How to Use Machine Learning to Predict Customer Behavior

We go everywhere from basic XG boost or boosted trees for when we look at data coming from a new customer, we're predicting how that data should map to our model. We've been building deep learning models for a group of supply chains that are coupled together. So what that does is it gives us a way to run through many, many different scenarios and show customers traces of what expected behavior would be without running huge Monte Carlo simulations.