3 smart factory trends that will boost productivity

When a system fails on the factory floor, each second of downtime equals dollars down the drain – about $22,0001 per minute for some automobile manufacturers.

With those stakes, advances in smart factory technology that enable efficiency, advanced machine-to-machine connectivity and high-speed communication – down to the microsecond – can't come fast enough.


  • A beverage factory that uses the same assembly line to fill bottles with different drinks.
  • An auto manufacturer with a modular production cell that can build different types of cars on the same line with near-zero downtime.
  • Alerts that tell technicians about potential part and system failures before they happen.
  • Machines that can sense objects and avoid collisions work collaboratively with humans.

“The factory of the future will be highly efficient and highly connected," said Thomas Leyrer, a system architect at our company. “Some of the latest innovations drastically improve communication while addressing increasing bandwidth requirements.”

 Design what’s next for Industry 4.0

Here are three trends adding intelligence to Industry 4.0:

1. Advanced industrial communication enables predictive maintenance

If the smart factory has a calling card, it's the level to which it has pushed machine-to-machine connectivity and communication – enabling a host of other capabilities.

While gigabit Ethernet time-sensitive networks (TSN) increase connectivity and the speed of data pinging between manufacturing devices, technologies like IO-Link and Sitara™ AM6x processors can harness that data from the factory floor and decipher it in real-time. Industrial Internet of Things (IIoT) related applications allow technicians to anticipate part and system failures before they happen and improve subsequent generations of product development.

“If a certain type of machine is deployed in 50 different locations, for instance, technicians can now compare their performance and control for variables like humidity, power supply and other environmental data,” Thomas said. “When one parameter for an individual machine goes out of limit, it triggers an alert signal to do predictive maintenance—remotely in the case of a software upgrade and onsite in the case of part repairs or replacements.”

2. Machine vision and human-machine interaction increase quality

Cobots – or collaborative robots designed to work alongside humans – represent one of the fastest-growing market segments in robotics, projected to reach nearly $9 billion by 20252. “These sophisticated machines can detect the proximity, speed and location of people or objects in defined zones through TI mmWave radar, giving robotic arms “vision” to safely help workers load machines or pick components out of bins, for example.

Machine vision can also enable greater product quality by testing tolerance, dimensions and other material attributes. “Now you can integrate quality assurance, which is typically done at end of the production cycle, as an integral part of the production process," Thomas said. “When you put machine vision into a TSN, it increases efficiency with identifying badly produced products early."

3. Edge analytics promotes efficiency

 On the factory floor, some critical movements can’t wait for machine learning in the cloud. Instead, they demand insights and decisions closer to action – such as a robotic arm that needs to maneuver around workers to do its job. Edge analytics puts intelligence and decision-making capability right into the robot arm.

“Edge analytics can improve overall efficiency in real-time, allowing technicians to measure and analyze the power consumption of each individual device and adjust it when it’s not operating,” Thomas said. “Edge devices also give users access to data that enables continuous monitoring of the efficiency and functionality of production cells remotely.”

Most modern factories are already benefiting from smart technology and IIoT to some degree. In the factory of the future, smart technology will add flexibility and modularity to even the most efficient single production line.

“With predictive maintenance alone, you can increase up-time from 80% to 95%," Thomas said. "That's a big deal."

1.    https://iiot-world.com/connected-industry/the-cost-of-one-minute-downtime-in-manufacturing/
2.    https://www.assemblymag.com/articles/94462-global-cobots-market-could-be-worth-9-billion-by-2025