Monday, June 4, 2012
Automated Store Replenishment – Volume XII
In the past, the question of Automated Store Replenishment (ASR) has been studied in the context that the entire process was to be managed by the supplier without the participation of the retailer (Vendor Managed Inventory—VMI). However, researchers obviously based their decisions on the assumption that retailers had become lazy, and suppliers and retailers would always maintain a one-way relationship between themselves, where inventory was ‘pushed’ to the retailer, and the retailer had no real say in the matter. This would have never worked because retailers do place orders, based on inaccuracies to be sure, but you cannot remove retailers from the equation entirely.
You could also deduce from their studies, their assumptions were grounded on the existing levels of technology prevalent at the time these studies were being conducted, and the environments between retailers and suppliers will forever remain unique and incompatible.
My research has proven just the opposite. Suppliers do not want the responsibility of determining what items are ‘pushed’ to the stores. Instead, they want retailers to submit orders, telling them exactly what to bring to their stores. This puts the onus on the retailer where it belongs and not on the supplier, and if the retailer has a complaint, he must first examine his own policies and shortcomings before placing the blame on someone else.
To understand this better, we must recognize when the suppliers’ trucks leave the warehouse, suppliers rarely make one delivery and then return to the warehouse to restock. Their routes consist of multiple deliveries and as the end of the day draws near, the last retailer on their route rarely gets the items, or the quantities to satisfy customer demand.
Sam Walton realized this early on, and demanded the company use its own fleet of trucks to make deliveries to his stores. Suppliers gained access to REAL-TIME DATA regarding product sales, and used that information to forecast future requirements. David Carter, a reporter for Baseline Magazine said it best when he wrote, “Wal-Mart has decided that sharing mission-critical information with its suppliers is in both their interest and their supplier’s interest.” The deciding factor that made all of this possible was the centralization of all of Walmart’s data through RetailLink. 10,000 suppliers accessing data in Walmart’s 2,500 stores (at the time the article was published), would have been a ridiculous exercise in futility—a nightmare of monumental proportions. None of this could have occurred otherwise.
Centralization or decentralization is a question that should be best decided according to where the data is ultimately needed. For example, if the data is most useful at the store level, decentralization is favored. If the data is required at headquarters, centralization is preferred, with the deciding factor resting on the location of the strongest information technology (IT) capabilities. Having all your data terminate at a central location provides you with the information you need to experiment and make decisions.
Traditionally, sales, manual ordering, stocking the shelves, accepting returns, handling cost and product damages are handled in the stores; while bookkeeping, analyses, upper-level decisions, pricing, purchasing, paying invoices, etc., are handled at headquarters. In circumstances where local store managers are not capable (or not allowed) to perform the latter efficiently, centralization wins hands-down.