There is a multitude of Supply Chain Planning solutions on the market, and the approach remains fairly similar, with the possibility of using forecasting algorithms based on classic time series methods or more recent Machine Learning techniques. The effectiveness of each approach depends on the use case: hierarchy level, product, location, and forecasting objectives (replenishment or store/warehouse… allocation.)

Regarding replenishment and allocation, we have encountered solutions that are based on a complete optimization of the logistics network, including a suggestion of all transfers to consider: supplier → warehouse, inter-store, warehouse → store.

Through our experience over the past 15 years, we have found that choosing a solution should prioritize scalability, ease of use, and the quality of support (especially for SaaS). Complex models that are difficult to understand do not help business teams improve their efficiency.

Today, it is also possible, thanks to tools from Microsoft Azure, Google Cloud, AWS, and others, to create custom models through smaller, human-sized partners. This is an interesting alternative for some retailers or CPGs who prefer to avoid packaged solutions.