Reflections on AI in the Supply Chain: Contributions by Gregorio Piccoli (CTO of Zucchetti) and Alice Kirchner (Commercial Director of Cybertec)

When talking about Artificial Intelligence, the dominant narrative is often extreme: systems that know everything, predict everything and that, sooner or later, will completely replace people. It’s a fascinating image, but not a very useful one for those who work every day in the real supply chain — made up of operational constraints, incomplete information, unexpected events and decisions taken under pressure.

It is precisely from this awareness that Zucchetti’s approach to Artificial Intelligence applied to the Digital Supply Chain was born. As Gregorio Piccoli, Chief Technology Officer (CTO) of Zucchetti, explains: “Artificial Intelligence must first and foremost be a concrete technology, capable of functioning within the operational reality of industrial processes, not a theoretical concept or a marketing promise.”

There Is No “One AI”: There Are Artificial Intelligences

The first misconception to overcome is conceptual. There is no single Artificial Intelligence, but many different intelligences, each designed to perform specific tasks.

The supply chain is a complex system that includes production, logistics, maintenance, warehousing, transportation, and planning. Each area has different data, rules, and objectives. Believing that a single AI can govern everything means ignoring this complexity. “If we have a health problem, we go to a doctor, not a lawyer — even though both are intelligent. In the same way, every process requires the right AI,” emphasizes Gregorio Piccoli.

Myths to Overcome About AI in the Supply Chain

One of the most common mistakes is imagining AI as an infallible technology, always correct and capable of eliminating all uncertainty.

Reality is different. We live in the age of prediction, not certainty. The supply chain is built on incomplete information: the context changes rapidly, many events are unpredictable, and decisions often need to be made under pressure.

A single truck accident can block a delivery and compromise a production plan. No algorithm, no matter how advanced, can completely eliminate this type of uncertainty.

For this reason, thinking of AI as a total replacement for the human component leads to fragile models. The real value instead arises from collaboration between people and technology.

An Ecosystem of Collaborating Intelligences

In the modern supply chain, there is no single AI that does everything. Instead, there is an ecosystem of specialized intelligences working together, including:

• Logistics and transportation flow optimization

• Dedicated forecasting models
• Optimization algorithms
• Simulation systems
• Predictive maintenance and anomaly detection analytics
• Logistics and transportation flow optimization
• Workforce optimization

“The true value of AI emerges when multiple intelligences collaborate with each other,” says Gregorio Piccoli.

The Role of Generative AI

In this scenario, generative AI does not replace existing systems but plays a coordinating role.

It is the layer that:

• Interacts with users in natural language
• Understands the operational context
• Orchestrates different models
• Simplifies technological complexity

A true central direction system that makes Artificial Intelligence accessible even to non-experts.

The 3A Model: Assist, Augment, Automate

Zucchetti’s vision for the Digital Supply Chain is embodied in the 3A model, which defines the proper role of AI in industrial processes.

Assist
AI supports people in their daily work, offering suggestions and decision support without replacing them.

Augment
AI transforms data into useful information, generating insights that improve decision quality.

How This Vision Translates into Zucchetti’s Digital Supply Chain Offering

This vision is the foundation of Zucchetti’s Digital Supply Chain offering, designed to bring specialized intelligences into different supply chain processes, according to varying data, objectives, and operational contexts.

The goal is to provide companies with the most suitable tools for each process, increasing decision effectiveness and supply chain resilience. From sales forecasting to predictive maintenance and logistics flow optimization, Artificial Intelligence becomes concrete support for daily operations.

“Zucchetti’s Digital Supply Chain offering follows this approach: not a single AI that does everything, but different intelligences designed to specifically support each process of the value chain,” adds Alice Kirchner, Sales Director of Cybertec, a group company specialized in Zucchetti’s Digital Supply Chain solutions.

In this way, Artificial Intelligence is not an end, but a means: a technology that adapts to real processes and everyday decisions.

This is where the supply chain truly becomes digital.