Beer Game Order Policy Optimization Under Changing Customer Demand

The Beer Game is a classic supply chain problem in which participants, playing the role of inventory holders, have to decide how many cases of beer to order. To win the game the participants have to develop an order policy that ensures enough goods to satisfy customers’ demand but, in the mean time, minimises the cost associated with stock maintenance. The main guide for their order policy is the Expected Demand (ED), but other important parameters are the rates of adjustment for the inventory and the supply line. The purpose of the present work is to analyse the optimal beer game order policy when the customers demand increases. Two different situations are analysed, depending on whether the sectors apply identical or different order policies. The optimal policy is found by means of a genetic algorithms (GAs) technique. GAs are specially suited for this problem because of the high dimension of the search space, mainly in the second scenario, and because the objective function i.e. the global score of the chain, has many local minima. Our results show that the best performance of the chain, for each change considered in the customer demand, is obtained when the sectors have different order policies. Moreover the advantage increases with increasing the change in the customer demand.

STROZZI Fernanda;  BOSCH PAGANS Jordi;  ZALDIVAR COMENGES Jose'; 

2007-02-05

ELSEVIER SCIENCE BV

JRC33542

https://publications.jrc.ec.europa.eu/repository/handle/JRC33542,   

10.1016/j.dss.2006.06.001,