������

BOMP uses machine learning to produce the optimal interface between the factory management end-user and raw data from material properties in an industrial production environment. It aims to optimize materials selection, design, and processing via two learning components: exploring and exploiting “small data” to discover hidden patterns. 

The machine Learning algorithm is based on probabilistic Bayesian reasoning, also known as Bayesian Optimization, and can be easily adapted to challenging problems in many engineering fields, including but not limited to the design of alloys, fluid mechanics, chemical reactions, and materials processing. The final proof-of-concept is an easy-to-use software interface to hard-core science where a layman can optimize material properties and process parameters without any training. This replaces the need for process/material optimization consultancy with a fraction of the cost. This research-to-business project maps the potential avenues to exploit the scientific research as commercial business.

For more information visit the