{"id":21860,"date":"2026-05-28T12:40:59","date_gmt":"2026-05-28T18:40:59","guid":{"rendered":"https:\/\/ulprospector.ul.com\/?p=21860"},"modified":"2026-05-28T12:40:59","modified_gmt":"2026-05-28T18:40:59","slug":"pc-ai-in-coatings-formulation","status":"publish","type":"post","link":"https:\/\/ulprospector.ul.com\/21860\/pc-ai-in-coatings-formulation\/","title":{"rendered":"AI in Coatings Formulation"},"content":{"rendered":"

Coatings formulators and manufacturers are facing pressures, old and new in order to stay profitable in today\u2019s business environment. Like many other enterprises, coatings companies are turning to Artificial Intelligence (AI) resources to meet these challenges. How can AI be used to develop coatings formulations? Can AI help overcome long-standing cost and performance challenges? Yes, by using AI coatings producers can meet the new challenges posed by sustainability initiatives and changing regulatory pressures.<\/p>\n

\"\"
Image created using ChatGPT<\/figcaption><\/figure>\n

According to the IBM website: \u201cArtificial intelligence (AI) is technology that enables computers and machines to simulate human learning, comprehension, problem solving, decision making, creativity and autonomy.\u201d <\/em>1 <\/sup>Let\u2019s see what an AI agent says about AI in coatings formulation. I asked ChatGPT to provide a summary of the topic, \u201cAI in Coatings Formulation\u201d The popular AI agents reply was, \u201cAI in Coatings Formulation<\/em><\/strong> refers to using machine learning and data-driven models to design, optimize, and troubleshoot paint and coating systems faster and more accurately than traditional trial-and-error methods.\u201d <\/em>ChatGPT offered to prepare a one slide graphic explaining AI in coatings formulation<\/a> which is shown as the header image for this article.<\/p>\n

Applications of AI in Coatings \u2013 Present and Future<\/strong><\/p>\n

For an expert\u2019s insights and opinions, I spoke with Dr. Erik Sapper, Associate Professor of Chemistry at California State Polytechnic University in San Luis Obispo. Erik is a leading academic researcher on the topic of using AI and Machine Learning (ML) to enable better polymer synthesis and coatings formulations. We discussed the topic of AI in coatings formulation and addressed areas such as current applications, what he thinks the future will bring, and how the use of AI in the coatings industry<\/a> will affect the workforce. Dr. Sapper teaches in Cal Poly SLO\u2019s undergraduate and graduate programs in polymers and coatings.<\/p>\n

The Current State of AI Implementation in Coatings<\/strong><\/p>\n

First Dr. Sapper and I talked about how AI is most commonly being utilized in coatings formulations today. \u201cWhere these tools are being used most effectively is as a lab assistant.\u201d In his experience \u201cthe biggest and easiest gains are going to be in adjusting well-defined data sets to solve formulation problems using limited existing data.\u201d This illustrates one of the common advantages to AI utilization cited by numerous sources, using AI as a tool to increase resource efficiency and reduce product development timelines.<\/p>\n

AI as a data gathering and analysis tool, AI as a predictive tool<\/strong><\/p>\n

When asked about what\u2019s next in the implementation of AI in the coatings industry, Dr. Sapper had this to say. \u201cThere will be a lot of options for data ingestion, taking existing data, i.e. spreadsheet data, and use data in different formats.\u201d Using technologies such as Generative AI and Large Language Models (LLM) \u201cdata wrangling will be sped up, combining and collapsing disparate data sets into a singular data set for training models\u201d. These models are then used to make predictions of formulation properties ranging from rheological and optical properties to film hardness and corrosion resistance.<\/p>\n

\n

ULTRUS\u2122 Product Stewardship enables material selection, compliance, sustainability and chemical management throughout the product life cycle to meet regulatory requirements and consumer demands. <\/span>Learn more here<\/a>!<\/h2>\n
\n

Cautions \u2013 AI in Coatings Formulation<\/strong><\/p>\n

Like any formulation technique based on predictions derived from data, data quality is all important. Small datasets can yield reliable predictions as long as those datasets are well-defined and documented. Specialized AI agents tailored to applications in chemistry or coatings can be very helpful as opposed to big data applications with very large, generalized datasets. Dr. Sapper remarked that development of new AI agents is taking place in both private industry and at universities like Cal Poly. Universities are taking the lead in areas like AI assisted polymer development while software developers are mainly working on generalized agents with broader market appeal. To quote Erik Sapper,\u201d University researchers aren\u2019t trying to sell anything, they can focus on the fundamentals.\u201d<\/p>\n

DIY AI Paint Formulation<\/strong><\/p>\n

A custom made of formulation targeted commercial AI agent isn\u2019t required to get your feet wet in AI paint formulation. Try using general generative AI programs to guide your coatings formulation<\/a> development efforts. Just as you might use a Google search to find raw materials or suppliers, you can use Googles AI Mode or ChatGPT to generate starting formulation ideas to build upon. A generative AI program is only as good as the input query and the data accessible to it. The user has no control over the dataset used by a general utility AI and therefore must tailor the input to get the most useful output.<\/p>\n

Different AI agents might give completely different answers to the same query. As an example, let\u2019s ask the Google AI Mode search and ChatGPT the same query. Using the query,\u201c \u201cFormulate a mid-market industrial coating\u201d I received these initial outputs.<\/p>\n

Google AI Mode<\/strong><\/p>\n

\u201cTo formulate a midmarket industrial coating, you must balance cost and performance. This formula uses a styrenated acrylic emulsion to offer good chemical resistance and gloss at an affordable price point.<\/em><\/p>\n

Base Formula Overview<\/em><\/strong><\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Component<\/em><\/td>\nFunction<\/em><\/td>\nWeight %<\/em><\/td>\n<\/tr>\n
Resin Base<\/em><\/td>\nStyrenated Acrylic Emulsion<\/em><\/td>\n45.0%<\/em><\/td>\n<\/tr>\n
Solvent\/Carrier<\/em><\/td>\nDeionized Water<\/em><\/td>\n18.0%<\/em><\/td>\n<\/tr>\n
Pigment<\/em><\/td>\nTitanium Dioxide (Rutile)<\/em><\/td>\n15.0%<\/em><\/td>\n<\/tr>\n
Filler<\/em><\/td>\nCalcium Carbonate (Fine)<\/em><\/td>\n10.0%<\/em><\/td>\n<\/tr>\n
Coalescent<\/em><\/td>\nTexanol (Ester Alcohol)<\/em><\/td>\n5.0%<\/em><\/td>\n<\/tr>\n
Dispersant<\/em><\/td>\nSodium Polyacrylate<\/em><\/td>\n1.5%<\/em><\/td>\n<\/tr>\n
Defoamer<\/em><\/td>\nSilicone-free Defoamer<\/em><\/td>\n0.5%<\/em><\/td>\n<\/tr>\n
Flash Rust Inhibitor<\/em><\/td>\nSodium Nitrite Solution<\/em><\/td>\n1.0%<\/em><\/td>\n<\/tr>\n
Thickener<\/em><\/td>\nHEUR (Polyurethane)<\/em><\/td>\n4.0%\u201d<\/em><\/td>\n<\/tr>\n
<\/td>\n\u00a0<\/em><\/td>\n\u00a0<\/em><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Step-by-Step Manufacturing Process<\/em><\/p>\n

    \n
  1. The Grind Phase (High-Speed Dispersion)<\/em><\/li>\n<\/ol>\n