Beer game psychology

Lessons of the game

During the game emotions run high. Many players report feelings of frustration and helplessness. Many blame their teammates for their problems; occasionally heated arguments break out. After the game I ask the players to sketch their best estimate of the pattern of customer demand, that is, the contents of the customer order deck. Only the retailers have direct knowledge of that demand. The vast majority invariably draw a fluctuating pattern for customer demand, rising from the initial rate of 4 to a peak around 20 cases per week, then plunging.

"After all, it isn't my fault", people tell me, "if a huge surge in demand wiped out my stock and forced me to run a backlog. Then you tricked me - just when the tap began to flow, you made the customers go on the wagon, so I got stuck with all this excess inventory." Blaming the customer for the cycle is plausible. It is psychologically safe. And it is dead wrong. In fact, customer demand begins at four cases per week, then rises to eight cases per week in week five and remains completely constant ever after.

This revelation is often greeted by disbelief. How could the wild oscillations arise when the environment is virtually constant? Since the cycle isn't a consequence of fickle customers, players realize their own actions must have created the cycle. Though each player was free to make their own decisions, the same patterns of behavior emerge in every game, vividly demonstrating the powerful role of the system in shaping our behavior.

Research reported in Sterman (1989) shows how this occurs. Most people do not account well for the impact of their own decisions on their teammates - on the system as a whole. In particular, people have great difficulty appreciating the multiple feedback loops, time delays and nonlinearities in the system, using instead a very simple heuristic to place orders. When customer orders increase unexpectedly, retail inventories fall, since the shipment delays mean deliveries continue for several weeks at the old, lower rate. Faced with a growing backlog, people must order more than demand, often trying to fix the problem quickly by placing huge orders. If there were no time delays, this strategy would work well. But in the game, these large orders stock out the wholesaler. Retailers don't receive the beer they ordered, and grow increasingly anxious as their backlog worsens, leading them to order still more, even though the supply pipe line contains more than enough. Thus the small step in demand from four to eight is amplified and distorted as it is passed to the wholesaler, who reacting in kind, further amplifies the signal as it goes up the chain to the factory. Eventually, of course, the beer is brewed. The players cut orders as inventory builds up, but too late - the beer in the supply line continues to arrive. Inventories always overshoot, peaking at an average of about forty cases.

Faced with what William James called the "bloomin', buzzin' confusion" of events, most people forget they are part of a larger whole. Under pressure, we focus on managing our own piece of the system, trying to keep our own costs low. And when the long-term effects of our short-sighted actions hit home, we blame our customer for ordering erratically, and our supplier for delivering late. Understanding how well intentioned, intelligent people can create an outcome no one expected and no one wants is one of the profound lessons of the game. It is a lesson no lecture can convey.

The patterns of behavior observed in the game - oscillation, amplification, and phase lag - are readily apparent in the real economy (figure 4), from the business cycle to the recent boom and bust in real estate. The persistence of these cycles over decades is a major challenge to educators seeking to teach principles and tools for effective management. Though repeated experience with cycles in the real world should lead to learning and improvement, the duration of the business cycle exceeds the tenure of many managers. In real life the feedback needed to learn is delayed and confounded by change in dozens of other variables. By compressing time and space, and permitting controlled experimentation, management flight simulators can help overcome these impediments to learning from experience.

But the biggest impediments to learning are the mental models through which we construct our understanding of reality. By blaming outside forces we deny ourselves the opportunity to learn - recall that nearly all players conclude their roller coaster ride was caused by fluctuating demand. Focusing on external events leads people to seek better forecasts rather than redesigning the system to be robust in the face of the inevitable forecast errors. The mental models people bring to the understanding of complex dynamics sytematically lead them away from the high leverage point in the system, hindering learning, and reinforcing the belief that we are helpless cogs in an overwhelmingly complex machine.