This paper presents some attempts to represent public transport services that are actually experienced, rather than intended, and allow for fluctuations in the provision of scheduled services. The following quantifiable elements are included in passengers' perception of the level of service offered: (1) cancellation rate; and (2) punctuality in terms of adherence to a timetable. For various reasons, both these parameters can vary significantly between routes, and passengers are considered to perceive a corresponding range of different levels of service. The paper describes how London Underground Limited (LUL) has handled the issues of cancellations and unreliability in its large public transport network public transport network covering much of South East England; these issues have featured little in previous modelling, as their data are difficult to collect. LUL has also been pioneering station capacity modelling to complement in-vehicle congestion studies. Congestion is being studied at nodes as well as links. Virtual reality can assist planning for positioning and quantity of station facilities. By attempting to approach issues guiding passenger behaviour, improved model calibration can be expected. Applications are being made to specific LUL services.
Abstract