In CRISP: In DETAIL: I've attached a PowerPoint that we give to the students when we start the Deicer and Solvent exercise. At this point, the class has been divided into teams consisting typically of 3-4 students. Each team is a company in the Deicer and Solvent industry.
The students start by assigning themselves roles within their companies; they can decide which roles to create within their firm and who should play each role. The roles are typically President, Director of Marketing, Director of Information Systems, etc. This does not affect the design of the program you are creating; it merely gets them thinking about how businesses are organized.
We give them about 10-15 minutes to get organized and think about various possible strategies. Most teams produce both deicer and solvent but some focus on just one. Some try to corner the market on one selected liquid so as to limit the other teams in their efforts to produce one of the products. Some teams become very aggressive; others see the benefits of cooperation, even with other firms in the same industry.
Then we open the market to allow sales of the liquids. Here is where linear programming comes in. They use the LP to find their optimal (profit-maximizing) product mix and then they use the associated sensitivity analysis to determine their firm's shadow prices (internal values) for each of the four liquids (raw materials).
The idea is that if a team's shadow price for a given liquid is low (relative to its market prices), it should consider selling some (or all) of that liquid; its revenue from sales will exceed the revenue from whatever increased production the liquid could be utilized. This may happen if the team's stock of that liquid is high; they can sell off the excess without hurting production much.
But if a shadow price for a liquid is high, that means that the liquid's value to the production process is likely to exceed its market price in which case the firm is better off using the liquid for production rather than selling it.
Since different teams have different shadow prices for a given liquid, the market will generally have both buyers and sellers of each liquid. The team's advertising (written on a blackboard) will help them find one another. A buyer might write on the blackboard "Team 3 is looking for Liquid Q!" while a seller might write "Team 7 wants to sell Liquid Q. Best offer!" These teams can then enter into a negotiation.
Team 3 wants to buy Liquid Q at a price that is lower than its shadow price for Liquid Q while Team 7 wants to sell Liquid Q at a price that is higher than its shadow price for Liquid Q. If they can work out such a deal, BOTH teams will increase their overall profit (profit from production PLUS profit from sales).
The fact that both teams do better is a revelation to most people. How can BOTH firms end up better off after such a deal? Don't all deals have a winner and a loser?
The answer is "No, some deals but not all deals." It is true if the deal is a "zero sum" deal; whatever one side gains equals whatever the other side loses. But in our game, the value of a gallon of a certain liquid may be higher for one side and, at the same time, be lower for the other. That creates an opportunity for BOTH teams to improve their profit simply by executing a well-designed sale.
If you have two cars and I have none, and if we each need only one car, it is in both our interests for you to sell me one of your extra cars. This is not a zero-sum game.
The students can see this in real time because, when they make a good deal, BOTH teams end up with increased wealth. We point this out to the class as soon as a (good) deal is executed -- "Hey, look everybody! Team 3 and Team 7 made a deal and BOTH of them ended up increasing their profit!" Quickly, the students catch on,the market gets very active and the room gets quite loud!
We often continue the market in the next class meeting, giving the teams time to rework their strategies if they choose and to plan what deals to go after in the next class. Ultimately, the room gets quiet and the dealing stops.
Then we close the market and discuss the results. It mostly happens that every team has increased its profit relative to its starting value but it happens that some teams actually lost money. We ask such a team to figure out what it did wrong. It always boils down to not understanding what the shadow price means and how to use it. They can usually identify which trade(s) were mistakes. We point out that even teams that increased their profits may also have made some "bad" deals but figured out why when they saw their profit drop after the deal and improved their deal-making.
The hope is that, at the conclusion of the game, they all have a greater appreciation for linear programming as an important business tool and that they more thoroughly understand the concept of shadow price. And they should also have a better understanding of the complexity of a business strategy formation.
Personally, I hope that they better understand the difference between zero-sum deals (my gain is your loss) and non-zero sum deals (we can both be better off). When we show them that the entire industry is more profitable, we explain that this is how an economy can grow through cooperation rather than cut-throat competition. And we remind them that everyone is better off and we did not need more resources for that to happen. We just needed cooperation!
Thanks for the help in getting this game online!