Abstract
A supply chain is composed of all the stakeholders and processes involved in satisfying consumer demand: wholesaler, retailer, warehousing, transport and so on. A classic method to understand the internal workings of a supply chain is the much-used beer distribution game that came out of MIT during the sixties. In this game, each player takes on the role of one of the members of the chain (consumer, retailer, wholesaler and manufacturer). The aim is for each of them to coordinate their actions in such a way as to satisfy the demands of the upstream member of the chain at the least possible cost. Sterman (1989) provided evidence of an effect that had already been described by Forrester (1961) whereby initial consumer demand is distorted and amplified as it passes along the chain. This increment is known as the Forrester or Bullwhip effect.
Key Terms in this Chapter
Bullwhip Effect: The variability in demand is magnified as we move from the customer to the producer in the supply chain
Beer Game: There are four levels (retailer, wholesaler, distributor and manufacturer) as well as the purchaser. Each player simulates being one level of the supply chains, so the purchasers order from a retailer, who has to decide how much to order from his supplier, given the stock that he has. This supplier in turn has to order from his distributor, who has to order from the factory.
Agent: An agent is anything that can perceive its environment through sensors and act on environment through effectors... An ideal rational agent performs a sequence of actions (life output history) which maximizes the expected and performs measure given evidence of the percept sequence and built-in knowledge... additional vital trait: autonomy, ability to learn.
Control Theory: In engineering and mathematics, control theory deals with the behaviour of dynamical systems over time. The desired output of a system is called the reference variable. When one or more output variables of a system need to show a certain behaviour over time, a controller tries to manipulate the inputs of the system to realize this behaviour at the output of the system.
ARIMA Models: An acronym for the auto-regressive integrated moving average models used in the Box-Jenkins forecasting procedure.
Genetic Algorithm: A genetic algorithm (GA) is a heuristic used to find approximate solutions to difficult-to-solve problems through application of the principles of evolutionary biology to computer science. Genetic algorithms use biologically-derived techniques such as inheritance, mutation, natural selection, and recombination (or crossover). Genetic algorithms are a particular class of evolutionary algorithms.