4 Ways that AI is Powering a More Sustainable Chemical Industry

The chemical industry is one of the planet’s most important employers and manufacturers. Unfortunately, it is also one of the world’s biggest polluters and consumer of resources. This means that if climate change is to be limited without lowering living standards, then chemical companies need to become more sustainable.

Embracing the circular economy is now seen as a key issue across the chemical industry, with nearly two-thirds of participants in an American Chemistry Council 2024 survey of executives stating that enhancing sustainability is their top priority for the ensuing two years, with nearly half of respondents believing it to be the biggest challenge the chemical sector is facing.

To achieve this means in-depth analysis, plenty of data, and a holistic approach to chemical production. As a recent report in the industry journal Specialty Chemicals Online, notes, “Whether the priority is to cut product carbon footprint, incorporate more renewable energy into processes, adopt more sustainable procurement practices and/or pursue other pathways to becoming a more sustainable business, chemical companies are going to need to rely heavily on fresh, high-quality data from internal and external sources—and on intelligent tools to manage and analyse that data—to meet that challenge.”

Facing up to the issue of sustainability will require a new way of thinking—an ideal role for artificial intelligence.

AI is already bringing massive changes to the workplace, driving efficiencies in data analytics, supply chain management, logistics organisation, chemical product pricing, and sales planning. However, when it comes to improving a chemical company’s sustainability, there are four key areas where AI adoption is yet to have the most significant impact.

1. Data and Reporting Compliance

The chemical industry, alongside other sectors, such as manufacturing, is on the verge of a new era of cooperative data exchange and standard-setting.

As individual chemical companies are increasingly required to provide sustainability-related information to their customers the need to compile data for regulator and adhere to stricter environmental laws will grow. As a frequent midstream player in receiving raw materials and supplying value downstream, chemical companies will need to invest more resources to manage information and comply with legislation. Such tasks are ideal for the application of AI tools.

One example of AI assisted data management can be found in the Together for Sustainability (TfS) initiative. As the Specialty Chemicals Online report notes, “… participants in TfS [have] separately created supplier information-sharing programmes for reporting emissions, including Scope 3, along with energy, water consumption and waste volumes.” The result is a Strategic CO2 Transparency Tool (SCOTT) methodology for calculating a product’s carbon footprint to help chemical companies conform to TfS standards.

This kind of action is setting the stage for digital passports that give consumers, authorities, and other interested parties access to a variety of details about particular products, such as the source of their industrial ingredients, a physical description, certification, carbon footprint, recyclability, and most notably content (renewable versus non-renewable).

“These passports will be required for products manufactured or sold into the EU from 2027 as part of the EU Green Deal,” explains Sergey Nozhenko, a chemical industry product specialist at SAP. “It probably will not be long before similar reporting requirements take hold in the US, as regulators, consumers and companies themselves push for a more sustainable chemical industry.”

2. Raw Material Sourcing

Artificial intelligence is also revolutionizing procurement in the chemical industry by enhancing data analysis, automating routine tasks, and improving supplier management. AI-driven tools can process complex datasets to forecast price trends, assess supplier performance, and predict potential supply chain disruptions, enabling procurement teams to make informed decisions and negotiate favourable contracts.

Furthermore, repetitive tasks, such as purchase order creation and invoice processing, can be automated to reduce errors and free up staff to focus on tasks better assigned to humans.

One notable example of AI integration in procurement is a specialty chemical company that implemented a digital material-number system to address inefficiencies in ordering low-value items. Unclear technical details and missing part numbers had been causing process delays and increased costs. By developing a simple algorithm linked to their requisitioner tool, the company streamlined the selection and ordering process. This initiative reduced maverick spending and enabled volume bundling for specific part numbers, leading to a 2% reduction in spending within the first year—a return on investment ten times the setup costs.

3. Process Optimisation

By enhancing process optimisation, chemical companies can increase efficiency, reduce costs, and improve sustainability. This is because AI algorithms are ideal tools for analysing vast amounts of data from chemical manufacturing processes to identify patterns and predict outcomes. This enables real-time adjustments that optimise operations and results in benefits such as reduced energy consumption, minimised waste, and enhanced chemical product quality.

A notable example of AI-driven process optimization is BASF's collaboration with Louisiana State University (LSU). BASF, the world's largest chemical producer, partnered with LSU to implement AI and machine learning techniques to better understand and predict production variations at its Geismar, Louisiana plant. The project involved developing data mining processes to organise and compare current operating conditions with historical data.

By employing unsupervised machine learning, the team aimed to identify intrinsic behaviours of processes and uncover unforeseen patterns. This approach allowed BASF to optimise production workflows, develop soft sensors for real-time quality parameter estimation, and enhance decision-making processes.

4. Chemical Research and Product Development

Artificial intelligence (AI) is transforming chemical research and product development by enabling more efficient data analysis, accelerating discovery processes, and fostering innovation. AI algorithms can process vast datasets to identify patterns and predict outcomes, facilitating the design of novel compounds and materials with desired properties. This approach reduces the reliance on traditional trial-and-error methods, thereby saving time and resources.

A notable example of AI integration in chemical research is Evonik Industries' development of a virtual formulation assistant for the paint and coatings industry. By utilizing AI and machine learning in a product branded as Coatino, Evonik is able to provide tailored additive recommendations based on specific user requirements.

By analysing extensive datasets and incorporating decades of expert knowledge, this AI tool can suggest optimal formulations to enhance efficiency and innovation in chemical product development.

AI is poised to play a crucial role in driving sustainability within the chemical industry. By optimizing procurement, enhancing compliance reporting, improving manufacturing processes, and accelerating research and development, AI enables companies to reduce waste, lower emissions, and create more sustainable chemical products.

With chemical industry leaders already keen to adopt AI for its inherent efficiencies, it cannot be long before they also embrace the advantages it offers for what they already see as the chemical sector’s greatest challenge—sustainability.

Photo credit: Freepik, Wirestock, Tom Fisk on Pexels, Rawkkim on Unsplash, & Omkar Jadhav