Transportation 27: 1–4, 2000  2000 Kluwer Academic Publishers. Printed in the Netherlands.

Travel behavior research Understanding and modeling traveler behavior in a variety of situations has been a major area of emphasis for this Journal. Traveler behavior lies at the core of procedures for analysis and evaluation of transportation-related measures aimed at improving urban mobility, environmental quality and a wide variety of social objectives. With over twenty years of development as an identifiable area of scholarly research, and of practical application to policy and planning questions, travel behavior analysis has established itself as a central area of fundamental intellectual significance in the transportation arena and in the social sciences. The meetings of the International Association for Travel Behavior Research (IATBR), which have taken place about every three years over the past two decades, have helped shape the development of this field. These international meetings have helped promote significant theoretical and methodological ideas and approaches, and have brought about a profound transformation in the foundations and rationale underlying travel demand analysis and forecasting applications. In addition, they have provided forums for the discussion of new areas of application, and emerging policy concerns that require understanding or modification of traveler behavior. The Eighth Meeting of the IATBR took place in Austin, Texas in September 1997. It continued the tradition of the early meetings to provide a focused forum for assessing state-of-the-art developments in travel behavior and applications, and identifying the principal emerging trends, challenges and opportunities in this important area of transportation research. The Austin meeting was organized around workshop themes that reflected emerging strands and directions of travel behavior research. The papers included in the present special issue were developed from presentations first given at the Austin meeting in two of the workshops: time use, and telecommunications-travel behavior interactions. The time use workshop addressed emerging developments and multidisciplinary perspectives on the analysis and prediction of travel behavior in the context of tripmakers’ time allocation to various activities. The interest in time use reflects a natural outgrowth of travel behavior researchers to link travel and mobility to activity participation and scheduling. Activitybased approaches to travel demand forecasting have steadily gained in interest and acceptance over the past decade. However, the methodological aspects remain to be more fully developed in conjunction with large-scale

2 planning and policy applications. The papers included in this special issue reflect the state-of-the-art in both theory and methodology in this area, including a novel survey instrument to measure activity scheduling patterns and processes. The workshop on telecommunications-travel interactions addressed the travel behavior implications of telecommunications, including the adoption of telecommuting, teleshopping, and other tele-activities that can be performed from the home or from virtually anywhere instead of at fixed locations. This area of travel behavior research is particularly timely in light of the rapid and global penetration, usage and impact of the Internet, and the fundamental structural changes it appears to be generating in the manner people plan and conduct their activities. While workshop participants at the Austin conference came up with a comprehensive conceptual framework to examine interactions between travel and telecommunications, the papers in this special issue are rather more narrowly focused on telecommuting. The latter has been the subject of extensive research in the travel behavior community because of its evident linkage to transportation policy. However, it is likely that the next IATBR meeting will feature a broader scope of contributions to travel-telecommunications interactions. Papers from the above two workshop streams have not been grouped in this special issue arbitrarily. It has become clear that telecommunications and travel play complementary roles in allowing individual households and businesses to pursue their economic and social activities. Coming to grips with this multifaceted interaction will require examining tripmaking decisions alongside telecommunications use, in the context of individuals’ time allocation to various activities. This calls for an activity-based perspective on these questions. As such, the strands of research represented in this special issue are converging towards a common theoretical framework that recognizes the natural inter-relations among time use, activity participation and scheduling, telecommunications use, and travel demand. Formulating such a framework, realizing it empirically, and translating it into policy insights and operational models is likely to provide rich intellectual and methodological challenges for the travel behavior community to address in future meetings and special issues. Seven papers are contained in this special issue. Harvey and Taylor address the relationship between social contact and travel behavior, which becomes of particular importance in connection with telecommuting, where working at home implies less social interaction at the workplace, prompting the worker to seek human contact in other social spaces. Their analysis of time use data offers perhaps one of the first empirical results relating social interaction and travel.

3 Wen and Koppelman represents a two-stage discrete choice model of activity-travel patterns. The model takes on the form of a nested logit model, whose upper level is concerned with the number of stops in household members’ daily activity-travel patterns, assignment of the stops to the household members, and the allocation of household vehicles to the members. The lower level is concerned with the formation of tours. The model is estimated using the 1994 Portland, Oregon, household activity and travel survey and effects of socio-demographic factors are shown. Kitamura et al. adopt a different approach in their attempt to generate daily activity-travel patterns. The probability that a particular daily activitytravel pattern is observed is decomposed into a series of conditional probabilities, each of which is represented in their model system by a discrete choice model or a duration model. A daily pattern is then generated by microsimulation, sequentially along a time axis. Results of the validation study indicate the need for incorporating rigidities in daily activities. Doherty and Miller point out that little fundamental research has been done into the scheduling process that underlies activity-travel behavior. They offer a conceptualization of the scheduling process and proposes a computerbased experimental survey approach to collect data for the analysis of scheduling behavior. Their analyses of the data from 41 households reveal characteristics of scheduling processes, which collectively form new insights into scheduling behavior. Pratt is concerned with the problems in survey that arise from the ambiguity of the terms, “work,” “home,” “telecommuting” or “teleworking.” This leads to problems in measuring, through surveys, the effectiveness of telecommuting in achieving such social objectives as reducing commute trips. After illustrating possible errors and confusion using examples taken from various surveys, Pratt proposes a set of questions phrased to reduce ambiguity and structured into four priority levels. Brewer and Hensher present a game-theoretic approach to the adoption of telecommuting. Namely, the process of adopting telecommuting at a workplace is modeled as an interactive game between the employer and employee, each having specific payoffs associated with different levels of telecommuting. How each player react at respective stages of the interactive game is described using discrete choice models. The study points to the roles of information and incentives to be offered by employers and employees for effective telecommuting. Yen performs an aggregate analysis of the demand for telecommuting based on a choice model estimated using survey results from Taipei. The results demonstrate the importance of cost elements, in particular, those incurred by the employee. Floor space available for working at home is another impor-

4 tant factor. The study also shows that commute mode is associated with the demand elasticity, with auto commuters exhibiting larger elasticities. Hani S. Mahmassani Department of Civil Engineering and Department of Management Science and Information Systems, The University of Texas at Austin Austin, Texas 78712, USA E-mail: [email protected] Ryuichi Kitamura Department of Civil Engineering Systems Kyoto University Sakyo-ku, Kyoto 606-8501, Japan E-mail: [email protected]