New solutions have emerged that bridge the gap between supply chain planning and execution. These solutions have a middleware layer that is critical. The middleware allows for a feedback loop that can allow the system to get smarter.
While speaking with a number of WMS product managers, I began to notice that machine learning was called out as a focal point for WMS product development efforts. In general, machine learning is a hot topic in the world of…
Demand planning is a good application for machine learning because the measure of success – the forecast accuracy – is clear. To learn, an application needs a clear measure of success. Having a clear measure of success sounds easy. But…
Machine learning has been successfully applied to demand planning, but leading suppliers of supply chain planning are beginning to work on using machine learning to improve supply planning. But architecturally and culturally, this is a much tougher problem than machine learning applied to demand planning.
There is an arms race to incorporate artificial intelligence in demand planning solutions. Many new data sources, features, and tools are being explored. A new demand modeling tool has been introduced which will make it easier to analyze new data sources to see if they can be used to improve forecasts.
Machine learning engines can take data on forecast accuracy and use that data to automatically improve the forecast model. However, three is a rub. Lost sales is a key piece of data on the accuracy of the forecast, but lost sales is generated with a demand forecast. This is circular reasoning. But that does not mean this analysis is without value.
