By Wolfram Keller (Contact), Ulf Stalmach, Ralph Wörheide
Today, raw material and paint manufacturers develop formulations based on the raw material portfolio they are familiar with. These products are intended for only one or a few market segments and applications. Finally, customers check to what extent the product meets their requirements. Especially in the case of global formulations, once a product has been developed, it is often manufactured centrally and distributed worldwide, as raw materials are usually not available in the same quality across all regions and product quality is often not identical at various production sites.
In the future, a customer will no longer just define product specifications, but requirement profiles that focus on the desired effect of the coated surface. This requires a better understanding of the parameters of the application, the properties of the coating and the underlying manufacturing and material properties.
Artificial intelligence (AI) is a powerful tool to propose several alternative options to meet the customer’s requirement profile. These options are based on an ever-growing pool of data on the origin, quality, specified and currently unknown or specified properties of the raw materials and their impact on the manufacturing process, equipment and application.
Not all of the data relevant to cover the entire value chain, from raw material suppliers to paint manufacturers and users to recycling companies and back, are generated yet. These closed data loops require a sufficient amount of relevant data, though, to ensure the quality of any machine and deep learning based simulation.
Smart companies in other industries are already using all kinds of connected technical, commercial and regulatory information to define and improve services, products, production sites and carbon footprint as well as total cost of ownership. Coatings and paints manufacturers can benefit in a similar way by collecting, archiving, using and, above all, sharing data just as consistently, i.e. thinking and acting “smartly”.
Figure D5: Centralized product development loops with AI and simulations that enable faster time-to-market and decentralized production of global formulations
By combining AI-powered predictions, simulations, and Design-of-Experiment 2.0, product development cycles and customer requirements can be drastically shortened. The operational benefits are shorter delivery times, lower total cost of ownership, resource consumption and CO2 footprint and higher customer satisfaction, e.g. by decentralizing raw material sourcing and production, even for “global formulations”.
This article is the fourth of six in our series on sustainable digitalization in the coatings industry, the concept of a Smart Paint Factory, and the Smart Paint Factory Alliance, SPFA