In this paper we analyze the performance of the main models developed to quantify the Bullwhip Effect (BE) focusing on the impact due to the adoption of different policies to treat inventory excesses, under the assumption of stochastic demand and lead time. In addition to the policy assumed in most BE quantification models available in the literature, where inventory excesses are handed back to suppliers at no cost, two alternative policies are investigated: (i) one where excesses are disregarded from the analysis, not motivating product devolution, and (ii) another where excesses are adjusted and used to supply for future demands. Policy (ii), considered as the one best reproducing the operational reality of companies, is incorporated to the BE modeling as a correction factor when quantifying the phenomenon. We thus present a new mathematical model to quantify the BE in situations where both demand and lead time are stochastic, and where inventory excesses are considered in the determination of future orders. The development of the new model is presented along with computational simulations run to compare its performance with models in Chen et al. (2000) and Fioriolli & Fogliatto (2007).
Bullwhip effect; quantification; stochastic modeling; stochastic lead times
