Creating win-win scenarios from the bullwhip problem-By design not accident
Abstract
We study the bullwhip problem in supply chains by relating it to the inventory levels and customer service provided by an Order-Up-To policy. We have analysed 16 real-world demand patterns from the consumer goods industry and matched them to the mathematical Auto Regressive Moving Average (ARMA) demand model. Using analytical results for the bullwhip metric, the variance of the inventory levels and a 99.5% fill-rate objective, we show that there are 4 possible scenarios to the bullwhip problem. We show that under some circumstances, we may avoid bullwhip and reduce inventory holdings simultaneously, a statement that is probably counter-intuitive. Sometimes extra inventory is needed to avoid bullwhip, or alternatively, we may choose to induce bullwhip because it allows further inventory gains to be achieved. Of course, with careless design, we may experience higher inventory holdings and unnecessary bullwhip.