Complex behaviour in a production–distribution model

The purpose of this paper is to show how the cascaded structure of a production–distribution chain can produce a wide variety of dynamic behaviours. Despite the apparent simplicity of the chain, the bounded rationality of human decision making interferes with the information feedbacks, nonlinearities and time delays of the system to produce exceedingly complex behaviour. When simulated with realistic parameter values, the model can exhibit stationary periodic, quasiperiodic as well as chaotic and hyperchaotic motions. Lyapunov exponents are used to characterize the different types of behaviour. Repeated simulation reveals how the profile of operating cost is related to the modes of behaviour and to the conditions that generate chaos. Detailed maps of parameter space graphically show the influence of inventory control policies upon dynamics and costs. The analyses illustrate why it is so difficult for decision makers to ‘navigate’ in low-cost policy parameter regions. Suggestions are made for decision heuristics that avoid high-cost, unstable solutions.