Special issue of the international journal of urban sciences on ‘activity-based modeling in urban sciences’

In the transportation research field, significant amounts of work have been and are being devoted to activity-based modelling. This modelling approach focuses on capturing the emergence of transportation demand from the desire of an individual to perform or participate in activities. As such, the demand for transportation emerges as a by-product of the location choices made by the individual when scheduling activities for the day.

The bi-directional link between transportation and urban sciences is apparent and has been extensively reported on in the literature. Testament to this being e.g. the work dealing with integrated land-use and transportation (LUTI) models. The microscopic modelling approach of activity-based models leverages new insights and opportunities at the level of the interaction between the domains of urban sciences, land-use and transportation.

The contributions in this special issue can be grouped around three main themes that emerge when (jointly) considering urban and transportation planning: monitoring, modelling and managing.

In the section on Monitoring, three papers are presented dealing with creative approaches towards data collection. Indeed, the enhanced level of detail that can be achieved with microscopic activity-based models yields a need to richer data sets for calibration and validation purposes. On the other hand, trends such as social media use, the emergence of smart transportation facilities and ubiquitous mobile communication facilities constitute potential alternative sources from which information on (transportation) behaviour can be harvested.

The first paper in this special issue, ‘Tracking the evolution of temporal patterns of usage in bicycle-sharing systems using nonnegative matrix factorisation on multiple sliding windows’, leverages the operational data that is collected in a bicycle sharing system to double as a ‘social sensor’ on space-time patterns. While this paper restricts its focus to the bike mode, a second paper, ‘Estimating activity patterns using spatio-temporal data of cell phone networks’, deals with extracting spatio-temporal patterns for all cell-phone users irrespective of their travel mode. ‘Incorporating social media in travel and activity choice models: conceptual framework and exploratory analysis’ in its turn investigates the influence of social networks on location choices by using data from an online social network.

The second theme in this special issue deals with modelling and groups three papers related to incorporating different facets from urban sciences into activity-based modelling.

Aggregated car ownership is traditionally linked to socio-demographic variables. In a microscopic activity-based context, one can go a step further and allocate the vehicles to individual trips. This requires a car allocation model that models which vehicle is used by which household member for which trip. ‘Who gets the key first? Car allocation in activity-based modelling’ investigates four alternative approaches to modelling car allocation in terms of accuracy and computational complexity.

An important component in peoples schedules is the ‘home’ activity. The home location, a result of the residential choice, of individuals has an impact on the activity-travel patterns that will be generated by the activity-based model. The paper ‘Understanding residential location choices: an application of the UrbanSim residential location model on Suwon, Korea’ investigates different preferences with regard to residential attributes (e.g. quality of neighbourhood, accessibility, …) using a land-use model for a concrete region. The output of such land-use models can serve as an input to activity-based models, effectively rendering them responsive to spatial evolutions and policies.

In ‘Planning in-home activities in the ADAPTS activity-based model: a joint model of activity type and Duration’, a different aspect of the impact of ‘home’ on activity-travel schedules is considered. Starting from the observation that the out-of-home activities (and hence properties of travel demand) can be affected by in-home activities, the in-home activity submodel of the ADAPTS model is further refined. Due to the interchangeable nature of in-home and out-of-home activities, reliably accounting for in-home activities will also prove to be relevant if models are used in policy assessment in a transportation or an urban management context.

The third and final theme in this special issue on activity-based models considers travel demand management. In this regard the paper ‘The influence of activity-travel patterns on the success of voluntary travel behaviour change (VTBC)’ investigates the impact of activity-travel patterns on the susceptibility of individuals to voluntary travel behaviour change as a response to sensibilising efforts promoting the use of an underutilised light rail service. It is found that the interdependency of trips in a trip chain impacts the propensity to use alternative transport modes, which is relevant information to account for in policy assessments.

As the guest editors of this special issue of the International Journal of Urban Sciences on Activity-Based models, we offer our gratitude to the many reviewers for their invaluable assessments and suggestions. Their joint effort contributed significantly to enhancing the quality and relevance of this special issue.