Predicting traffic volumes and estimating the effects of shocks in massive transportation systems

摘要: Public transportation systems are an essential component of major cities. The widespread use smart cards for automated fare collection in these offers a unique opportunity to understand passenger behavior at massive scale. In this study, we network-wide data obtained from the London transport system predict future traffic volumes, and estimate effects disruptions due unplanned closures stations or lines. Disruptions, shocks, force passengers make different decisions concerning which enter exit. We describe how changes lead possible overcrowding model will be affected by given disruptions. This information can then used mitigate shocks because authorities may prepare advance alternative solutions such as additional buses near most stations. statistical methods that leverage large amount smart-card collected under natural state system, where no take place, variables indicative find features extracted regime successfully exploited disruption regimes, our framework general tool any similar complex system.