Solving Material Development Bottlenecks With AI

Test tubes in material development AI

Hitachi High-Tech developed an AI tool that enabled them to reduce their research on biodegradable additives from 120 days to two days.

Researchers have used AI in the development of new materials and additives for the production of plastics. Some companies are looking for alternatives to suit their customers’ needs. Normally these searches take place in huge databases accompanied by extensive phases of experimentation. However, new tools have allowed some advances in the field.

AI for materials research

In the development phase, the Hitachi High-Tech team set out to find an additive that would accelerate the decomposition of plastic to reduce environmental impact. Their goal was to guarantee the additive preserved the plastic’s conventional performance, including durability, flexibility, and moisture resistance.

To achieve this, the team explored a combination of polylactic acid (PLA), commonly used in biodegradable plastics, metal hydroxide, metal oxide and organic additives, and ultimately narrowed the selection to three optimal candidates. In this case, the achievement of expanding the consideration of candidate compounds from 600 to 60,000 was a significant advance.

Simultaneously, Hitachi drastically reduced the time required for material selection and simulation from 120 to only two days. The number of prototyping, measurement, and evaluation cycles was also reduced by 80%.

Reducing experiments to increase efficiency

Scientific biological laboratory for the study of biological materials and viruses in material development AI

In the past, researchers conducted many experiments in a physically meaningful way, but with uncertainty about the outcome. However, thanks to the creation of experiment plans using AI, it is now possible to carry out less experiments. AI-generated experiment plans enable selective, high-confidence experiments, reducing unnecessary trials.

These analyses use previous experimental data from customers, but can also be supplemented with data from patents, research papers and images.

Finally, this is just one example of how AI is gaining ground in the area of materials research and development. Companies need to start integrating it as part of their processes to solve bottlenecks and increase productivity.

By Juliana Montoya | December 29, 2023

Share Your Thoughts

Your email address will not be published. Required fields are marked *

Stay updated
Each week, receive a summary of all the latest news from the world of Plastics
Choose Language