Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
A key phenomenon of the past 30 years is a change in the scope of decision making. When I first became involved in travel modeling, the primary decisions con- cerned investment analysis. This analysis focused on identifying what should be built, where it should built, and how it should be designed. We have seen a very dra- matic expansion in the number and the scope of issues included in the transportation decision- making process over the years. While capital investment is still a central issue, a wide range of other decisions have achieved a much higher level of importance. Such decisions include system operations and policy decisions for both transit and highway systems, pricing and the impacts on the environment, energy consumption, and urban and regional development. We need to recognize and take account of the linkages among all these components of the context as we develop and implement models. We have seen many important developments in travel modeling over the past 30 years. These developments can be divided into the broad categories of conceptual, econometric, spatial, computational, transportation ser- vice, and land use. I will briefly discuss recent develop- ments in each of these areas. A key conceptual development over the past 10 to 20 years is the organization of travel behavior as part of a daily pattern of activity- based travel pattern analysis. This analysis considers all travel by household members during a portion of the day, the entire day, or longer time periods. It takes account explicitly of intraperson and interperson consistency as well as joint choice. This imposes a variety of constraints on travel analysis, includ- ing not starting an activity until the preceding activity and necessary travel are completed, and coordination of joint travel and joint activity participation between individuals and with other travel and activity participation. Further, it ensures that travel resources, primarily cars, are assigned to no more than one tour at a time. Model design issues that need to be addressed in activity- based travel pattern analysis include generation of activities, scheduling activities, location of activities in time and space, assignment of activities to individuals within a household, and development of the travel- activity tour structure. All of these elements fit together into the daily travel- activity pattern and must be designed in a consistent and integrative way for each per- son and each household. Other issues associated with activity- travel modeling are the conflict between realism and feasibility. A model cannot be a perfect representation of the real world; part of the art of modeling is deciding which components can be ignored without seriously undermining the usefulness of the model to represent behavior and inform decision makers. An effective structure balances the need to repre- sent activity and travel components; balances activity loca- tion, scheduling, and tour structure to satisfy timeâspace constraints; and relates travel pattern and mode choice. Figure 3 highlights the tour modeling dimensions. An important observation about this diagram is that the com- ponent models are generally estimated distinctly. Each component is estimated separately in both application and most research and is linked analytically. An important step in the advancement of activity- based travel models is to integrate the information and estimation process; this is a very difficult, but critical, task to ensure the consistency of relationships between the model elements. Integrating the activity generation and scheduling process is important for entire- day schedule consistency. This process needs to recognize the dynamics of individ- ual behavior. We tend to reflect behavior based on what individuals have done in a specific period of time. But we do not know how much of those activities were planned and how much are the result of changes in activities or conditions during early portions of the day. The major economic developments in this field are adoption of disaggregate analysis and discrete choice mod- eling. The historical argument over disaggregate versus 3OVERVIEW OF THE POLICY ISSUES The Model Modeler Decision Maker Factors Other FIGURE 2 Modelersâ view of decision making. FIGURE 3 Tour modeling dimensions. (TOD = time of day.) Primary and secondary activity and tour TOD Entire-tour mode Stop frequency Stop location Trip mode Trip departure time
