Emerging algorithmic paradigms for supply chain optimization (with David Simchi-Levi)


00:07 Introduction
02:35 Prof. Simchi-Levi, you recently launched a course on Supply Chain and Demand Analytics where you mention digitization, analytics and automation. Why are these concepts so popular right now?
05:20 Joannes, what do you think about the compromise between resilience and efficiency?
16:00 The idea that we can predict or supposedly predict what is going to happen in our supply chain 6-7 weeks from now is interesting. At Lokad we embrace the uncertainty and look at the likelihood of various futures. How exactly can you predict the future, and do you then only look at one output?
25:35 Prof., you mentioned the laws of Physics in your course. What do you mean by laws of physics in the supply chain perspective?
35:24 You often talk about a framework applicable to many problems rather than just one solution alone in your research and publications. How come?
39:36 Joannes, what do you think about the S&OP process that Lokad has quite a different approach to?
48:16 What emerging algorithms do you see at the moment in supply chain learning and optimization and what negative trends do you see becoming popular?
51:41 What is the main difference between online and offline learning?
59:16 What are your thoughts on Joannes’ opinion, Prof. Simchi-Levi?

Description

Forbes describes that a shift in paradigm has led supply chain planning systems to become more intelligent by learning to adapt to ever changing circumstances by sensing and analyzing data. Where the decision making process is more and more handled by machines than humans. In this episode we are discussing paradigms that are now emerging for optimization and learning in Supply Chain, and we are joined by Prof. David Simchi-Levi,author of over 300 publications, in which a common topic is algorithmic paradigms for supply chain.

We discuss concepts such as trends in technology, laws of physics in supply chains, the constant balance between flexibility and service level, and how to find that ideal balance. The pandemic showed us that the future is here and its disruptions will be felt for years to come. One cannot make decisions about the future anymore based on what one’s used to due to the new normal. Furthermore, the idea that relocating manufacturing closer to market demand will guarantee resilience is a misguided concept.

The key to succeeding in this game of supply chain poker is the ability to integrate machine learning and optimization to make better decisions - through a combination of online and offline learning.