The impact of network connectivity on performance in production logistic networks

Recent publications on traffic control in urban road networks have presented strategies for adaptive flow control under varying network load. Traffic lights are modeled as independent, periodic devices, which take decisions on a local level. Despite the not actively coordinated decisions at single network nodes, such a strategy can perform better in terms of waiting time for the whole system than applying a standard optimization approach. First results from an ongoing simulation study indicate that adaptive control algorithms can also be applied successfully to different network topologies, e.g., logistic networks. This article presents the outcome of an extended simulation study. The simulation experiments have been carried out on artificially generated networks as well as networks derived from real manufacturing environments. We are able to show that the promising findings from traffic control regarding waiting time reduction and the emergence of synchronized behavior can be reproduced for production logistics. Furthermore, we illustrate how the variance of the network degree as an indicator for network connectivity influences the logistic performance.