Line capacity in metro and high-frequency suburban railways is as much determined by station stop times as by factors such as line speed or train acceleration. This paper applies the method developed by London Underground to estimate the time that trains spend at stations, as a function of the physical characteristics of the situation (e.g. train door width) and the numbers of passengers involved. Analysis was carried out on a number of alternative designs for refurbishment of South West Trains' Class 455 inner-suburban rolling stock. Whilst there is indeed an interaction between boarding and alighting passengers, this paper demonstrates that the LUL relationship breaks down at the highest passenger loads. Moreover, results indicate that passenger flow is not equal between different parts of the same group of boarders or alighters.

Although there is considerable engineering and demand planning analysis in the development of urban railways around the world, less attention is paid to the understanding of key operating conditions such as station stop times. The current paper takes forward research by London Underground, and shows that it is applicable to situations around the world without substantial changes in parameter values or the resulting passenger flowrates. Where passenger flow is substantially different, this can generally be explained by particular differences in the physical characteristics of the site, rolling stock, or passenger demand.

The reliability of train service is a key objective of metro management and a major part of a successful operation. An occurrence of incidents in the network is likely to cause delays to the train service and disorder in the punctuality and regularity of the metro operation and hence affects the service reliability of the metro system. This result suggests a way to improve train service reliability by reducing the occurrence of incidents in urban metro systems. This paper used statistical techniques to identify the main factors that explained the variation in the number of delay incidents across 42 metro lines of 15 different metro systems from 2005 to 2009. The main factors that explained differences in incident performance across urban metro lines were the technology of the mode of train operation, the level of passenger demand, the service level operated during peak periods, and the practical capacity available. In contrast, engineering and usually fixed factors such as the type of track support, the type of rail connection, and the type of rolling stock wheel did not affect the level of incidents. The findings also suggested that metro-specific factors (e.g., differences in maintenance and management practices, operations management, and health and safety procedures) helped to explain the variation in incident performance across urban metros.

No abstract

No abstract