Safety and AI at the Belfiore High-Speed Rail Construction Site (Florence)

Introduction

At the construction site of the new Belfiore high-speed rail station, worker safety and personal protection are the top priority. In an environment where ground crews, moving vehicles, suspended loads, and tight timelines coexist, risk rarely depends on a single mistake. It more often arises from the interaction of multiple simultaneous factors.

To address this complexity, NCM developed a Computer Vision and AI-based safety solution with a clear goal: prevent hazards before they become accidents by supporting crane operators with immediate, actionable information while managing suspended loads across the six cranes deployed on site.

The Challenge

Construction sites involve dynamic conditions that change minute by minute:

• People move in and out of operational zones

• Vehicles cross the site with variable visibility

• Loads travel on trajectories that can intersect with moving targets

• Operators must make fast decisions, often with limited visibility

Traditional safety procedures remain essential, but in real operations they can be challenged by:

• limited reaction time when multiple risk factors occur together

• blind spots due to equipment position and site layout

• lack of objective, real-time situational awareness at the point of operation

The Solution: Real-Time AI Vision for Crane Operators

The solution combines a camera installed on the crane jib with a tablet inside the operator’s cab. Using the camera feed, the system continuously monitors the operational area and detects in real time:

• people within the maneuvering area

• covered and/or uncovered vehicles moving across the site

• real-time distances, trajectories, and overlap zones between moving elements and suspended loads

This transforms complex visual information into clear guidance the operator can act on immediately.

How It Works

Processed data is converted into a simple operational representation that crane operators can consult instantly to avoid potential hazards.

The interface provides a schematic view of the site area with concentric risk zones:

• Safe zone (Green): safe area

• Warning zone (Yellow/Orange): pre-alert area

• Danger zone (Red): high-risk area

As people or vehicles progressively approach the suspended load, the system detects transitions between zones and triggers:

• visual alerts on the operator interface

• audio alerts to warn the operator and support situational awareness

This progressive logic supports early decision-making, not just emergency reaction.

Real On-Site Setup

The system was deployed directly in live site conditions and designed to operate continuously throughout crane operations. The combination of camera on the jib + tablet in-cab ensured that operators received a consistent, real-time picture of risk—even when site activity was dense and rapidly changing.

Operational Evidence Observed On Site

During real deployment, the system successfully detected and managed multiple scenarios typical of complex construction operations, including:

• Ground operators entering the warning zone beneath suspended loads

• Simultaneous presence of vehicles in safe areas and personnel approaching danger zones

• Material drop events within high-risk zones

In each scenario, the system generated progressive alerts as conditions evolved, enabling crane operators to react before a critical situation escalated.

This confirmed the shift from a reactive safety model to proactive risk prevention supported by objective data.

Technical Value for HSE Managers, RSPP, and Site Management

The Belfiore high-speed rail site application highlights concrete benefits for safety leadership and operational management:

• greater risk awareness during critical work phases

• early interception of risk scenarios through progressive alerts

• reduction in near misses (dangerous situations that could have caused injury but did not)

• more effective coordination between crane operators and ground personnel

• objective, verifiable safety support, strengthening the effectiveness of existing procedures

NCM’s technology does not replace HSE procedures, it strengthens them, making prevention more timely, measurable, and consistent.

Value for the Company

Putting people’s safety first also improves operational resilience and performance:

• reduced unplanned stoppages caused by critical events

• greater reliability of site operations

• improved execution quality and production continuity

• stronger reputation with clients and stakeholders

• a modern, structured, and demonstrable safety approach

In this context, safety is not just a requirement, it is a strategic lever that protects people, operations, and results.

2025 Regulatory Framework and Future Direction

As regulations evolve and accountability mechanisms become stronger on construction sites, prevention is increasingly central. Today, “being compliant” is no longer enough: companies need tools that make safety continuous, measurable, and truly effective in the field.

The Belfiore new high-speed rail station case demonstrates how Computer Vision and AI can support this shift, from a reactive model to a proactive one, delivering tangible benefits for workers and for the organization.

Key Outcomes 

• Real-time monitoring of people, vehicles, and suspended load risk zones

• Progressive warning logic (safe → warning → danger) with audio/visual alerts

• Increased situational awareness for crane operators

• Early interception of hazardous scenarios and fewer near misses

• Stronger, evidence-based safety management aligned with evolving compliance expectations

Contacts

Would you like to explore this application further or assess a similar project for your site?
Contact us at info@ncm.ai.