10: A Distributed Computing Approach for Transportation Modelling and Simulation: Application to an Airport System

This chapter introduces a modelling and distributed simulation approach for the analysis of transportation data flows. The research is based on a conceptual and logical mapping between a real-world system, in which the interest of study is the modelling of transportation data flows, and a distributed computing environment. The dynamic and static properties of a transportation system are modelled using a hierarchical-based approach at different levels of granularity. The mapping between the application and the computing level is ensured by a graph that represents a structural decomposition of the transportation system on the one hand, and the computing nodes of the distributed system on the other. Our computing platform supports time-driven distributed simulations whose dynamic properties replicate the behaviour of transportation data flows. One of the peculiarities of the platform relies in the concept of migration by group that supports the coordinated displacements of objects between the computers of the distributed system, thus replicating the transportation data flows identified at the application level. Simulation objectives are the analysis of emergent transportation behaviours that result from the non-linear interactions and displacements of mobile objects within the distributed system. The potential of our distributed computing approach is illustrated by a case study that models and simulates passenger flows between the different halls of an airport terminal.