There are a number of causes of the inventory/service crisis described in HP: •The first cause was the demand variability of printers. For example, the demand for model A in Europe had the monthly standard deviation of 32. 4 while the monthly mean was 42. 3. Combined with increasing pressure from resellers that forced HP to offer the resellers flexible distribution, fluctuations in demand resulted in the high levels of safety stocks and high holding cost. •The large variations in printer models (i. e. with different power supply module) in Europe form another issues for the inventory crisis as different models require customized assemble.
The sum of individual models increases the safety stock significantly (as opposed to one universal model with the same demand). •The lack of the demand forecast accuracy imposed a negative impact on the distribution process. In fact, the target inventory levels at the DCs were set based on rule of thumb. The inaccuracy made it difficult for HP to track down the appropriate amount of inventory. •The lack of scientific approach regarding performance measures for the DCs was another cause of the crisis. For example, as for the target line item fill rate, the target (i. e. 98%) was not decided from scientific estimates but from marketing. The long lead-time in Europe and Asia DCs was a burden for HP in quickly reacting to unexpected changes in demand. Due to ocean transit and the procedures at port of entry, it took 4-5 weeks to send printers to Europe and Asia. Therefore, in order to correspond to customers’ needs, HP had to keep substantial amount of safety stocks. 2. INVENTORY REPLENISHMENT The Deskjet printers are dispatched to European Distribution Centre (DC) in six different options and are be shipped from the Vancouver plant by either sea shipment or air shipment, with five weeks and three days lead time, respectively.
In order to assess safety stock, cycle stock and pipeline stock associated with the two shipment methods, a periodic review model is used with a condition that the replenishment review is being taken place at a weekly interval. The following assumptions were used for the stock modeling: -The periodic demand for each computation cycle is sum of demand during review time (DDT) and demand during lead time(DDLT) -The daily demand distribution follow the monthly demand distribution -Cycle and pipeline stocks are averaged based on the periodic demand -Weekly replenishment to meet 98% service level
A summary of the computation for 5-week sea shipment and 3-day air shipment are shown in Figure 1 and Figure 2. Detailed excel computation is included in Appendix 1. It can be seen from the computation that the inventory levels in terms of safety stock, cycle stock and pipeline stock using the 3-day air shipment is significantly less than those using the 5–week sea shipment. This is mainly due to the fact that a longer shipment time is required for the sea shipment and therefore a higher Order-up-to-level (S) and safety stocks are needed to ensure a high service level.
