Civil Engineering Reference
In-Depth Information
Without direct data, we must use some method to estimate the zones
to which people go. Let us start with trips taken to work. By the most com-
monly used method, the modeler assumes that the work-travelers leaving
Zone A distribute themselves among Zones B, C, D, etc., by those zones'
attractiveness (how many work attractants they have) and their distance
from Zone A (road distance, sometimes adjusted by tolls or transit fares).
The farther the work from Zone A, the less likely that the Zone A residents
will work there. Let's switch now to shopping trips: the farther the shopping
site from Zone A, the less likely that Zone A residents will shop there.
Known as the
friction factor,
the rate at which attractiveness declines over
distance must be estimated to make the model work. (Because attractive-
ness is assumed to decline with distance, it is usually called a
gravity model.)
We know that these are crude assumptions. Doctors and business
executives in a secluded neighborhood of mansions may travel far from
their suburban homes to hospitals and corporate offices, while blue-collar
workers from an old urban district may travel far in the opposite direc-
tion to work in an industrial area, while young or poor service workers
may find jobs closer to their homes. As matters of income and profession
greatly affect destination choices, modelers have over the years tried to
incorporate ever more factors, beyond attractiveness and distance, into the
models, though it has not always been clear that multiplying the numbers
of variables increases accuracy.
In Square City, everyone is of the same economic and social group,
very much simplifying our model.
Modal Split
Now that we have estimated where the current inhabitants of a zone are
going, we ask how they get there, whether in a single-occupancy car, car
pool, or bus, or by rail, walking or bicycling—each of these ways called a
transportation
mode.
As hopes have grown that Americans will quit their
cars and instead walk or cycle, and solutions have been sought for conges-
tion and petroleum reliance, ever newer ways have evolved for trying to
estimate modal choice.
As a start, we need to find out what modes are at all available between
one's zone of origin and the destination. If there is no subway, no bus, and
no sidewalk, but everyone owns a car, then there is not much of a modal
choice. The modeler will still want to know likely car occupancy, since
that affects the volumes of vehicles that will load the highways. The easiest
way to estimate is to consult survey data on car occupancy by income, by
number of cars owned, or by type of destination.
