As the complexity and features of bus and rail electronic ticketing machines (ETMs) and associated management software have evolved, the available data have become increasingly useful as input data to the development of public transport demand matrices. Initially data were of poor quality but the rigours of market disciplines in the transport industry have improved the level of detail and accuracy. Some software now allows origin-destination details to be recorded rather than just boarding point information. MVA has been at the forefront of using of ETM data in a number of studies where the data have been at differing levels of quality and detail. This has necessitated the development of a number of parallel techniques, particularly those for merging ETM data with data from other sources. This paper demonstrates a number of possible solutions, drawn from a number of case studies, when working with data over a typical range of data-types: (1) Dublin - bus boarding information on a prescribed farescale which makes estimation of a lighting point relatively easy given integrated ticketing and rail journey information; (2) Leicester - boarding-only information in a deregulated environment where fares are charged on a service by service basis and city centre surveys were necessary to pick up the bulk of interchanging passengers; (3) Bristol - where boarding and alighting data were available directly from the ticket equipment for individual journey legs; and (4) Singapore - where a magnetic card system has been in use since 1990 following MVA's involvement in the design, procurement and implementation process and where full through-journey information has been used as part of a matrix update facility for data originally gathered by survey.
Abstract