ABSTRACT
The evaluation and monitoring of transport systems is critical for planning and for supporting decision making related to a city and its transport system. Traditionally, these procedures have been supported with survey data that contain detailed information based on respondents’ declarations and measurement data of specific aspects of the transport systems of interest. However, due to the high costs of surveys and measurements, these methods are typically sparse in terms of time and space coverage. Conversely, the incorporation of technological devices in the daily operation of a public transport system has provided significant
quantities of passive data. These databases are provided (and limited) by the technological devices available, which generate large databases with regard to specific aspects of passenger and vehicle movement. Additionally, because these devices require limited human intervention, they are usually particularly reliable in what they detect; however, what the technological devices detect is not necessarily what planners and decision makers require. The objective of this book chapter is to show that it is possible to estimate the level of service, mobility and accessibility indicators using passive data. Specifically, from a real case in the public transport system of Santiago, Chile, we will focus on the use of passive data from automatic fare collection (AFC) and automatic vehicle location (AVL) systems that are complemented with geographic information system data (GIS). Bagchi and White (2005) recognise the potential of smartcard data to analyse boarding transactions, time-space distribution and turnover rates over time; however, they have also identified a series of problems. In AFC systems where passengers are not required to tap their cards when exiting the transport system, trip destination information is not available. In addition, no journey purpose, attitudes or quality of service information is recorded, and trip identification is based on rules that require validation. Researchers have observed these limitations as opportunities, and different authors have proposed methods to estimate an alighting bus stop (Zhao et al. 2007; Trepanier et al. 2007; Munizaga and Palma 2012), to link trips and analyse transfer behaviours (Seaborn et al. 2009; Charikov and Erath 2011; Devillaine et al. 2013) and to estimate a trip’s purpose (Charikov and Erath 2011; Devillaine et al. 2013; Lee and Hickman 2014). Certain authors have also analysed travel behaviours at different levels: walking access behaviour (Utsonomiya et al. 2006), travel patterns and variability (Morency et al. 2007; Ma et al. 2013) and the location of regular activities (Charikov and Erath 2011; Amaya and Munizaga 2014). In the next section, the levels of service indicators found in the literature are briefly discussed. Then, using passive data, the construction of the different levels of service indicators for the public transport system of Santiago is explained and shown at different aggregation levels. The final section presents some concluding remarks.
