Zerynth is an Italian deep-tech company that develops an IIoT platform for the rapid and scalable digitalization of industrial production. The company is currently evolving its solution into an Industrial AI Copilot platform, equipped with a conversational agent that enables natural interaction with production data and processes. The development of this artificial intelligence component is considered strategic for the evolution of the product and the company’s future positioning.

Modern factories are gold mines of data.

IoT sensors capture every vibration, energy consumption pattern, and production cycle. However, this information often remains “silent” or accessible only to those who know how to interpret complex software tools.

There is also the issue of documentation. User manuals, standard operating procedures (SOPs), and technical guides are often confined to static PDFs hundreds of pages long, making them difficult to consult when information is urgently needed.

At the core of the project is an advanced digital assistant that does not simply “answer” questions but actually reasons over the data. Unlike traditional chatbots, the system relies on an Agentic RAG (Retrieval-Augmented Generation) architecture.

In practical terms, this means:

Real-time data access: The system directly queries the platform’s APIs to retrieve KPIs related to machines, energy consumption, and efficiency (OEE)

Intelligent document consultation: Through a vector database, the Copilot “reads” PDF manuals and provides precise answers, citing the exact source and page.

Memory and context: The agent remembers the conversation. If a user asks “Compare it with yesterday”, the system understands exactly which machine or metric is being referenced.

The experimentation placed strong emphasis on data sovereignty. The core engine relies on Azure OpenAI models to ensure compliance with European regionalization and data privacy requirements, orchestrated through advanced frameworks such as Agno and LangGraph.

One of the system’s key strengths is Explainable AI (XAI). Every generated response is not a “black box”: it includes transparent references to the content used to formulate the answer, ensuring that technicians can always verify the accuracy of the information provided.

During testing, the integration of the platform enabled several critical use cases:
1. KPI analysis: Immediate responses on availability, performance, and quality across production lines.
2. Technical support: Automatic opening of support tickets via chat when anomalies are detected.
3. Energy efficiency: Comparative analysis of energy consumption across different periods to identify inefficiencies.

The results are significant. The ability to reduce machine downtime (up to 70% in optimized contexts) and drastically cut the time required to find technical information makes AI an essential tool for the competitiveness of manufacturing SMEs.