Data Science for Supply Chain Forecasting 2nd ed. Edition

Nicolas style to write books displays his strong academic foundation as well as his vast experience in consulting various companies. The synergy of theory and practical application makes the content lively. The language is supply chain and data analysis specific, but yet simple enough and easy to understand.The book is well structured and the chapters build upon each other. I really enjoy the sub-structure of first being presented with the context, then theory and finally the call-to-action to apply the knowledge either in excel or python - this definitely helped to fortify each chapter's content. Nicolas challenges you to not only "having heard about it" but actually "know how it's done" - of course it's up to you what you choose to do.I work in aerospace and used the content of this book to develop a machine learning forecasting model which uses fleet flight hours, meantime between failures, meantime between unscheduled repairs and known maintenance intervals to forecast the predicted influx of returned parts from the field. This helped better prepare any required inventory to ensure short turnaround times and bring parts back into service as quickly as possible.Using this forecasting model and knowing the accuracy/error of your forecast, you can simply take this parameter as "demand deviation" and optimize your inventory to account for the known fluctuations, while achieving a cost optimal inventory level or target service/fill rate level. If you want to learn more about inventory optimization, I strongly recommend to check out Nicolas' book "Inventory Optimization", which is my personal favorite.To summarize, the book "Data Science for Supply Chain Forecasting" is great resource for any data analyst who wants to increase exposure to simple and advanced forecasting methods - in my opinion this book is useful in many other disciplines than only supply chain, e.g. also in planning, sales, operations, etc.

As a consultant and as a teacher, I have truly enjoyed reading this book. Apart from the Python code it provides, which is easily understandable even for beginners in Python, Nicholas also gives many interesting points of view about commonly confusing concepts in supply chain. I highly recommend it.

It should be a standard position for all people working in supply chain. Very accessible language giving strong fundamentals to deep dive into more complicated SCM topics

Other than the last couple of pages where the author talks strictly about supply chain issues, how to deal with stakeholders, and specific KPIs he found useful in his vast experience with the subject, I'm really struggling to find anything in the remainder of the book to justify a higher rating.There's really nothing here that cannot be found in better introductory texts on foresting in general. The application to Supply Chains is basic at best, and the models (while proven and effective) are nothing a data scientist worth his/her salt would not already be familiar with. So while the author makes the case that this is a text aimed at DSs, it's really aimed at junior, excel based analysts/planners looking to get baseline forecasts - there are actual excel recipes for every statistical model.There's some ML in the second half of the book but very basic and done better elsewhere for those interested in that side of things.If you are really looking for great resources on your forecasting journey, I can recommend the following (for practitioners, not theorists):Forecasting:Advanced Forecasting with Python by KorstanjeTree Methods:Hands on Gradient Boosting with XGBOOST and ScikitLearn by WadeNeural Networks:1. Hands on Machine Learning with ScikitLearn and Tensorflow by Geron2. Deep Learning with Python by CholletFeature Engineering:Feature Engineering for Machine Learning by Soledad Galli(please make sure to check out the UDEMY course as well!)Feature Selection:Feature Selection for Machine Learning by Soledad Galli(please make sure to check out the UDEMY course as well!)Regression Analysis:Regression Analysis by Jim Frost

Excellent introduction to demand forecasting models and it's applications, from traditional and simple models to more advanced models. The author explanations are very clear, with lot of practical insights and without including a lot of math and theory and providing useful code. The book is oriented mostly for data scientists with little experience in demand forecasting and demand forecasters or demand planners looking to apply data science in their role.I think the only drawback of the book is that it didn't include ARIMA family models and deep learning models for forecasting. There is also little information about time series analysis and feature transformation.

Content is good but the pages are falling out no matter how you handle it

U can find all those simple concept online…