Urban Rhythms and Travel Behaviour

Spatial and Temporal Phenomena of Daily Travel

Stefan Schönfelder, Kay W. Axhausen

• Publisher: Routledge
• Year: 2016
• Print ISBN:
  9780754675150
• eBook ISBN:
  9781317003458
• Topic: Politics & International Relations
• Subtopic: Local & Regional Planning Public Policy
• Index: Politics & International Relations

The recent availability of longitudinal data on individual trip making and activity behaviour has provided analysts with new insights into the structures and motives of daily life travel. Multi-week travel diary data-sets and GPS observations are exciting sources of information for the description and modelling of the variability of individual travel patterns. Through an analysis of these strong new data sets, this book questions what are the most suitable methodological tools to represent the structures of long-term travel behaviour. It also examines what the data tells us about the travellers' motives and looks at how planning should translate the findings into forecasting tools and transport strategies. In doing so, the multifaceted and ambiguous character of daily life travel is revealed, illustrating how, while sound routines in time and space seem to dominate daily life, individuals show a considerable amount of variability and flexibility in travel and activity behaviour.

Chapter 1
Introduction

The patterns and rhythms of urban life have fascinated observers ever since urban life began (Mumford 1961; Pahl 1970; Lilley 2002). Industrialization profoundly changed the scale of urban life from the early 19th century onwards, inspiring works by Honoré de Balzac, Friedrich Engels, Charles Booth (1889)1 and Auguste Comte, for example. The rapid growth of cities exacerbated urban problems, making it necessary to provide and manage urban infrastructure more professionally. Transport planning and traffic engineering were two allied disciplines which emerged out of these processes relatively late, in the 1930s and 1940s, but both had been concerns of municipal engineering and urban design from the start in the 1860s (see, for example, Baumeister 1876). Their purpose has always been to provide urban actors with a reasonably priced transport system for the pursuit of spatially distributed activities (see Chapter 11 for empirical evidence).

Early success in the 1950s by social-physics researchers using large-scale, computer-based models narrowed the focus of transport planning to the peak hours of the average working day (see Weiner 2008; Hutchinson 1974; Martin, Memmott, and Bone 1961; Leibbrand 1957). While this focus was understandable against the concerns of the day, it became increasingly restrictive as time progressed. Then research in the tradition of the activity-based approach to understanding travel behaviour (see Chapter 4 for an introduction) emerged as the most important tool for widening the scope of analysis for transport planning. Not the isolated trip (see Chapter 3 for definitions), but the whole activity programme with its rhythms and interdependencies became the object of analysis (see Jones et al. 1983 and the proceedings of the relevant conference series, especially: www.iatbr.org). Better knowledge of the structures and motives of longitudinal travel behaviour has enabled transport policy and planning practice to design better measures to influence travellers according to current transport policy priorities (Long 1997; Miller 1999). Since demand management, information and counselling play more important roles in transport policy today, the traveller as an individual decision maker and his or her travel habits have been receiving more attention.

The work reported here belongs to the tradition of activity-based analysis and advances it in one crucial aspect: For the first time it employs multiday geocoded2 observations to address questions which had been raised earlier by our discipline, but never pursued for lack of data:

• What are the multiday rhythms of activity participation?

• How variable is behaviour from day to day?

• What is the size and shape of human activity space in the urban area?

• How is innovation in spatial choice mixed with well-known routines?

This is not to say that such data have never been collected or that such analyses have never been undertaken before, but either the data was much more limited in scale and scope, or the questions were different. The only early geocoded long-duration survey (the five-week Uppsala survey of 1971; see Chapter 6) was never analysed in its spatial dimension, but employed to identify the typical day or answer the reverse question of inter- and intrapersonal variability. Analyses of various two-week surveys also focused on the latter question (see, e.g., Webber 1978; Yun and O’Kelly 1997; Miller and O’Kelly 2005).

