09/09/2026 | Press release | Archived content
In day-to-day production, getting the right answer at the right moment is everything - nobody can afford machine downtime. The ENGEL Virtual Assistant (EVA) understands questions asked in natural language, places them into the context of the machine, draws on relevant expertise, and points to the next step. And it doesn't stop there: once the operator confirms, EVA can pass approved changes straight through to the machine control system - moving beyond a simple question-and-answer tool to a genuinely capable assistant that sets itself apart from conventional chatbots.
Feeling a sense of déjà vu? ENGEL's autonomous injection moulding cell at Fakuma 2026 looks just like last year's. But the value it delivers to customers keeps growing, step by step. This year, trade visitors can experience first-hand that the machine doesn't just respond - it acts. In their personal lives, most people already turn to AI by text or voice for everyday questions. Now, that same kind of dialogue is making its way onto the shop floor. Operators simply describe the issue, and once they confirm, EVA carries out the approved action itself. This is how ENGEL is turning artificial intelligence into measurable production performance.
Artificial intelligence is becoming a tool with genuine practical value in daily production - for the part, the process, the machine, service cases, and everyday decision-making. Every production environment is different, shaped by the machine and mould, as well as by material, order and quality requirements. EVA takes this individual context into account and follows a simple principle:
Understand: Operators ask their question in their own words - by text or voice. EVA places it in the context of the machine and the current job.
Interpret: The data is linked to ENGEL's machine, process and service expertise.
Explain and recommend: EVA makes complex relationships easy to understand, and suggests the next sensible step is.
Act with control: EVA prepares approved changes and, once the operator confirms, passes them on to the machine control system.
The technological foundation for these new capabilities is the ENGEL AI Interface. It makes machine context available in a controlled way for approved ENGEL use cases, relying in part on the Model Context Protocol (MCP), which provides machine information to AI applications in a structured, context-aware format. It is available for every new ENGEL and WINTEC injection moulding cell - as an optional feature on new machines, and as a retrofit for existing machines equipped with a CC300 or C3 control unit. This means customers can plan the virtual assistant as a standard equipment option for their production.
The cell on show at Fakuma 2026 embodies the bigger vision behind inject AI: automated, self-regulating production that, in future, could even run lights-out. To achieve this, ENGEL is bringing machines and digital solutions together into a single system via EVA and the ENGEL AI Interface, making AI increasingly usable in real production environments. Working together with customers and selected partners, ENGEL is developing new use cases, validating them step by step, and introducing them into real-world production settings.
"The skills shortage in production really bites when a fault occurs and the right expert isn't available. At the same time, valuable knowledge is often held by just a few people, and not always available in the language you need. EVA brings ENGEL's expertise directly to the people on the shop floor - around the clock, in plain language, and tailored to the situation at hand. That means faster problem-solving, more targeted process optimisation, and production that keeps running reliably."
"We're seeing AI change the way people work and access knowledge. Asking a question and getting an instant, understandable, context-aware answer is fast becoming the norm. With inject AI, we're bringing that shift into injection moulding production. Our goal is to give users a tool that makes them more efficient in their daily work - from simple knowledge queries, through support during faults, to optimising the machine itself."