Laurent Couillard, founder and CEO of InUse, emphasizes that the combination of generative artificial intelligence and data processing algorithms represents a true "game changer" in production. He explains that "restart time following an incident is a key factor in increasing a line's OEE (Overall Equipment Effectiveness). However, as several of my clients have pointed out to me, this time increases considerably when the outage occurs at night. Why? Because the manufacturer's technical support is unavailable. Restarting a line requires two things: a diagnosis and a problem resolution procedure. It is in this second area that conversational assistants are invaluable, as they provide the information the maintenance team needs as clearly as a human could on the phone, and in multiple languages if necessary. With these tools, it becomes possible to have a dialogue with the machine."
Laurent Couillard believes that "it makes perfect sense for AI assistants to be developed by manufacturers and integrated into the machines. They have the best knowledge base since they record all their service tickets." He adds that these conversational assistants also benefit the manufacturer's after-sales service technicians: "It changes their lives in the field. And it gives them a real reason to document every intervention in the company's information system." Initial observations from InUse's clients indicate that the development of tools using generative AI could reduce technical support calls by 30%
He encourages manufacturers to integrate AI into their businesses, provided there is appropriate oversight: “If you develop an application within a well-defined context, limiting access to a closed knowledge base, there is no risk of the AI going hallucinatory. If it doesn't have the answer to a question, it will tell you. And you can ask the same question a hundred times and get the same answer a hundred times,” he explains. The reliability of the troubleshooting instructions provided by the AI is indeed just as crucial as the accuracy of the fault diagnosis
Regarding the integration of chatbots on production lines, Laurent Couillard emphasizes two aspects: “It’s important to be able to geolocate the question, because this allows you to find the answer in the right knowledge base. On a single line, you might have five machines with error code 42; you need to be able to use the correct procedure in the right place. I think we also need to provide end users with a prompt engineering function; this allows them to develop autonomy over their applications by building assistants adapted to each context and each role.” He anticipates that intelligent agents will soon appear in factories: “AI agents will soon be able to analyze the question asked and chain the different operations to be performed, both in terms of data and information retrieval, to formulate the appropriate answer. Tomorrow, an operator could even use them proactively, for example, when starting a batch, by asking them what problems they might encounter.”
