摘要：Road networks are complex interconnected systems. Any sudden disruption canresult in debilitating impacts on human life or the economy. In particular,road systems in mountain areas are highly vulnerable, because they often donot feature redundant elements at comparable efficiencies. This paper addresses the impacts of network interruptions caused bylandslide events on the (rural) road network system in Vorarlberg, Austria. Based on a landslide susceptibility map we demonstrate the performance ofagent-based traffic modelling using disaggregated agent data. This allowsus to gain comprehensive insights into the impacts of road networkinterruptions on the mobility behaviour of affected people. Choosing anagent-based activity-chain model enables us to integrate the individualbehavioural decision-making processes into the traffic flow model. Thedetailed representation of individual agents in the transport model allowsoptimisation of certain characteristics of agents and including theirsocial learning effects into the system. Depending on the location of the interruption, our findings reveal mediandeviation times ranging between several minutes and more than half an hour,with effects being more severe for employed people than for unemployedindividuals. Moreover, results show the benefits of using agent-based traffic modellingfor assessing the impacts of road network interruptions on ruralcommunities by providing insights into the characteristics of the populationaffected, as well as the effects on daily routines in terms of detourcosts. This allows hazard managers and policymakers to increase theresilience of rural road network systems in remote areas.
