Trip Generation
Introduction
Trip generation is the first step in the traditional, sequential four step process of transportation
modeling. It establishes a relationship between land use, socioeconomic and demographic data and trip
productions and attractions.
Background
Model demand is aggregated into zones. Trips originate and terminate at zone centroids. A trip
has 2 ends, a production and an attraction. A trip also has an origin and a destination, and the origin
of a trip is not always the production end. This will depend on trip purpose. For example, because
we generate work trips in two parts (the production end and the attraction end), we arbitrarily assign
the production end to the home zone. Well, as you know, after a day at work, you leave work (the
attraction end) and depart for your destination (which is the production end of the trip, by definition).
We must learn about trip purposes, and methods of trip generation which are best suited for each purpose.
Three Trip Purposes:
Three trip purposes are used (at a minimum). Many other types are possible. Some large are models use 8 or more.
Home Based Work (HBW), only trips from home to work or visa versa. Most urban area peak hour trips are this type.
Home Based Other (HBO), any other kind of trip with one end at the home.
Non Home Based (NHB), any trip that doesn't either come from or go to a home.
Cross Classification Method:
(Also called category analysis). Used mainly for trip productions. Used
for Regional Studies not Traffic Impact Analysis
Advantages:
This technique allows the transportation planner to get a good understanding of the importance of the variables of trip
generation.
There is no need for an assumption of linearity between independent and dependent variables.
Disadvantages:
Requires very detailed data to construct and predict trip generation.
The independent variables selected for the study may not be independent.
The following is an example of a cross classification model. Note, this example uses income/auto ownership.
You can also use income/household size or auto ownership/household size (or others) depending on availabilty of data.
Determine Income Distribution - This part predicts the number of households within various income divisions (based on average zone income level). (%low,% medium and %high income)
Determine Auto Ownership - This part gives percent of households that fall into a particular auto ownership category (based on average zone income level). (%0, %1 and % 2+ autos/hh)
Determine Trip production - This gives the number of trips by income/auto ownership category
Trip Purpose submodel -This step gives the percent of trips in each trip purpose (e.g., HBW, HBO and NHB), for
each income class category.
See Example
Rates Based on Activity Units Method:
Used for Traffic Impact Analysis or very detailed regional models.
The graph below shows the number of trip ends vs. the number of dwelling units. If the number of dwelling
units is known, the number of trips can be determined (or vice versa) from the graph.
see ITE trip gen book(Institute of Transportations Engineers)
The following image depicts the screen of a computer program (Trip Generation Software) that will
determine the number of trips produced depending on the type and "size" of establishment. see example TripGen screen
Regression Method:
Regression is mostly used for attractions
Graphical Solution to Trip Generation given values for variables.
see graph
A linear or nonlinear relationship can be assumed.
The dependent variable is the number of trips. (y-axis)
The independentvariables are assumed to be independent of each other although that is
not always the case. If they are not independent of each other they are " colinear ".
The independent variables can be population, autos, number of dwelling units, etc.
see NCHRP regression equations