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.
been recommended for the Bay Area. However, in a wider sense, the final decision has not been made on the extent to which the additional accuracy of explicitly modeling household interactions will merit additional complexity. For that reason, such models will not be included in the Denver system, at least in the initial version. ESCORT TRIPS LINKED EXPLICITLY ACROSS HOUSEHOLD MEMBERS Another type of joint travel, known as an escort trip, occurs when two or more household members travel together to or from (or both to and from) an activity location but do not participate in the same activity there. The most common example is a parent driving a child to school and then either returning home (an escort tour) or else driving on to work (an escort stop on a work tour). Because these types of tours are partly joint and partly independent, it can be very complex to link them explic- itly across persons. For that reason, explicit modeling of escort linkages has not been done in any of the applied models or recommended for the models under design. Most of the models, however, do include a separate escort purpose, so that the most important special char- acteristics can be capturedâ particularly the fact that the mode is nearly always by automobile, with the exception of infrequent cases of walk escort. Furthermore, chil- drenâs school locations can easily be included as special alternatives in the parentsâ escort tour destination choice sets, so that at least the location is accurate, even if the exact trip timing and car occupancy are not matched. ALLOCATED ACTIVITIES DIVIDED EXPLICITLY AMONG HOUSEHOLD MEMBERS Certain types of activities, such as grocery shopping, escorting, and some other maintenance chores, are likely to be allocated across individuals in a household, show- ing a negative correlation across frequencies within a household day. The Columbus and Atlanta model sys- tems include explicit models of the generation of these activities at the household level and then allocation to particular individuals. In the Atlanta case, this model was estimated jointly with the model for household joint travel generation. Compared with explicitly linking peo- ple who make joint tours together, predicting which peo- ple within a household perform allocated activities appears less important to the model results: nothing fun- damental about the tours is being changed, only which person makes them. So, in relation to the tradeoff between accuracy and complexity, these models seem less crucial than the joint travel models, and thus they 17DESIGN FEATURES OF ACTIVITY- BASED MICROSIMULATION MODELS Brief Note on FAMOS Florida Activity Mobility Simulator Ram Pendyala, University of South Florida The Florida Activity Mobility Simulator (FAMOS)was completed in 2004 with a full- fledged devel- opment and application in southeast Florida. Since that time, work has progressed to reengineer FAMOS and integrate it with the land use model UrbanSim in con- junction with an ongoing 3-year EPA project. Table 1 (page 12) compares model features of FAMOS and var- ious models from other locations throughout the United States. REFERENCES Pendyala, R.M. Phased Implementation of a Multi- modal Activity- Based Travel Demand Modeling System in Florida. Final Report, Volume II: FAMOS Users Guide. Research Center, Florida Department of Trans - portation, Tallahassee, 2004. www.eng.usf.edu/~pen dyala/publications/report/FAMOS%20Users%20Guide .pdf. Pendyala, R. M., R. Kitamura, A. Kikuchi, T. Yamamoto, and S. Fujii. FAMOS: The Florida Activity Mobility Sim- ulator. In Transportation Research Record: Journal of the Transportation Research Board, No. 1921, Trans- portation Research Board of the National Academies, Washington, D.C., 2005, pp. 123â130.