From document automation and voice commands to machine learning for estimating and production planning, B+B International is bringing artificial intelligence into the packaging and display industry through Packway AI. We spoke with Davide Dal Col, CEO of the company, to understand how this technology is already creating tangible value for businesses across the sector.
When discussing artificial intelligence, the focus often falls exclusively on the technology itself, overlooking the fact that its true value lies in the knowledge and expertise that power the algorithms. This is precisely the foundation of Packway AI, the project developed by B+B International to integrate artificial intelligence into its software ecosystem dedicated to packaging, displays, and digital printing.
Founded in 1993, B+B International develops ERP and CAD/CAM solutions specifically designed for the packaging industry and is now recognized as one of Italy’s leading technology providers in this field. More than 220 companies use the Packway ERP system, while over 1,000 ArtiosCAD licenses and an equal number of Kongsberg cutting tables have been installed throughout Italy. With annual revenues of approximately €19 million and a team of 65 professionals, the company has built, over more than three decades, an extensive heritage of expertise, data, procedures, and industry knowledge that now serves as the foundation for its new AI applications.
Another key strength is the company’s ISO 9001 and ISO 27001 certifications, which ensure high standards in process quality, information security, and data protection, factors that are becoming increasingly important in the age of artificial intelligence. To better understand the potential of this technological evolution and its practical applications within the packaging industry, we interviewed Davide Dal Col, CEO of B+B International.
Davide, let’s start with the basics. What is artificial intelligence, really?
“Artificial intelligence is a branch of computer science that develops software and hardware systems capable of simulating some of the core human abilities, such as learning, reasoning, planning, and creativity. Simply put, it is a technology that enables machines to process information, learn from data, and perform actions or provide answers in order to achieve specific objectives. What makes AI particularly revolutionary is the speed at which it is spreading. Electricity took around 50 years to reach 50% of the world’s population, automobiles 45 years, personal computers 20 years, the Internet 10 years, and smartphones 7 years. Artificial intelligence is reaching the same level of adoption in less than three years. The real difference, however, is that AI is a horizontal technology, it can be applied across virtually every industry and business activity”.
How does artificial intelligence work in practice?
“AI processes enormous volumes of data using sophisticated mathematical algorithms. It analyzes information, identifies recurring patterns, learns from experience, and uses this knowledge to generate responses, make predictions, or support decision-making processes. However, it is important to understand that the quality of AI’s output depends directly on the quality of the information available to the system. This is why experience, historical data, and deep industry expertise become fundamental success factors”.
What is an AI Agent, and how does its architecture work?
“An AI Agent is a software system capable of perceiving its environment, understanding a user’s request, making decisions, and autonomously performing a series of actions to accomplish a specific objective. Its architecture consists of several components working together. On one side is the user, who can interact through a keyboard, voice, or visual input tools. On the other side is the AI agent, which interprets the request, plans the necessary activities, and coordinates the execution of tasks. To accomplish this, the agent can access an extensive knowledge base that includes user manuals, operating procedures, technical documentation, assistance tickets, corporate databases, and calculation tools. All this information is used to build the appropriate context, allowing the AI to generate responses that are truly relevant and useful. This is precisely where B+B International’s more than thirty years of experience become a strategic advantage”.
How is this architecture implemented within Packway AI?
“When users interact with Packway AI, the system does not simply query a generic language model. Our AI agent has been trained using millions of lines of Packway ERP source code, together with more than thirty years of technical documentation, assistance tickets, manuals, and operational procedures. In essence, we have embedded the company’s accumulated knowledge directly into the system. This enables the AI to understand the specific context of the packaging and display industry and deliver answers that are significantly more accurate and relevant than those produced by a general-purpose AI system”.
Can you share some practical examples of Packway AI?
“We have currently developed three primary application areas. The first is document automation. The system can receive documents from customers or suppliers, automatically interpret them, translate them when necessary, and import the information directly into Packway without any manual intervention, regardless of the ERP platform used by the other party. Consider a PDF purchase order received from a customer or a delivery note sent by a paper mill. Packway AI reads the document, identifies the relevant information, translates it if required, and automatically records it within the ERP system. This eliminates duplicate data entry, reduces the risk of errors, and significantly improves operational efficiency. Today, Packway AI Document Automation already processes thousands of transactions every day. We have active integrations covering both sales and purchasing processes, involving multinational companies, major industrial groups, and leading paper manufacturers. This is not a pilot project or proof of concept, it is a fully operational solution already deployed by numerous customers”.
Beyond document automation, what other AI applications have you developed?
“The second area focuses on voice commands. Users can interact directly with the Packway ERP system using natural speech. They can ask the system to create a quotation, retrieve customer information, check inventory levels, or access order details. Our objective is to make interaction with enterprise software increasingly natural and intuitive. The third application area is machine learning for quotation management and production planning. By collecting data directly from production machinery, the system can become progressively more accurate in determining sales prices while simultaneously improving production planning. This enables better on-time delivery performance for customers and more efficient allocation of internal resources”.
How does machine learning improve quotations and production planning?
“Although companies can define highly detailed production standards, estimating processing times has traditionally depended heavily on the experience of individual operators. This often results in variations between estimators and makes it difficult to consider every production variable. With machine learning, the system continuously learns from actual production results and uses this information to automatically suggest processing times for each manufacturing operation. The more production data the system analyzes, the more accurate its predictions become. In this way, AI provides increasingly reliable support for both quotation preparation and production scheduling”.
Product first, then communication
The topics discussed during this interview with Davide Dal Col were recently presented in a webinar organized for Gifasp members, representing the first major public introduction of the Packway AI project. The response exceeded expectations. Following the presentation, participating companies submitted numerous questions and requests for further information, which continued in the days that followed.
“I believe the success of our initiative stems from a collaborative and educational approach aimed at explaining what artificial intelligence really is, how it works, and the tangible benefits it can bring to organizations. We consider this an essential step in helping companies understand a technology that is often presented either too superficially or in overly theoretical terms”, comments Dal Col.
The project itself reflects a clear philosophy: first develop practical solutions, deploy them at customer sites, validate their effectiveness in real operating environments, and only then begin communication and educational activities. Today, Packway AI is no longer an experimental initiative. It is a fully operational platform that processes thousands of transactions every day and supports mission-critical business processes across packaging companies.
“Over the coming months, we will continue expanding the platform’s capabilities while intensifying our educational and training initiatives. New webinars for customers are already scheduled, along with additional events organized in collaboration with industry associations. Our goal is to promote greater digital awareness and support companies in the effective adoption of artificial intelligence technologies”, concludes the CEO of B+B International.
The challenge of the company extends beyond developing new software tools. Its ambition is to transform AI into a practical, measurable business asset for the packaging industry, one that drives efficiency, strengthens competitiveness, and creates new opportunities for sustainable growth.
