Supply Chain science and tech
BACK TO LOKAD TV ›
Differentiable Programming in Supply Chain (Part 3/3)
Differentiable Programming is a hybrid perspective between statistical learning and numerical optimization. This mix is uniquely suited to deliver predictive optimization for supply chains, which also require a mix of learning and optimization.
Differentiable Programming in Supply Chain (Part 2/3)
Yann LeCun, the director of AI research at Facebook, argues that ‘Deep Learning’ has out-lived its usefulness and has coined Differentiable Programming as a fresh machine learning perspective. In particular, this perspective proves itself of prime relevance to address supply chain challenges.
Differentiable Programming in Supply Chain (Part 1/3)
Differentiable Programming is the descendant of Deep Learning. It has unlocked a series of challenges that were previously seen as unsolvable and has paved the way for considerable progress and superior numerical results in the world of supply chains.
Why DDMRP Is Fundamentally Flawed
Demand Driven Material Requirements Planning (DDMRP) is a multi-echelon planning and execution method. This technique is a further development of MRP and works through strategically placed decoupling points and stock buffers in a supply chain. It has been described as being ‘Built for People, Not Perfection’. In this episode of LokadTV we try and learn whether this method really works in practice and why.
The Problem With Flowcasting
Flowcasting has been previously described as 'The Holy Grail of demand-driven supply chain planning'. But just what is it exactly? 'Flowcasting the Retail Supply Chain', a book published in 2006, presents a series of techniques that were intended to revolutionize the retail industry. In this episode of LokadTV, we learn a little more about this concept and debate why a technique that was published in 2006 is still of interest today.
Data Security in Supply Chain
Data is both an asset and a liability. Supply chains require extensive historical records for tracability purposes and to ensure the accuracy of demand forecasts. However, data leaks are damaging events both for the company and its clients. Supply chains have to protect both their physical and software infrastructures.
Blackboxing and Whiteboxing
Any nontrivial demand forecasting model becomes a black box for supply chain practitioners, that is, an opaque subsystem that produces numbers that are difficult to understand and to challenge. Whiteboxing, as part of the Supply Chain Management practice, is the answer to this problem. Practitioners don't need to understand the 'how' but need to understand the 'why'.
Pricing Optimization and Supply Chain Management
Pricing optimization is typically not considered as part of the Supply Chain Management (SCM) practice. Yet, pricing is a factor that strongly influences customer demand. Thus both production capacities and stock levels are highly dependent on prices, and must be jointly optimized.
Data Lakes in Supply Chain
Data lakes are data storage technologies intended for bulk reads and bulk writes. They are particularly well suited to address supply chain challenges, because many situations require an inspection of the company's entire history of orders and stock movements.
POCs (Proofs Of Concept) Don’t Work For Supply Chains
Supply chains are complex systems made of many moving parts: goods, people, machines. POCs (Proofs of Concept) routinely fail when attempting Quantitative Supply Chain initiatives because problems get displaced instead of getting solved.
Why You Should Subscribe to LokadTV
Keep up with the latest changes in the Supply Chain industry.
Terabyte Scalability for Supply Chains
The relevant amount of historical data when considering large supply chains frequently exceeds one terabyte. As a result, inventory control requires two distinct flavors of software: transactional software (e.g. an ERP) to manage the resources, and predictive software (e.g. Lokad) to optimize the resources.