The structure of this book

This book has four main parts. Part I presents the foundations, including a discussion of theoretical perspectives on time, space and travel, an introduction to data-collection methods and terminology and a presentation of the longitudinal data sets used in our analysis. Part II focuses on the temporal aspects of day-to-day travel behaviour, which includes the concept of activity scheduling, basic results on human mobility and an analysis of the temporal rhythms of human space-time behaviour. Part III provides a framework for the analysis of spatial aspects of day-to-day mobility and presents results on the variability of human activity space. Part IV concludes with an interpretation of our findings.

PART I – Foundations

In Chapter 2 we introduce the subject of this book by sketching the interaction between social networks, activity spaces and traffic growth. These linkages influence the adoption of intelligent transport systems, particularly with respect to their management and control technologies. We then develop qualitative models of personal activity space and commercial markets. These suggest that any decoupling between economic and traffic growth will be difficult. They also suggest that any change in trends is difficult to achieve because existing travel patterns reflect the social-capital structures of society, and society is reluctant to change without good reason or external pressure.

Chapter 3 provides a consistent definition of the units of measurement for human movement. The scope of transport and activity surveys is described, and the strengths and weaknesses of the possible implementations of such surveys (stage, trip, journey and activity-based) are highlighted, as are similarities with comparable issues in time-budget surveys.

Chapter 4 provides the theoretical and empirical background for the analyses in this book, with a focus on the introduction of the activity-based-analysis (ABA) tradition within mobility research, the relevant terminology of the space-time travel relationship as well as concepts and findings of earlier work on the subject.

Different levels of reporting in travel and activity diaries (stage, trip, journey and daily activity chain) result in various non-reporting strategies by the respondents that interact across the sequence of contacts between the survey researcher and the respondent. Chapter 5 summarizes what is known about these non-response processes and suggests expected values for the most crucial of them: reports of staying at home.

The analyses in this book are based on a range of individual panel-data sets derived by different data-collection methods and survey areas, which provides a great range of behavioural patterns and regional peculiarities. Chapter 6 gives a synopsis of the data bases used to reveal the structures of daily mobility. It provides a detailed description of the different data sources and clarifies differences between the observation approaches, especially between travel-diary surveys and in-vehicle GPS tracking.

Part II – The temporal aspects of day-to-day travel behaviour

The authors of this book take the view that people plan and schedule their day. Chapter 7 suggests a framework for understanding this process and summarizes current knowledge about how travellers commit to trips and activities over the preceding days.

Chapter 8 frames the other results by providing an overview of current patterns of travel behaviour in its key dimensions (share of out-of-home travellers, number of journeys and trips, distances travelled and the durations involved in conjunction with the modes chosen) while addressing the temporal (day-of-week and time-of-day) and spatial (urban, suburban and rural) dimensions.

Chapter 9 is about variability and periodicity of personal day-to-day travel. We develop a conceptual framework to investigate the day-to-day variability in activity demand and a suitable modelling approach to capture the variables driving it.

Part III – Human spatial behaviour and the analysis of activity spaces

In Chapters 10 and 11 we describe the development of approaches to visualizing and measuring human activity space and we outline a (comparative) analysis. We then provide a synopsis of the results gained by the two main analytical approaches (enumerating trips and locations over time and continuously representing and measuring space usage).

Part IV – Conclusions

Chapter 12 concludes with a summary of the results and a methodological and policy-relevant review of the major findings. The analysis of the longitudinal travel data will show a distinct ambiguity between strong habits and the aspiration for variety seeking, especially in spatial behaviour.

1 See also booth.lse.ac.uk.

2 Geo-coded data add an exact geographical reference (coordinate) to the item observed or reported, which allows it to be mapped, to be linked with other spatially referenced data and analysed.

