The purpose of this game is to study the behavior of firms in a supply chain. Students are embedded in four person groups where each student takes a unique role in a supply chain. Each group contains one producer, one distributor, one wholesaler, and a retailer. Goods flow downstream: factories sell to distributors, distributors sell to wholesalers, and wholesalers sell to retailers. On the other hand, purchase orders flow upstream through the supply chain. The retailer has information on consumer demand. This informs purchase orders submitted to the wholesaler, which in turn influences wholesaler purchase orders submitted to the distributor, and so on.
Each firm in the supply chain seeks to minimize the cost of holding inventory (Carry Cost) and penalties from not having enough product to supply their customer (Backlog Cost, e.g. reputational costs from unmet demand). The goal of each firm is for each firm to minimize the costs they incur.
Demonstrate that sudden changes in customer demand generate larger and larger changes in order submissions upstream through the supply chain.
Show the consequences of the bullwhip effect on total supply chain costs.
Identify the underlying causes of the bullwhip effect through manipulation of game parameters (e.g. logistics improvements that decrease time delays).
Each student is assigned a role as producer, distributor, wholesaler, or retailer. Students participate in fixed roles and groups across a fixed number of rounds. In this setting, each round is the equivalent of a week. Regardless of role, rounds proceed in the following way:
You can set the quantity of Initial Demand and the duration (i.e. Rounds or “weeks”) for which retail customers demand that quantity. In the first round, the incoming delivery equals the Initial Demand. This remains true until the firm receives their first placed order. All firms begin the game with zero inventory. The instructor can set Demand Shocks by adding new weeks at lower or higher demand.
With our default parameters firms incur a Carry Cost of $1 per unit for holding inventory and a penalty of $2 for each unit of their customer’s demand (the downstream firm they deal with) they are unable to meet. The goal for each firm is to minimize these costs.
Manipulate the cost of maintaining inventory or the penalty to unmet demand.
Alter the amount of time it takes for an order to travel upstream and goods to be delivered downstream. Longer delays amplify the bullwhip effect, as firms must attempt to forecast demand over a longer time horizon and seek to avoid more costly backlog costs.
The bullwhip effect is characterized by how sudden changes in consumer demand generate larger and larger changes in order submissions upstream through the supply chain. This is best shown in the following two graphs. Figure 1 plots order submissions across rounds, and is color coded by firm type. A dashed vertical line between rounds indicates a demand shock.
Because there is a delay in receiving newly placed orders firms dip into available inventory and move deeper and deeper into backorders before those larger deliveries arrive. You can show this with Figure 2, which plots a line-graph of inventory across rounds for each role (negative inventory indicates there are backorders). As before, firm types are color coded and dashed vertical lines between rounds indicate a change in customer demand.
Table 1 is organized by role and shows information on orders, as well as inventory. From left to right the roles go downstream to upstream to organize discussion about how decrease in customer demand information leads to higher average orders, higher “largest orders”, greater numbers of “small orders”, and a higher variance in inventory.
Figure 3 plots cumulative total cost across rounds for each supply chain. Total cost refers to all the carrying costs for inventory, and the penalties from unmet demand incurred by all members in a supply chain in a given round. You can choose to focus on one group, a subset of groups, or all the groups by using our radio buttons. This graph should help to facilitate comparisons between total costs across different parameterizations, but also strategies for attenuating the bullwhip effect.