Notícias

Radix Machine Learning Solutions accelerate processes

Machine learning Artificial Intelligence

In the Digital Revolution era, it is imperative to invest in Artificial Intelligence, making processes and software associated with learned behaviors smarter. It is in this context that Radix has developed Oil X!Pert, a technology that uses Machine Learning techniques to accelerate and optimize processes in industrial laboratories and LIMS systems, allowing tasks typically performed by collaborators to be automated and optimized.

According to Radix engineer Rafael Ladeira, the technology applies mathematical algorithms used by companies such as Google and Facebook. "The system uses a large amount of data to automatically learn the process dynamics. The process of interpretation of fluid analysis can be accelerated and treated more objectively by the analysts with the knowledge acquired by the mathematical model," said the engineer.

Oil X!Pert is being used in the oil and fluid analysis laboratory of Sotreq, one of the largest Caterpillar equipment dealer in the world. For Sotreq's COO, Marcelo Orberg, the technology developed by Radix has multiplied three times the capacity of interpretation of the results by analysts. "This tool, in addition to exponentially increasing the ability of our specialists to generate recommendations, has practically eliminated the errors of interpretation in a manual process and considerably reduces our time in delivering the service. Radix was able to put into practice what we have always dreamed possible with the advancement of Artificial Intelligence," Orberg said.

The next step is to include Deep Learning technology in Oil X!Pert. "We want to develop an even more powerful mathematical model using the latest advances in science in ​​Deep Learning, which has opened new horizons in the field of Artificial Intelligence. We are working together with researchers from the Federal University of Minas Gerais (UFMG) to bring these scientific advances to Oil X!Pert and to other Radix projects," said Ladeira.