Walking the exhibit hall and attending educational sessions at CAMX 2018, it was hard to miss one of the recurring themes – industry automation. Dozens of companies displayed automated solutions, ranging from turnkey robotic solutions to automated inspection systems and computerized cutting machines. Educational sessions focused on Internet of Things (IoT) technology, Industry 4.0, artificial intelligence (AI) and smart manufacturing.

Here are four highlights from CAMX 2018 related to industry automation:

  • AI on the Production Floor – In the tutorial “Implementing Artificial Intelligence in Composites Manufacturing,” Avner Ben-Bassat shared opportunities, challenges and best practices. “The overall dynamics in composites today – whether it’s aerospace or other industries like wind or automotive – is that production volumes are going up,” said Ben-Bassat, CEO of Plataine, a creator of industrial IoT software for manufacturing optimization. “And while everybody is looking for economies of scale, what we’re really seeing is that actual unit costs are out of control.”

    Implementing AI on the production floor could offer a solution, with its ability to make quality projections based on autoclave data, assign jobs to specific machines, make tool maintenance recommendations, provide warnings about expired materials or misplaced parts and much more. If you think AI is the stuff of sci-fi, think again. Many companies have field tested the technology for composites production, including Airbus, Boeing, GE, Renault Sport and General Atomic.

  • Digital Twin of the Factory – “With IoT, I can collect a ton of data from products, from consumers, from analytical tools, from manufacturing equipment. What do I do with that?” asked Brench Boden at a session on IoT, Industrial IoT and digital twin of the factory. One answer is to use it to create a digital twin of your factory. Boden is a technical advisor to the Digital Manufacturing and Design Innovation Institute (DDMII), which partners with UI Labs and the U.S. Department of Defense to equip factories with digital tools and expertise.

    “We’re talking about building a representative model of an actual existing factory,” said Boden. “A digital twin is a simulation. It’s not enough to just record events and move materials. I want engineering and physics on equipment of interest to me.” Boden provided four reasons for creating a digital twin of factory operations. First, companies can use the model to assess rough design concepts for producability and predict yield coming out of operations. Second, it can help companies make proactive changes to schedules and resource allocation. Third, companies can collect data on individual processes to build libraries of data that may help improve aspects of the business. Finally, manufacturers can evaluate anomalous signal data for risk identification to head off cybersecurity issues.