The Evolution of AI in the Coatings Industry: From Subjective Observation to Predictive Analysis
With the passage of time, Artificial Intelligence and its remarkable developments are increasingly permeating various aspects of our daily lives. Regardless of the ethical or practical debates surrounding AI, its widespread adoption—especially in professional sectors—is now a reality, revolutionizing traditional operational methods.
Companies in the coatings sector must proactively approach these new AI-driven solutions to integrate their workflows (production, R&D, safety, marketing, etc.) into a global labor market transformed by the digital revolution.
In the field of Paint and Coatings R&D, AI tools are already proving invaluable for enhancing the design of new formulations and evaluating laboratory and application results. While the efficacy of a coating was once tested exclusively through subjective observation, today the integration of Computer Vision and predictive modeling allows for high-precision analysis of material properties.
Case Study: Comparative Analysis on "Sabbia" (Sand-type) Porous Ceramic
In this study of a new coating for porous flooring, AI was utilized to evaluate the efficacy of three experimental formulations compared to a market competitor. After 120 minutes of exposure, the liquid-solid contact angle—which scientifically defines the degree of surface hydrophobicity—was calculated.
Methodology:
The substrates (porous "sabbia" type ceramics) were coated with the four products. After 24 hours, an equal volume of water was deposited on each substrate to form a droplet. After 2 hours, the droplets were photographed to record the visual result. The images were processed via the Gemini AI interface with the following technical prompt:
"Evaluate the water repellency of the droplets in the photos according to the UNI EN 15802 standard. Generate a report by collecting the data in a table and providing conclusions. The tested material is a 'sabbia' ceramic tile treated with water-repellent products."
Analytical Results
| Sample | Visual Description of Droplet | Estimated Contact Angle (θ) | Hydroperellency Level |
| A | Well-defined droplet, flattened spherical shape with sharp edges. | Medium-High (70-80°) | Good |
| B | Very flat droplet, spread across the surface. Tendency toward soaking/imbibition. | Low (<40°) | Poor/Insufficient |
| C | Droplet with excellent surface tension and pronounced curvature. | High (85-90°) | Excellent |
| Comp. | Flat droplet, very low leading-edge angle. | Low (45-50°) | Sufficient |
Technical Conclusions
The comparative analysis after 2 hours of exposure reveals the following:
Maximum Efficacy: The sample treated with "Formula C" shows the best performance. Despite the 120-minute interval, the droplet maintains a compact shape and a high contact angle, indicating the product created an effective barrier that prevents substrate wetting.
Intermediate Performance: "Formula A" provides valid protection, exceeding common market standards by maintaining good surface tension.
Criticalities: "Formula B" and the "Competitor" show performance degradation over time. The droplet tends to spread (wetting), indicating that the surface energy of the treatment is too high, allowing water to expand or begin penetrating the micropores of the ceramic.
Final Result: Formula C is the most suitable candidate for guaranteeing long-term protection against standing liquids on this specific substrate.
This case study highlights the potential of AI in coatings research, specifically leveraging Digital Vision to analyze visual outcomes according to industry standards.
Will the future development of AI-based software—increasingly tailored to formulation and lab evaluation—provide tools capable of significantly improving the daily work of researchers and companies?
Will we become increasingly "dependent" on these technologies, perhaps casting aside "human" technical experience?
Commenti
Posta un commento