Maybe A Solution For All Glass-Made Products From IIT-Delhi

Saanvi Araav - Sep 03, 2019

The solution from IIT Delhi comes with the name PyGGi (Python for Glass Genomics). They use this software to optimize and predict glass compositions

These days, many are wishing for window panes, glass utensils, smartphone screens that could resist damage. On the other hand, glass manufacturers also want a mechanism to predict glass compositions to make products with customized properties, and researchers from IIT Delhi have a solution for that.

Optimize And Predict Glass Compositions

Human has been making and using glasses for more than 2000 years already. Despite that, making glasses with customized properties is still a big challenge. Thus to solve this problem, researchers from IIT Delhi developed a new machine learning software. It comes with the name PyGGi (Python for Glass Genomics). They use this software to optimize and predict glass compositions.

glass-AI — Human has been making and using glasses for more than 2000 years already

It will allow companies and researchers to predict glasses with better properties like crack and scratch resistance with just a simple click.

N M Anoop Krishnan is an IIT Delhi's professor and one of this project's project investigator. He said that to develop novel glasses with better properties such as scratch-resistant and bulletproof, we had to be able to predict and understand the composition-structure- property relationship of the glasses.

A Data-Driven Method

Krishnan also added that we could use data-driven methods like artificial intelligence and machine learning to exploit our current knowledge to predict glasses for customized applications. He and the team have used python to develop this PyGGi software to optimize and predict inorganic glasses' properties.

The main objective of this software is reducing the cost of predicting novel glasses for customized applications.

Hariprasad Kodamana (another professor at IIT Delhi) said that they would keep update and upgrade PyGGi constantly to keep up with the academic and industrial challenges in glass science. They also open to making raw modules base on the requirements of users. Then they could give these modules to the supporters of the project.