Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 30
D E S I G N F E AT U R E S O F A C T I V I T Y- B A S E D M I C R O S I M U L AT I O N M O D E L S 19
that allow some sensitivity of time-of-day choice to net- model systems include such a model as the "lowest" one
work conditions. All the models have used at least four in the system. It is also possible to include such a model
network assignment periods: a.m. peak, midday, p.m. for car trips only so as to predict the shape of the demand
peak, and off peak. In some cases, free-flow conditions profile within the broader peak periods.
are assumed for the off-peak period, so no traffic assign-
ment is needed for it. In some models, a fifth period has
been added by splitting the off-peak period into early ACCESSIBILITY MEASURES IN
morning and eveningnight. The more recent models, UPPER-LEVEL MODELS
beginning with Columbus, use more precise time win-
dows so as to schedule each tour and trip consistently Last, but certainly not least, is the issue of how to include
during the day. This scheduling involves keeping track of most accurately the accessibility and land use effects in
the available time windows remaining after blocking out the upper-level models. Calculation of full log sums
the time taken by each activity and associated travel. The across all possible nests of lower-level alternatives is
time windows can also be used in the activity generation clearly infeasible with so many levels of choices. The ear-
models. The Sacramento model and perhaps other mod- liest Portland models came the closest to including
els are moving to half-hour periods to provide even more "proper" individual-specific logsums, but the structure
detail. The main constraint on how small the time peri- of that model was relatively simple and the effect on
ods can be is the adequacy of the self-reported times in model run time severe. The San Francisco models include
the diary survey data. There is evidence that people often mode-specific measures with set boundaries, such as the
round clock times to 10-, 15-, or 30-min intervals. number of jobs accessible within 30 min by transit. The
rather arbitrary cutoff boundaries in such measures can
result in unexpected sensitivities when the models are
TOUR TIME-OF-DAY RELATIVE TO MODE AND applied. The New York and Columbus models use
DESTINATION CHOICE MODELS mode-specific travel-time decay functions that approxi-
mate the log sum from a simple destination choice
It is not obvious whether activity and departure times model. Such measures perform better but still have the
should be predicted before both mode and destination problem that they are mode specific and that automobile
choices, between them, or after both. There is some and transit accessibility tend to be correlated, so it is dif-
empirical evidence that shifts in time of day occur at two ficult to estimate model parameters for both of them. A
levels: the choice between broad periods of the day (e.g., method that solves this problem and is more consistent
morning, afternoon, etc.) is made fairly independently of with discrete choice theory is to approximate joint
accessibility, while smaller shifts of up to an hour or two modedestination choice logsums. However, the mode
are more sensitive to travel times and costs--the peak- choice log sums tend to vary widely across the popula-
spreading effect. Because all the models use broad net- tion, so it is best to calculate different accessibility mea-
work time periods, the tendency has been to model the sures for different population segments. The Sacramento
choice of these periods for tours at a fairly high level models use such an approach, with aggregate accessibil-
above mode and destination choices (although in most ity logsums for each combination of seven travel pur-
cases the usual destination for work and school tours has poses, four car availability segments, and three
already been predicted). In some models, time-of-day walk-to-transit access segments--as those tend to be the
choice is predicted between the destination and mode most important segmentation variables in the mode
choice levels, which allows the use of destination-specific choice models.
mode choice log sums in the time-of-day model but
requires that the destination choice model assumes (or
stochastically selects) a specific time of day for the REFERENCES
impedance variables.
Bowman, J., and M. Ben-Akiva. Activity-Based Disaggregate
Travel Demand Model System with Activity Schedules.
DEPARTURE TIME CHOICE MODELED Transportation Research A, Vol. 35, 2001, pp. 128.
SEPARATELY AT TRIP LEVEL Bradley, M., J. Bowman, and K. Lawton. A Comparison of
Sample Enumeration and Stochastic Microsimulation for
Perhaps the placement of the model that predicts the Application of Tour-Based and Activity-Based Travel
choice of times for the overall tour is not as crucial if Demand Models. Presented at European Transport Confer-
there is a separate model that predicts the departure time ence, Cambridge, United Kingdom, 1999.
for each trip to the more detailed periods, conditional on Bradley, M. A., J. L. Bowman, Y. Shiftan, K. Lawton, and M.
the mode and origindestination of each trip. Some E. Ben-Akiva. A System of Activity-Based Models for Port-
