While many people work from home or face extended furloughs during the Covid pandemic, essential manufacturing operations must continue, and the so-called “smart factory” presents a way to help manufacturers keep their doors open to support customers.

“The crisis has accelerated the need for smart factory initiatives,” said Steve Shepley, U.S. aerospace and defense consulting leader at consultancy Deloitte, during an FIA Connect session held on Wednesday. Deloitte smart factory leader Lindsey Berckman joined Shepley for the session, titled Implementing the Smart Factory - Accelerate the Journey to Digital.

It turns out that the benefits of a smart factory have become even more apparent during the pandemic. By connecting factory assets such as machinery, IT infrastructure, inventory, people, and processes, smart factories help manufacturers make products more efficiently and provide the ability to predict bottlenecks and shortcomings, all the while freeing people up to do more important work. During the pandemic, Shepley explained, the concept can make it easier to let people work remotely while keeping the factory humming. “People can be safer, and technology can drive safety in the factory,” he said. “You’re connected to supply chains so you can deal with issues before they hit the [factory] floor.”

“The key is how they come together,” said Berckman. “To have a valuable predictive smart factory, those attributes come together so workers act on signals and make worthwhile decisions.”

Implementing a smart factory involves more than just deciding to do it and hiring consultants. It is important to consider the use case and whether the organization could benefit from a smart factory. One critical factor, Shepley noted, is a high technology readiness level, in which case “the comfort level to implement a smart factory is quite high.”

Companies that might be ready should consider the following factors, he said:

• Focus on value and results and don’t get mired in technology and hope that technology will create the solutions to your problems. Determine the burning issues. Laser focus on the desired outputs.
• Think about user experience. If you want people to adopt new solutions, you must make it easy to use. Don’t just involve technologists in solutions but get front-line workers involved and get their feedback into the design.
• Establish an ecosystem. This is not a silver bullet. Bring in partners, take full advantage of the cloud, use sensors to bring in data, all on a common platform or ecosystem of technology.
• Build a roadmap of use cases. A larger basket of implementations allows a company to hedge the risk over a larger population of processes. 
• Plan for scale. You might pilot the smart factory with one use case, but don’t build on an environment that doesn’t scale. If you can’t scale larger, it’s not worth it. Think big, start small, but scale fast once you see results.

“This isn’t the only way to do a smart factory,” Shepley said. “But a cloud provider is critical, particularly for scale. It allows you to put all the data into a single structure and gain benefits at one plant and apply them to another facility.” Deloitte worked with Amazon Web Services to create Smart Factory Fabric, a pre-configured suite of cloud-based applications to facilitate development of smart factories. “They’re willing to partner and they have a spirit of innovation, which allows the solutions to grow with the manufacturer,” he said. "It gives us a chance to continue to innovate.”

In one example of a smart factory that Deloitte helped establish, a U.S. aerostructures manufacturer faced a big problem with material shortages and getting workers aligned with the work that needed doing. The company had planned to build new facilities to improve the situation, but the smart factory eliminated the need for a new plant.

Berckman explained that the first step centered on helping the company improve asset tracking. “Workers felt like they were on a treasure hunt to get what they needed,” she explained. Deloitte helped the company create RFID tagging and more comprehensive tracking to improve visibility and tracing of parts.

The next step involved making sure workers were in the right place at the right time. The smart factory setup included a dynamic scheduling system that replaced whiteboards and sticky notes filled with tasks and made it much easier to reschedule tasks and reassign workers efficiently.

It was important to include the employees in the design of the smart factory, she added. “The feedback was incredible.”

The result was a vastly more efficient factory, and no more looking up tasks on sticky notes, plus the company managed to apply the smart factory technology to many other use cases, further improving efficiency.  

It takes about 16 weeks to implement smart factory processes, according to Berckman. The exercise includes about two weeks developing the business case, six weeks building a prototype to solve the particular problem at hand, then eight weeks running a pilot to prove the value of the system. Once the first process gets created, it can expand to other areas of the factory. “Think about the next use case to give better lift, efficiency, and metrics,” she said.