5GEM use case: Ford optimizes quality with 5G and hybrid cloud

Manufacturing errors are expensive and time consuming, resulting in defective parts and machine downtime. Being able to efficiently capture, analyze and process data in real time on the shop floor helps manufacturers identify and fix issues before faulty parts start rolling off the production line. It can also enable them to perform predictive machine maintenance to minimize downtime.

Many factors can impact machine accuracy, including environmental ones such as temperature, the condition of the input materials, and the state of repair of the machine itself. Accurate, automated inspection and maintenance can be greatly enhanced by sensors connected to low latency 5G networks that gather data from multiple sources across the shop floor.

Cloud data management and edge computing play an important role in real-time data analysis, which makes these key technologies for the 5G Enabled Manufacturing (5GEM) project, a consortium of eight partner organizations: ATS Global, Ford, HSSMI, Lancaster University, TM Forum, TWI, Vacuum Furnace Engineering and Vodafone. Together we are exploring use cases for 5G private networks in different manufacturing environments, including real-time process analysis and control for laser welding of Ford’s electric vehicle components, and intelligent maintenance for welding machines and vacuum furnaces at TWI near Cambridge, UK.

The automotive industry performs automated, computer-controlled laser beam welding during the manufacture of multiple metal components, including battery tabs for electric vehicles. The technique allows for incredible accuracy and speed as well as enabling deep welds that would be unachievable by traditional means.

As part of 5GEM, Atlas from ATS Global is delivering a hybrid cloud platform directly to the shop floor using 5G networking and edge computing to collect data from laser beam welding equipment at high speed with low latency. Using the data and robotic process automation (RPA), Ford can make adjustments and improvements to the laser welding process without the need for costly downtime.

What is a hybrid cloud and why is it important here?

A hybrid cloud environment is one that combines offsite cloud servers with onsite infrastructure such as computers, sensors, edge devices and more. In this particular case, the most important elements are the edge computing devices; the 5G sensors that have been retrofitted to the equipment and our offsite servers.

With the 5G sensors able to collect and transmit data at near real time speeds, we can process information about individual welds every 20 milliseconds. That means quality assurance can happen during a weld rather than as a post-process activity.

Information about the welds is passed to edge computing devices where an initial ordering and understanding of the data happens on-site. This information can be passed back directly to the welding equipment with detailed data being sent to cloud servers for further data processing and future analysis.

How is edge computing helping this project?

Edge computing is the keystone that links the cloud servers to the onsite infrastructure. 5G technology is orders of magnitude faster than what has come previously, with lower latency than we’ve experienced before. However, to keep things as close to real time as possible we need edge computing.

The 5G network enables on-site data collection with no latency. That data can then be processed in an edge device that is constantly being monitored and updated by cloud technologies.

Preliminary data processing can take place within the edge device, making transmission of data to the cloud easier and more efficient. At the same time, our cloud-based data processing can continuously update these edge devices with new and better data models.

Insight based on the data processed at the edge, using constantly updated data models and understanding developed automatically in the cloud, can be fed directly back to the equipment as it runs. This means that, by using RPA, equipment can run, repeat and improve processes automatically.

By enabling this intelligent improvement of repeatable processes, quality control can take place during production, increasing uptime, productivity and the output of higher quality components.

How can these technologies affect the wider world of manufacturing?

With the 5GEM Project we have the opportunity to demonstrate how 5G networking can produce a new level of efficiency in manufacturing. With low-latency data collection combined with data processing at the edge, backed up by larger scale data modeling and digital twins in the cloud, we can create an environment where quality assurance is a part of the manufacturing process rather than apart from it.

What are the roadblocks and opportunities ahead?

As with any innovation project, the roadblocks are really only ever opportunities for learning. Much of what we’re testing here is new technology and augmentations to existing technology. As such, we can theorize how we want things to go but it's in the application that we see how accurate our theories were and how well they prove out in a real-world environment.

Any roadblock we encounter in terms of processes not working as well or as efficiently as we expected is an opportunity for us to improve our technology and understanding. With this project, we get to see how cutting-edge technology can become more advanced even as we’re using it. The joy of a hybrid cloud environment combined with edge computing is that the processes themselves are at once the test bed, the laboratory and the accomplishments.

The 5GEM project is supported by the UK government’s 5G Testbeds and Trials Program.