PART I
Foundations

Chapter 2
Theoretical Framework

Travel behaviour research works with implicit theoretical assumptions which are rarely made explicit. One which is very explicit in most work is the microeconomic assumption that more travel will be consumed ceteris paribus when the generalized cost of its consumption falls. But note that travel is interpreted and therefore measured in any number of ways, for example: number of movements, number of kilometres travelled (person kilometres or vehicle kilometres) and number of minutes of movement (person minutes or vehicle minutes). Given that these three basic dimensions are not perfectly correlated, it is unclear whether a particular change in any of the dimensions of the generalized costs of travel will have an effect of the same sign or size. The generalized costs of travel measure only the movement part of the day, but the overarching framework of the activity approach takes interactions into account and balances them with the activities undertaken during the day. The activity approach is based on Becker’s (1965) model of leisure- and work-time allocation, expanded to account for travel (de Serpa 1971; Bates 1987) without imposing the strict rationality assumptions of Becker’s neo-classical formulation on every instance of observed behaviour.

The hypothesis adopted here is that travellers trade off the generalized costs of travel with the generalized costs of the activities undertaken, which in sum make up the generalized costs of their (daily) schedules. “Generalized costs” are defined as the risk- and comfort-adjusted weighted sum of time spent on travel or activity, the associated expenditure and the social content of the movements and activities. We assume that travellers try to reduce these costs on any one day, but especially from day to day. Over longer time horizons travellers have the opportunity to change the constraints under which they must operate on a given day, some of which they choose themselves, others of which are imposed by others or are the by-product of earlier choices: home and work (or educational) locations, the availability of mobility tools (driving licences, bicycles, cars, public-transport season tickets, planes and boats), the availability of telecommunication tools (land-line phones, mobile phones, fax machines, email accounts, data lines, etc.), their network of social contacts and their contacts’ home locations (relatives, friends, work colleagues, members of clubs and churches, etc., to name the most important ones). With each choice a traveller has the opportunity to trade off an up-front investment against later lower generalized costs of travel and activity. Consider, for example, the acquisition of a car, which will lower the generalized costs of travel, especially by reducing time costs, but requires a substantial outlay at the time of purchase1.

The vast literature on transport-mode choice has shown that the time needed for the different stages of a trip and for the different elements of each stage are valued differently by travellers, as they involve different levels of comfort and risk. (See Wardman 2001, for example, for a summary of British results and Hess, Erath, and Axhausen 2008 for recent Swiss experiences.) Walking stages are valued more negatively than riding the bus or driving a car. The more likely it is that a stage along a particular route will take longer than expected, the less likely it is to be chosen, i.e., its generalized costs are deemed higher due to its unreliability. Equally, the comfort of the private vehicle is judged by most travellers to be superior to travelling by bicycle, quite apart from the higher risk of accident while cycling.

Monetary expenditures for travel and activity are obvious parts of the generalized costs of a schedule. In both cases it is appropriate to assume that travellers will only consider those expenditures which they can avoid over the time horizon of the choice at hand. The streetcar entails no additional cost to a season-ticket owner deciding on the mode for the next trip. Cost is a factor when a traveller decides whether to acquire a season ticket for the year or month ahead. Equally, the next visit to the health club only involves add-on expenditures for tips, drinks afterwards, etc., but not the membership fee any more.

The social content of the elements of the schedule are both familiar and unfamiliar to travel behaviour analysts. Social content is defined here as the sum of the social signals which a traveller sends and receives by participating in a particular trip or activity – signals which position the traveller in the social spa...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents
• List of Figures
• List of Tables
• List of Abbreviations
• Notes on Authors
• Acknowledgements
• 1 Introduction
• PART I FOUNDATIONS
• PART II THE TEMPORAL ASPECTS OF DAY-TO-DAY TRAVEL BEHAVIOUR
• PART III HUMAN SPATIAL BEHAVIOUR AND THE ANALYSIS OF ACTIVITY SPACES
• PART IV CONCLUSIONS
• Appendix
• Bibliography
• Index