Defining Demand Forecasting Demand forecasting as the term suggests is predicting the need for a...
Stock replenishment is a routine activity in most retail stores, warehouses and other catchment areas. It can be defined as the process of ordering stock to cover the demand (with some safety cover) for the next period. The definition of “period” and quantity to order depends on the products, supply chain configurations, time it takes to fulfil the order and whether the order is from a store or a warehouse. Consider the example of a hypermart, which typically would have a large selection of daily use items, food and grocery. Depending on the perishable nature of the food items, the replenishment requirement might be a couple of days. Whereas for a slow-moving item which has long shelf life the replenishment might be done in a larger time cycle. This, of course, is an oversimplified view of how replenishment works.
Different strategies are deployed for replenishment by different retailers, even a retailer might have different strategies for replenishment for different products. However, there are some common strategies:
Before the products are placed and sold in stores, they go through multiple hops, which includes raw material procurement, manufacturing, specialized job works, warehouses, distribution centres and finally stores. This is a description of a typical supply chain, the configuration and complexity of these interactions might vary. No matter how good a replenishment strategy is, it would not work if all the hops work in synchronization. A 100 stores asking for a product with variable demand, might overburden a warehouse if it is not adequately stocked. So to be run a smooth replenishment cycle, it makes sense to have a replenishment strategy for each hop of the supply chain. The best method to do that especially for fashion retailers (which has high variability in demand) is to implement a multiple horizons forecasting system. Multiple horizons demand forecast lends itself pretty well to planing for each product at each store, with backward integration with the rest of the supply chain. Because each of the hops in the supply chain has different lead time (a warehouse would have a longer lead time than a store), the multiple horizon demand forecasts relate directly to the demand at each hop.