Assessing the spatial impacts of unreliable public transport systems: A quasi real-time data-driven approach

Sustainable transportation networks require monitoring systems and corresponding sets of performance indicators (PI). Next to network reliability also accessibility and network capacity are well-known and important properties which have to be addressed by such indicators. On the network level, corresponding PIs are required to be easily interpretable to fulfill their purpose of informing operators and local authorities about the network’s performance and its effects on users’ mobility. Moreover, a simplistic formulation will increase transparency and reduce maintenance costs for corresponding monitoring and evaluation systems. Emerging technologies such as automated passenger counting (APC) and automated vehicle location (AVL) systems can provide necessary data to derive performance indicators automatically in real-time for transport systems. This paper introduces a network reliability indicator which computation is based on open data, data which is usually accessible to PT operators and non-commercial software. The methodology can thus readily be implemented and its relation to the well-established accessibility indicator allows for a fast interpretation. It does not only show a quantitative measure for reliability but also its spatial impact can be evaluated. This allows analyzing if critical points of interest such as train stations can be reached also when PT services are unreliable during congestion. Potential traffic management measures or transport policies can be evaluated with respect to that. Also, the fact that it can be calculated in quasi real-time allows for an immediate evaluation of certain bus control strategies in case of severe delays in the system.