The development of pervasive personal digital devices such as phones, watches, and headphones, interconnected by technologies such as Bluetooth, has led to a disruptive change in the ways in which local governments can monitor both vehicular and pedestrian traffic flows within their cities. In modern vehicles, navigation systems interconnect the personal devices of drivers and passengers typically via Bluetooth. By continuously monitoring such devices when they are in discover mode, traffic flows can be estimated almost in real-time. This paper examines traffic data collected from a Bluetooth Traffic Monitoring Systems installed by the Wellington City Council. Potentially such an installation could assist a local authority in real-time monitoring of normal traffic, as well as events including concerts and sport events, or in reaction to unanticipated events such as disasters. The limitation in this technology is that the data collected is of low fidelity, that is: not every vehicle has a detectable device, some have many, and there are devices carried by pedestrians and non-motor vehicles as well as stationary devices. This paper enumerates and investigates these challenges through statistical modelling, cleaning and data analysis. We present two novel algorithms for the processing of Bluetooth traffic data and validate our algorithms against a physical road counter. A case study is of a major earthquake is then presented as a proof of concept. The earthquake led to road closures, building collapse and other infrastructure damage and we examine three weeks of BTMS data and visualise how this earthquake impacted daily traffic flows.

In this paper we present a novel and comprehensive simulation framework that we have named AMEM (A Multidisciplinary Evacuation Model) for vehicle traffic modelling in urban areas – with a specific focus on large-scale evacuation scenarios. In general, the value of a comprehensive urban traffic modelling system is that it can assist authorities in identifying parts of a road transportation network that exhibit poor performance, or unanticipated and negative emergent properties under a variety of conditions. Such conditions can arise from planned or projected changes to the road infrastructure, or more interestingly, in reaction to uncommon or rare 100 year events. These are not the typical day to day traffic events that can be monitored and measured directly. In AMEM, we combine a number of different elements in our modelling, including routing, car-following, behaviour, driving culture, traffic light signalling, and psychological patterns. We validated the AMEM framework using real traffic data harvested from a network of Bluetooth and road sensors deployed in Wellington, New Zealand, and used this data as the basis for 13 scenarios. We also included in our study a unique socio-technical factor – the use of navigation systems, in part to address the question as to if such systems help or hinder traffic movement during an evacuation. Our results suggest that the use of navigation systems, as currently implemented, have a potentially negative impact on the evacuation process in dense urban areas.