Machine Learning can be applied to help complete missing data, so it doesn’t require a lot of data. There’s lots of research to improve Machine Learning Transparency, but even if it’s not transparent, it’s still worth it. The use of machine learning is very broad, including in the industrial sector. In its application, several methods can be used, one of which is theory-based modeling, which is very often used. Youssef Marzouk said that “there are many situations where a theory-based model is not sufficient.”
Before the advent of practical computing, engineers and scientists began to apply the first principles of the laws of physics built on observations (the first paradigm) and theory-based models (the second paradigm) that aim to describe the physical world with computers. As computers become more powerful and accessible in the third paradigm, computational modeling and simulation are applied to theory-based models that provide a new level of prediction. The 4th Paradigm, This computational transformation combines theory-driven models with data using high-performance computing, large data sets, and powerful new algorithms, including machine learning. Which is certainly better than the previous paradigm.
Machine Learning is great for speeding up the modeling process and providing shortcuts to solving problems and knowing the rules. So that there are many uses of Machine Learning in industrial fields such as; in Fluid Mechanics: Knowing the mechanism of “essential flow” at a lower cost, in Oil and Gas Exploration : Predicting how formations will react to certain drilling techniques, in Mechanical Engineering: Extending the life of the tool by predicting the maintenance on the tool, and many more applications in the industrial field.
Read further information here