No abstract

In this paper an estimation is made of a joint household-level model of the number of vehicles owned by a household, the vehicle type choice of each vehicle, the annual mileage on each vehicle, and the individual assigned as the primary driver for each vehicle. A version of the proposed model system currently serves as the engine for a household vehicle composition and evolution simulator, which itself has been embedded in the larger Simulator of Activities, Greenhouse Emissions, Energy, Networks, and Travel (SimAGENT), an activity-based travel and emissions forecasting system for the Southern California Association of Governments planning region.

No abstract

The population of Jerusalem, Israel, can be divided into three distinct ethnic sectors: secular Jewish, ultra-Orthodox Jewish, and Arab. Not only do these population sectors tend to inhabit and work in different areas of the city, but they each have unique household structures, activity patterns, mobility tendencies, and, ultimately, travel behavior. These substantial variations in behavior, largely driven by differences in culture and lifestyle that are not captured by other personal characteristics, are essential to representing travel behavior in the Jerusalem travel model. In this paper, sector differences were traced through the activity-based travel demand model framework by using the 2010 Jerusalem Household Travel Survey. Significant variations in behavior were seen both in direct relation to the population sector and in interactions with other socioeconomic and demographic characteristics such as income and gender. This is the first known travel demand model in practice to incorporate ethnic differences so extensively in its application.

Most modern activity-based travel demand models (ABMs) in practice and research do not fully capture the central idea that travel is derived from activities. The basic unit adopted in ABMs for travel analysis is the tour, which is borrowed largely from tour-based travel demand models. To a certain extent, this approach contradicts the basic idea of ABMs in which the unit for travel analysis is the activity. In reality, individuals plan to participate in various activities in a day, and the tours and corresponding trips emerge from activity participation, potential activity location, and activity sequence choices coupled with time and space constraints imposed by activities with relatively lower spatial and temporal flexibility. The model discussed in this paper is an effort to better mimic this decision-making process. This model is a part of the latest version of the coordinated travel and regional activity modeling platform (CT-RAMP) adopted for the Jerusalem, Israel, and Phoenix, Arizona, ABMs.

This study represents a research effort to capture explicitly the intrahousehold interactions involved in the decision to participate in a joint activity. Joint activity participation is a lesser-explored step in activity-based travel demand modeling, since enlisting all possible subsets of household members in a large household results in many alternatives. For example, the number of possible subsets of members out of 10 persons is 210 = 1,024. After the exclusion of one empty subset and 10 subsets with a single member, 1,013 distinct subsets should be considered with two or more members for joint activity participation. Even more important, a joint choice model formulation is behaviorally unappealing and would require the formulation of a complicated utility function for each possible subset. Additionally, different subsets would have a highly different degree of similarity that would require a sophisticated error structure. This paper analyzes three methods to model joint activity participation that are relatively easy to estimate and implement for households of any size. In all three methods, the travel party is constructed on the basis of the individual and pairwise propensities of the household members to be engaged in a joint activity. These propensities are statistically estimated on survey data in the form of relatively simple binary choice models. The travel party emerges in the process of microsimulation as a result of the reconciliation of the decisions of different household members. This approach is an example of the use of the agent-based modeling paradigm to frame an intrahousehold decision-making mechanism in addition to econometric models.

Mobility attributes such as driver's license, car ownership, reserved parking at work, and transit pass have a very strong impact on travel choices, in particular, mode choice. Mobility attributes are not acquired for a particular trip but rather are driven by the entire set of individual travel needs (commuting being the most basic of them). Some mobility attributes, for example, car ownership and transit pass, are substitutable; others, for example, car ownership and reserved parking at work, are complementary. For this reason, mobility attributes have to be analyzed and modeled jointly. The purpose of the current research is to analyze a wide set of mobility attributes and incorporate them in an operational activity-based model as a set of midterm choices. The approach suggested in this paper is based on an iterative application of three interlinked choice submodels: (a) joint choice of person driver's license, usual driver role (priority in using one of the household cars), car type choice, reserved or reimbursed parking at work, and transit pass; (b) household car ownership choice by type; and (c) intrahousehold car allocation by type. Model estimation results confirmed strong cross- attribute effects as well as revealed many impacts of person, household, and travel accessibility variables. In particular, historical and cultural differences between three population sectors in Jerusalem—secular Jewish, Orthodox Jewish, and Arab—manifested themselves quite strongly. Application of these models for future scenarios is discussed.

This paper describes a modified approach to modeling an individual daily activity-travel pattern (DAP) coordinated at the household level. The model was primarily introduced to handle large households that are typical for the city of Jerusalem. However, the developed method proved useful in adding more behavioral aspects to the model. The study introduced daily modality and added the emphasis on modeling it for all household members simultaneously. It is of special practical value for Jerusalem since such population sectors as ultra-Orthodox Jewish and Arabs are characterized by a large share of persons who have an entire day of travel implemented in a non-motorized fashion. In addition to daily modality, this paper presents a simple but useful approach to understand at-home time-use. A binary choice model was formulated for the main reason of being inactive (working-at-home or other). The paper also discusses possible extensions of the current approach as well as alternative approaches to frame DAP type choice in the context of a complete activity-based travel model (ABM). In particular, this model can be easily extended in order to better address the joint nature of the choices using the Gibbs sampler. The paper discusses how DAP type choice can be framed in many ways and with different levels of detail subject to the ultimate model system design.