Modelling Urban Bus Service and Passenger Reliability.

Abstract

Reliability of public transport systems has been considered critically important by public \ntransport users, operators and the government. Lack of reliability in public transport results in \nuncertainty and delays aggravating anxiety and discomfort for the passengers and increases \ncosts due to lost mileage and lower fleet utilisation for the operators. The government \nemphasises a reliable bus-service in order to ensure its competitiveness to car use such that \nit results in modal shift. Though reliability is of significance to a range of actors such as the \nbus operators and the passengers, its interpretations vary across different groups. For \nexample, quality of service identified by the demand-oriented indicators, the travellers, is \nsystematically different from what the supply-based quality indicators show. Reliability, as a \nmeasure of quality of service, is related to the operational characteristics of the public \ntransport system. Passenger’s wait time, on the other hand, is much more sensitive to \nschedule reliability than to service frequency. \nThis paper investigates some of the measures used to assess reliability of an urban bus \nnetwork using a dynamic microsimulation model framework. The framework incorporates \nmodels of bus operation, passenger arrival and boarding for buses within a network model of \ngeneral traffic. It therefore allows explicit modelling of the interactions among cars, buses \nand passengers and of their effect on bus reliability. The model framework is applied to a \ncase study based on a bus route in the city of York. Empirical analysis of the test route was \ncarried out to understand the background of reliability issues and to establish a sound base \nscenario of the network. The simulation model is used to help understand the causes and \nprocess of occurrence of unreliability and to study the impacts of possible scenarios such as \nincreased congestion, rise in passenger demand, and reduction in ‘per passenger boarding \ntime’ on the reliability of the test-route. The study measures reliability in terms of travel time, \nheadways and excess passenger wait time, and sheds light on factors affecting them. \nThe results of the study reveal that excess passenger wait time is strongly correlated with the \nvariation of headways and an increase in passenger demand intensifies the ‘bunching and \nspreading out’ of buses. With increasing congestion and passenger demand, reliability is \nreduced. Reduction in per passenger boarding time proves to be an effective measure in \nimproving reliability. \nBeyond these specific case study results, the modelling framework developed may serve as \na useful tool to assess reliability levels for a future year and to evaluate proposed strategies \nto improve reliability before implementation. Such a modelling framework, if extended to \ncover larger bus networks can serve as an important decision-making tool in the hands of the \ngovernment and bus operators for offering more reliable bus services to the public.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 65 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.