NET@PRO Digital Manufacturing Suite: the role of AI in making production data visible

What is NET@PRO and why mid-market manufacturing companies need it today

In Italian manufacturing, the real competitive advantage is not having more data — it is being able to read it while production is happening. NET@PRO was born as Italy's first MES (Manufacturing Execution System), later evolving into a Digital Manufacturing Suite developed by Qualitas to give manufacturing companies complete, real-time control over their operations. It is a comprehensive system that collects, processes, and distributes strategic information from operators, machinery, and external systems, transforming the factory from a black box into a transparent, connected, and proactive environment.

The NET@PRO MES solution is designed to turn shop-floor data collection into a continuous decision-making foundation: every production event is captured, contextualised, and made immediately usable — for those working on the line, for those planning, and for those governing. In this article we analyse:

• the context of Italian mid-market manufacturing and its current competitive challenges
• why production data exists yet often remains unused
• how AI can transform process data collection into preventive decision levers
• the key features of NET@PRO
• up-to-date figures on AI adoption in Italian SMEs

Italian mid-market manufacturing: excellence under pressure

The Italian manufacturing fabric is among the most articulated and competitive in Europe. Its beating heart is not the large listed groups, but mid-market companies: businesses with deep industrial culture, often family-owned, operating in sectors such as precision mechanics, metal processing, packaging, food, plastics, and wood-furniture. This is the context Impresoft knows and serves: companies with complex processes, variable job orders, high quality expectations from end customers, and the need to optimise every cent of margin in a market that does not forgive inefficiencies.

The current environment is challenging. The Italian Manufacturing PMI index closed 2025 in contraction territory (47.9 points in December 2025, source: S&P Global HCOB), with growing pressure on new orders and operating margins. The companies managing to defend their competitive position are those capable of doing one thing better than others: seeing what is happening in the factory, in real time, and acting accordingly.

The data paradox: plentiful, but unused

In mid-market manufacturing companies, the problem is not data availability, but its structure. Machines generate signals, operators enter information, management systems record events. But without a layer that collects them coherently and ties them to the production context (order, phase, machine, shift), this data remains isolated fragments. The result is a missing unified operational view: no-one truly understands what is happening on the line, or why.

The AI market in manufacturing: 2025 figures

This phenomenon is at the centre of the debate on Italian industrial innovation. According to data from the Artificial Intelligence Observatory of the Politecnico di Milano (2025), the Italian AI market reached 1.8 billion euros, with growth of +50% compared to 2024. The manufacturing sector is among those with above-average growth, driven by demand for analysis, optimisation, and process-prediction solutions.

Yet the gap between large and small companies remains wide: 71% of large Italian companies have already launched at least one AI project, while among SMEs the figure falls to 8% (source: Polimi AI Observatory 2025). A gap that is, however, rapidly narrowing — precisely starting from the manufacturing sector.

Factory productivity: from reactive control to continuous oversight

How productivity is managed in manufacturing SMEs today

In many mid-market manufacturing companies, productivity is still managed reactively: inefficiency is detected downstream of the process, when the cost has already been incurred. The problem is not just organisational, but informational: process data exists, but is not collected and correlated continuously along the production flow. Without a structured and contextualised data foundation, productivity remains a verification activity, not a governance one.

The introduction of advanced analytics models changes this paradigm: data collected in real time allows deviations to be identified, anomalies anticipated, and interventions made while production is still under way. According to the Confindustria report "AI for the Italian System — 2025", an analysis of more than 240 real-world use cases shows that in manufacturing, AI applications go well beyond predictive maintenance: the most impactful include early detection of defect trends and the correlation between process parameters and business productivity.

NET@PRO: the Digital Manufacturing Suite for mid-market manufacturing

What NET@PRO does

NET@PRO is built to do one thing in a structured way: collect, normalise, and contextualise shop-floor data. Every event — production advance, machine stoppage, process parameter, operator action — is captured the moment it occurs and linked to the correct production context.

All operational features are built on this unified data foundation: quality control, traceability, KPIs, ERP integration. The solution covers the entire productivity perimeter in an integrated way:

Automatic data collection and production advance. Real-time acquisition of data from machines and operators, with precise monitoring of cycle times, machine states, and the progress of production phases.

Lot and component traceability. Reconstructs the complete history of a product at any moment, making every response to audit requests, certifications, or customer complaints fast, precise, and documented.

Non-conformity management. Guided workflows for root-cause analysis, definition of corrective actions, and verification of effectiveness — all in a single system, without spreadsheets or informal communications.

Real-time process parameter control. Configurable alerts on critical thresholds before defects materialise in the product. No waiting for end-of-line: intervention happens during production.

Integrated production KPIs (OEE, scrap, rework). Quality and production speak the same language. OEE is not an isolated number, but an indicator connected to the real causes of losses.

Native ERP integration. Quality data flows automatically into job order and cost management, eliminating duplication and information delays.

The interface is designed for those who work in the factory: not a boardroom dashboard, but an operational tool, accessible from any device, engineered to reduce the friction between the machine and the decision.

AI in NET@PRO: from data to operational intelligence

From data to decision: AI scenarios in the factory

The integration of AI capabilities within NET@PRO enables practical applications that mid-market companies can implement without overhauling their processes or requiring in-house data science teams.

1. Predictive detection of quality anomalies. Machine learning models continuously analyse production parameters and learn the patterns that precede a quality anomaly. The alert arrives before the defect is physically present in the part, reducing scrap and upstream rework.
2. Automatic causal correlation. When a non-conformity emerges, AI helps identify the root cause by relating variables that an operator would struggle to correlate manually: a supplier change, a temperature variation on the line, an out-of-specification raw material batch.
3. Natural language querying (GenAI). Production managers can query the system directly: "What was the main cause of non-conformities last week on product code X?" — and receive a contextualised answer based on real plant data, without opening reports or consulting IT.
4. Intelligent OEE. OEE stops being a descriptive metric and becomes a governance tool. AI connects availability, performance, and quality, identifying the relative weight of each component on inefficiencies and indicating where to intervene as a priority.
5. Operational production optimisation. AI analyses production patterns, timings, and operational sequences to identify bottlenecks, suggest priorities, and improve plant utilisation, reducing inefficiencies and downtime.

Why Italian manufacturing SMEs cannot wait

The most common objection is: "AI is for big companies." The 2025 data says otherwise. Among medium-sized enterprises (50–99 employees), AI technology adoption has almost tripled in a single year — from 5.6% to 14% (source: Istat "AI 4 Italy" 2024). And 47% of companies that have already adopted AI report productivity increases of more than 5%.

The gap between those who have adopted AI tools in production and those who have not is widening. Its consequences are already measurable today: lower scrap rates, better delivery punctuality, more satisfied customers, defended margins.

In companies operating with small batches, high product mixes, and variable supply chains — the typical scenario of Italian mid-market manufacturing — the ability to correlate events of different natures and respond in real time is not an additional competitive advantage: it is a condition for survival.

Conclusion: the factory that sees itself improves

The factory is not governed at the end of a shift, with the next day's report. It is governed while it is happening: when every piece of data is collected at the right moment, linked to the correct context, and made immediately readable.

In this scenario, quality is no longer a separate activity: it is the natural effect of a production process that can finally see itself. NET@PRO is Impresoft's answer to this need — built for Italian mid-market manufacturing companies that want to move from reactive production management to continuous, data-driven governance.

Because the factory that truly manages to see its data — not just collect it — is the one that builds quality before a defect is ever born.

Would you like to discover how NET@PRO can integrate with your production system? Contact an Impresoft expert and request a personalised demo.