How to Efficiently Distribute Scarce Resources: Volatile Decision Behavior and Optimal Guidance through Information Services

Abstract

The coordinated and efficient distribution of limited resources by individual decisions is a fundamental and unsolved problem. When individuals compete for road capacities, time, space, money, etc., they normally take decisions based on aggregate rather than complete information, such as TV news or stock market indices. In related experiments, we have observed a volatile decision dynamics and far-from-optimal return distributions. We have also identified ways of information presentation that can considerably improve the overall performance of the system. In order to determine optimal strategies of decision guidance by means of userspecific recommendations, we have developed a stochastic behavioural description. These strategies manage to increase the adaptibility to changing returns and to reduce the deviation from the time-dependent user equilibrium, thereby enhancing the average and individual payoffs. Hence, our guidance strategies can increase the performance of all users by reducing overreaction and stabilizing decision behaviour. Our results are highly significant for predicting decision behaviour, for reaching optimal behavioural distributions by decision support systems, and for information service providers.