, and costs and enables fault and failures to be predicted before they happen in the physical world.

A Cobot demonstator with huge potential

With a focus on collaborative robots (cobots) and visual simulations using digital twins, Ali Malik demonstrated how the factories of the future can be designed and how robot systems can be made available as ‘colleagues of humans’.

Read more about MADE DIGITAL

MADE DIGITAL

In april 2017 Ali Ahmad Malik started his Ph.D. as part of MADE platform MADE SPIR, where he worked on flexible automation. He has worked on building an assembly cell prototype to quantify and demonstrate how LINAK’s operators and collaborative robots can be integrated at LINAK with the use of digital twins. The work with digitalizing supply chains is something that will further strengthen the Danish manufacturing industry.

The human-robot assembly cell is developed to replace a manual assembly cell. Data connected simulations are used as a digital twin from design to operation for validation, estimation of cycle times, process balancing, robot program generation, and human-safety assessment.

“I was focused on increasing the level of automation in processes that are conventionally difficult to automate. An example of this is assembly in manufacturing which, besides its economic significance, is often difficult to be automated through robots, due to the variety and complexity of the task,“ says Ali Malik on his work with LINAK, the Danish company that manufactures actuators from their factories all around the world.

Together with experts from The University of Southern Denmark (SDU), Ali has developed a collaborative assembly system to produce actuators to showcase the potential of collaborative robots and digital twins.

“The assembly system was aimed to be developed with a higher level of engagement between human and robot and also to utilize digital twins from design to development,” says Ali, who is now an assistant professor at SDU.

The assignment was to provide LINAK with insight to, how they can achieve next level automation regarding high-mix low-volume manufacturing operations, and to create a cost-benefit analysis, so the company has a better overview of potential limitations of collaborative robots.

”The main goal was to boost our competitiveness. Ali has been a part of this development that has given us a better understanding of how and when to use cobots. Where does it make sense to automate and how to share work in production between human operators and robots,” says Martin Vejling Andersen Senior Manager, Logistics & Production Development at LINAK.

The benefits of using digital twins

Cobots have the potential to automate nearly 50 % of the assembly tasks in the factory according to Ali Malik’s research.

“But if robots are desired to be used in close proximity to humans in a variant oriented production, it becomes challenging. They need to be easy to implement, reconfigure and safe for fellow humans,” Ali Malik explains.

A collaborative assembly system needs to be agile, adaptable and safe for human colleagues. In the industry there are growing market challenges and desired product varieties. A human-robot collaborative assembly system can be managed with a digital twin which enables you to run multiply scenarios and optimize the system.

“This makes it possible to see the future behavior of the real system and optimize it based on the results achieved. Change in production parameters are integrated in the digital twin enabling you to predict future behavior. The behavior can be visualized with 3D-models and results are assessed without the risk of any financial loss or human injury,” Ali Malik explains.

Future factories need to be immune to global crises

How can factories of the future be designed with a digital twin-approach?

“Future factories need to be resilient and responsive. They also need to be immune to global crisis e.g. COVID-19. The solutions often discussed in Industry 4.0 domain have the potential to address the challenges but also tend to increase complexity of the production floor. But the advancement in virtualization, sensing technologies and computing power has evolved the concept of digital twins to design, develop and control complex production systems,” says Ali Ahmad Malik.

How will your work benefit the industry?

“Although, during the last years the industrial landscape has enjoyed great success in using cobots for a variety of tasks but, I believe, the potential is much higher than what has been achieved. Still, the deployment of cobots in assembly is complex, time consuming, difficult to reconfigure and the deployment time is often long,” Ali judges and adds:

“A digital twin as a virtual copy of the assembly system is conceived earlier than a physical system and thereby makes it possible to visualize, experiment and test the cobot system well before making any investments. Besides fast and reliable decision making, the virtual twin generates robot programs, helps in online collision optimizations, safety validation and makes live interfaces between the physical system and the digital system for data backed optimizations throughout the system lifecycle. It also makes it possible for an operator to use virtual reality (VR) headsets and ‘get into’ the scene and interact with the robot, and equipment in a virtual environment.”

Ongoing work

The researcher from Pakistan will stay in Denmark as he is to assume a new role as an Expert for Robotics and Automation at the MADE-member Siemens Gamesa Renewable Energy.

“It is an exciting role where I will explore robotic automation and their development through data enabled simulations,” he says.

Even though the project with SDU and LINAK is finalized, the work is still ongoing. Several publications have been made such as in Journal of Computer Integrated Manufacturing, Journal of Industrial Robot and Journal of Assembly Automation. A recent development together with The University of Cambridge has just (June 2020) been published in Science Robotics.  

The same can be said about LINAK: The global company with headquarters in Guderup has in parallel to Ali’s study been addressing the challenge of achieving next level automatization internally – and the company has developed a guideline that answers the question: Where and how can we profit from implementing cobots?

“We have seven cobots in production globally out of approximately 80 robots, and I believe that number will increase,” says Martin Vejling Andersen Senior Manager, Logistics & Production Development at LINAK.

The company will closely monitor how the technology evolves, and the guideline that Ali was a part of will dictate LINAK’s next steps of becoming factory of the future.

Other news

Article
Article
Article
Article
Article
Video