A simultaneous model of household activity participation and trip chain generation

https://doi.org/10.1016/S0191-2615(99)00028-4Get rights and content

Abstract

A trip generation model has been developed using a time-use perspective, in which trips are generated in conjunction with out-of-home activities, and time spent traveling is another component of overall time use. The model jointly forecasts three sets of endogenous variables – (1) activity participation and (2) travel time (together making up total out-of-home time use), and (3) trip generation – as a function of household characteristics and accessibility indices. It is estimated with data from the Portland, Oregon 1994 Activity and Travel Survey. Results show that the basic model, which has 10 endogenous time use and trip generation variables and 13 exogenous variables, fits well, and all postulated relationships are upheld. Test show that the basic model, which divides activities into work and nonwork, can be extended to a three-way breakdown of subsistence, discretionary and obligatory activities. The model can also capture the effects of in-home work on trip chaining and activity participation. We use the model to explore the effects on time use and trip chaining of GIS-based and zone-based accessibility indices.

Section snippets

Background and objectives

From a standpoint of consumer behavior, time is the ultimate resource constraint. Financial constraints can be overcome by increasing income and wealth, but there are severe limits in how far one can go in reducing time constraints by purchasing time-saving goods and services. Engel et al. (1990) note that more money allows consumers to buy more of everything, but consumers cannot conceivably do more of everything.

In the 1970s, researchers in the field of travel demand modeling began to realize

The model concept

Our models are based on a three-level causal structure, which is depicted in Fig. 1. We propose that the demand for activities generate trips. Trips then generate travel time. Trip chaining behavior provides a feedback loop from trip demand to activity demand, as people find ways of satisfying activity demand by arranging their travel. Finally, time spent traveling cuts into time available for certain activities, thus limiting activity demand. This last set of feedback loops can be called “time

The data

The data are from the Portland, Oregon 1994 Activity and Travel Survey, conducted in the spring and autumn of 1994 and the winter of 1995. This survey involved a two-day activity diary, which was designed to record all activities involving travel and all in-home activities with a duration of at least 30 min, for all individuals in the household. Our sample consists of 3217 households with 6872 individuals (an average of 2.14 persons per household). This sample represents all of the households

Methodology

Structural equations modeling (SEM) with observed variables is defined by the system:y=By+Γx+ζ,where y is a column vector of p endogenous variables, x is a column vector of q exogenous variables, and ζ is a column vector of the error terms. The structural parameters are the elements of the three matrices:

B is the matrix (p × p) of direct effects between pairs of the p endogenous variables; Γ the matrix (p × q) of regression effects of the q exogenous variables, and Ψ=Eζζ′ the symmetric

The base model

Our base model has the first nine endogenous variables listed in Table 4 and 13 exogenous variables listed in Table 5. There are 28 direct effects among the nine endogenous variables, and each of these effects is shown as an arrow in the flow diagram of Fig. 2; these are freely estimated elements in the B matrix of system (1). In addition, there are 58 regression effects from 13 exogenous variables (Γ matrix elements), nine error-term variances (diagonal elements in the Ψ matrix), and one

Summary

These models are founded on the testable hypothesis that demand for out-of-home activities, broken down by type, causes households to generate trips in simple and complex chains, and the combination of demand for different activities determines, in part, the complexity of trip chaining. The trip chains in turn distribute travel time, broken down by travel to sites for the different types of activities and return-home travel. The activity demand and travel time variables account for all

Acknowledgements

This research was sponsored by the University of California Transportation Center. The author wishes to thank his colleague Mike McNally of the University of California, Irvine, for his intellectual stimulus and technical support in the conduct of this and related research conducted under the auspices of the Center for Activity Systems Analysis of the Institute of Transportation Studies. The author also wishes to thank Keith Lawton and the staff of Portland Metropolitan Service District

References (39)

  • Golob, T.F., Kitamura, R., Lula, C., 1994. Modeling the effects of commuting time on activity duration and non-work...
  • J. Gould et al.

    Shopping without travel or travel without shopping? An investigation of electronic home shopping

    Transport Reviews

    (1997)
  • K. Goulias et al.

    Practical method for the estimation of trip generation and trip chaining

    Transportation Research Record

    (1990)
  • Gunn, H.F., 1981. An analysis of travel time budgets into mandatory and discretionary components. Presented at the PTRC...
  • Jones, P., Koppelman, F., Orfeuil J-P., 1990. Activity analysis: state-of-the-art and future directions. In: Jones, P....
  • R. Kitamura

    An evaluation of activity-based travel analysis

    Transportation

    (1988)
  • Kitamura, R., Robinson, J.P., Golob, T.F., Bradley, M.A., Leonard J., van der Hoorn, T., 1992. A comparative analysis...
  • L.P. Kostyniuk et al.

    Life cycle and household time–space paths: empirical investigation

    Transportation Research Record

    (1982)
  • Kraan, M., 1997. In search for limits to mobility growth with a model for the allocation of time and money. In: Ettema,...
  • Cited by (167)

    • Telecommuting and trip chaining: Pre-pandemic patterns and implications for the post-pandemic world

      2022, Transportation Research Part D: Transport and Environment
      Citation Excerpt :

      For instance, travelers seem to engage in more complex work tours during peak hours, possibly due to the fact that most commute tours happen during peak hours (J G Strathman & Dueker, 1995). In addition, travelers tend to have different travel purposes and schedules on weekdays and weekends (Golob, 2000). Day of the week also influences the relationship between mode choice and trip chaining decision-making.

    View all citing articles on Scopus
    View full text