Rethinking How Operators Interface With the Line

Estimated reading time: 1 minute

Jurgen Schmerler, CEO of WaveOn, reveals how AI and large language models are revolutionizing electronics manufacturing. By integrating AI with machinery, operators can access real-time, multimodal information for troubleshooting and maintenance, significantly reducing training time and enhancing efficiency. He discusses the industry's challenges, the customizable knowledge bases, and the future of proactive maintenance and process control.

Nolan Johnson: Jurgen, how does the knowledge transfer work you’re doing fit inside electronics manufacturing?

Jurgen Schmerler: With the advent of large language models, we can communicate with machinery almost like with humans. The machine comes with an inherent set of knowledge, typically in the form of standard operating procedures (SOPs) or manufacturer user guides. Also, the experience from other users can go into how to operate a piece of machinery.

To get the operator to proficiency, we propose a solution based on AI. These large language models digest the information available to answer questions in natural language using existing information graphics, diagrams, and even videos in the answers. This information can help perform troubleshooting steps or complex maintenance procedures and guide them through the steps. Users get just-in-time knowledge delivery, which is up to date on the latest ways to operate a piece of machinery. That's the premise of the idea.

Johnson: Let’s say an EMS company is running an assembly line, and all the equipment is tied together end to end, when suddenly, one unit starts to show the amber status light, asking for operator attention. Is that where you become involved?

Schmerler: That could be the troubleshooting use case. If something goes wrong, you have to perform a certain procedure to troubleshoot. But we would get involved even earlier in the process.

From a macro level, we have a big retirement wave coming up in the industry. Experienced operators will need to be replaced. Also, as machines get more complex, you need operators to get up to the job quickly and efficiently. We are working with the machine manufacturers, providers, and vendors who have a lot of knowledge. We are adding the knowledge of the experienced operator to a knowledge base.

To read the entire interview, which originally appeared in the June 2025 issue of SMT007 Magazine, click here.