Dynamic Behavior Of Supply Chains

The bullwhip effect has been well known since many years and often takes place in supply chains. It is caused by wrong order policy in real systems. The bullwhip effect can be demonstrated easily through the beer game. It was developed by the MIT in the 60s. All stages of the supply chain operate independently. The only decisions that can be taken in the particular stages of this simulation game are the size of the orders. This beer game was depicted as a simulation with two additional simple control algorithms for the decision of the order quantities. The first algorithm set the time to compensate a difference in the inventory to a specific value, in the second algorithm the quantity of the order is limited. Both algorithms were taken to compensate a sudden and steady increase of the orders. Both control algorithms generated a bullwhip effect. The limited duration strategy for compensation a difference in the inventory led to rising orders in the upstream stages. The limited order strategy led to an increasing time to compensate a difference in the inventories of the upstream stages. The limited duration strategy was applied to an ordering behavior with a linear trend und to a random ordering behavior. For the linear trend it has been found that the shorter the time for compensation the lower the difference to the nominal stock and the higher the bullwhip effect in the stages upstream. In contrast to that for random orders the time for compensation has to be longer to minimize the bullwhip effect. That is why the setting of the closed-loop control of the stocks in supply chains is an optimization problem. It can only be solved for a real supply chain, if the ordering behavior of the customer is known.