No abstract

Over the last two decades, passively collected data sources, like Global Positioning System (GPS) traces from data loggers and smartphones, have emerged as a very promising source for understanding travel behaviour. Most choice model applications in this context have made use of data collected specifically for choice modelling, which often has high costs associated with it. On the other hand, many other data sources exist in which respondents' movements are tracked. These data sources have thus far been underexploited for choice modelling. Indeed, although some information on the chosen mode and basic socio-demographic data is collected in such surveys, they (as well as in fact also some purpose collected surveys) lack information on mode availability and consideration. This paper addresses the data challenges by estimating a mode choice model with probabilistic availability and consideration, using a secondary dataset consisting of 'annotated' GPS traces. Stated mode availability by part of the sample enabled the specification of an availability component, while the panel nature of the data and explicit incorporation of spatial and environmental factors enabled estimation of latent trip specific consideration sets. The research thus addresses an important behavioural issue (explicit modelling of availability and choice set) in addition to enriching the data for choice modelling purposes. The model produces reasonable results, including meaningful value of travel time (VTT) measures. Our findings further suggest that a better understanding of modes choices can be obtained by looking jointly at availability, consideration and choice.

Recent work in transport research has increasingly tried to broaden out beyond traditional areas such as mode choice or car ownership and has tried to position travel decisions within the broader life context. However, while important progress has been made in terms of how to capture these additional dimensions, both in terms of detailed tracking of movements and in-depth data collection of long term decisions or social network influences, surveys have tended to look at only a handful (or often one) of these issues in isolation, especially at the data collection end. Making these links is the key aim of the data collection described in this paper. We conducted a comprehensive survey capturing respondents' travel, energy and residential choices, their social environment, life history and short-term travel patterns. The survey is composed of a detailed background questionnaire, a life-course calendar and a name generator and name interpreter. Participants were also required to use a smartphone tracking app for two-weeks. We believe that this is an unprecedented effort that joins complexity of the survey design, amount of information collected and sample size. The present paper gives a detailed overview of the different survey components and provides initial insights into the resulting data. We share lessons that we have learned and explain how our decisions in terms of specification were shaped by experiences from other data collections.

An understanding of activity choices and duration is a key requirement for better policy making, in transport and beyond. Previous studies have failed to make the important link with individuals' social context. In this paper, the Multiple DiscreteContinuous Nested Extreme Value (MDCNEV) model is applied to the choice of activity type and duration over the course of two days, using data from the Chilean city of Concepción. In common with other studies, heterogeneity across decision makers is accommodated in the model by analysing the impact of different sociodemographic, mobility and residential location variables on both the activity choice and the time allocation decision. In addition, different social network and social capital measures are found to be significantly correlated with the choice and duration of different activities, and we show how these relationships seem to differ from the effects of socio-demographic variables. Finally, we perform a forecasting exercise using the MDCNEV model, highlighting the differences in substitution patterns from a standard MDCEV model.

The increased interest in time use among transport researchers has led to a search for flexible but tractable models of time use, such as Bhat's Multiple Discrete Continuous Extreme Value (MDCEV) model. MDCEV formulations typically model aggregate time allocation into different activity types during a given period, such as the amount of time spent working and shopping in a day. While these applications provide valuable insights into activity participation, they ignore disaggregate activity-episodes, that is the fact that people might split their total time spent working in multiple separate blocks, with breaks or other activities in between. Insights into this splitting into episodes are necessary for predicting trips and understanding time use satiation. We propose a modified MDCEV model where an activity-episode , rather than an activity type , is the basic choice alternative, using a modified utility function to capture the reduced likelihood of individuals performing a very large number of episodes of the same activity. Results from two large revealed preference datasets exhibit equivalent forecast accuracy between the traditional and proposed approach at an aggregate level, but the latter also provides insights on the number and duration of activity-episodes with significant accuracy.