There are so many AI tools for chemical engineering jobs that it takes years to find the right one. So researchers at North Caroline State University have trained a ‘virtual lab’ to choose the best AI tools for each job.
Having the right tool for the job makes the job a lot easier, less expensive and faster.
They simulated 600,000 experiments, assessed 150 AI-guided decision-making strategies and claim to have cut 7.5 years of continuous robotic operation into one month’s work.
Chemical engineering researchers have now developed a virtual laboratory that can be used to determine the artificial intelligence (AI) tools best suited for addressing various chemical synthesis challenges in flow chemistry systems.
Autonomous systems can accelerating chemical R&D and manufacturing, but they are not in widespread use for two reasons. It’s hard to select the right hardware for automated synthesis – and it’s impossible to find AI-guided decision-making algorithm, says Milad Abolhasani, assistant professor of chemical engineering at North Carolina State University.
There are three reasons why: One, there’s a huge choice of AI tools available. Two, there’s not information about them – which is add omission for artificial intelligence – so it’s hard to decide which tool is the best for each material synthesis problem. Three, the admin doesn’t stop once you have made your decision because each tool, once selected, still needs to be fine tuned.
“Recently, there has been increased interest in using off-the-shelf AI programs for modelling and optimisation of chemical reactions,” Abolhasani says. “But those off-the-shelf AI techniques are not one-size-fits-all – they’re not all equally good at solving whatever material synthesis problem you want to address.
“Ultimately, we want to find the best AI model architecture for determining the best material formulation that gives you the target properties you are looking for. Not just identifying the best material, but the best way of producing that material so that it has the best possible combination of characteristics. And the best AI model architecture is going to vary depending on the material and the complexity of the challenge.”
So Abolhasani and his collaborators took an AI-driven approach to finding the best AI tool for each material synthesis problem.
“It would be impossible to do the millions of experiments necessary to determine which AI tools do the best job for addressing different kinds of material synthesis problems,” Abolhasani says. “So, we wanted a model that simulates a real-world microfluidic experimental platform to effectively run those millions of experiments for us.”
The researchers ran 1,000 experiments using their automated Artificial Chemist platform and used those experimental data points to train the virtual experimental platform.
The paper, Accelerated AI development for autonomous materials synthesis in flow, is published in the journal Chemical Science.