This final blog on multi-echelon “push” and “pull” will summarize the uses and differences between the two techniques discussed throughout the series.
Inventory management in a multi-echelon channel environment is determined by the structure and objectives of a firm’s distribution network. Companies create a channel of distribution because they have determined that the best way to serve the marketplace is to have inventory close to the customer. The goal is to determine the best method of ensuring the highest level of customer service while pursuing lowest inventory and transportation cost alternatives. There are basically two ways to replenish inventory in a distribution channel: push or pull.
In a “push” system, channel resupply activities are conducted by the supplying facilities. Channel resupply is performed first by aggregating the historical demand from all lower echelon stocking points. Forecasting methods are then applied, aggregate quantities determined, and replenishment inventory pushed down the channel from the originating supply facility using some form of allocation algorithm.
In contrast, in a “pull” system each downstream facility performs its own inventory planning and determines the timing and quantity of replenishment from designated upstream channel supply points. When a “pull” system strategy is selected, inventory managers can choose from two possible techniques: reorder point (ROP) or distribution requirements planning (DRP). The decision to choose a pull technique depends on how inventory moves through the channel and the required level of control.
A ROP technique is mostly chosen by channels that stock finished goods acquired from outside suppliers. When the reorder point of any facility in the channel is triggered, a resupply order is created and sent to the designated supply facility (or the outside supplier). A key factor to note in the selection of the ROP technique is the relative shortness (expressed in days or even less) of the replenishment lead times separating each channel facility. Outside of some order preparation time, almost the entirety of lead time in a ROP-driven channel is composed of transportation time. ROP works because the resupply lead times necessary to replenish inventory in the supplying facility are the same or less than the lead time to transport those items from the supplying to the satellite facility.
For distribution channels supplied from an intracompany production plant or have long lead time items made-to-order at outside suppliers, planners would most likely choose the DRP technique. The decision to use DRP is purely one of lead time. Unlike the short lead times characteristic of a ROP supply channel, DRP is best used to manage the much longer manufacturing lead times required by production plants at the beginning of the supply chain. DRP works well with the time-phased manufacturing resources planning (MRP) systems used to plan production. By providing a schedule of planned demand that extends out into the planning horizon, DRP is able to directly input actual and anticipated demand into the production plant’s master schedule. As a result, planners at the production plant have the necessary visibility to total channel demand through the planning horizon so that component inventories are purchased and production orders are started and completed on time. As a note, even companies on a ROP system can use DRP as a simulation tool to plan future inventory requirements and costs.
Choosing a multi-echelon replenishment system requires a firm understanding of the needs of the supply chain, the objectives of the organization, and the nature of the product. In the final analysis, the objective is to ensure the highest customer service, at the lowest cost, with the widest margin of profitability for all members of the supply chain.
