Nicolas Vandeput is a supply chain data scientist specializing in demand forecasting and inventory optimization. 

Episode Notes:

Data science in SCM 4:56.
Small data 6:09.
Historical sales/stockouts/marketing expenses data 6:40.
Importance of data culture 9:11.
Challenge of developing in house machine learning expertise 11:32.
Statistical models for determining demand patterns 12:12.
Open source, Python and R 14:47.
Forecasting models 17:24.
Building your own machine learning model 18:08.
Overfit 18:49.
Underfit 22:10.
Selecting KPIs 23:04.
Mean supply error 24:48.
Bias 25:01.
Forecast added value 26:02.
Code used in production 30:15.
Process of machine learning 30:38.
Unsupervised learning 33:52.
Norwegian car sales35:35.
Off-the-shelf solutions vs having models built 38:22.
What are outliers in the COVID-19 pandemic 42:21?
Inventory optimization 45:24.
ROI from machine learning 47:57.