Regional Travel Demand Model Overview
To understand how people will make trips, modelers look at the reasons why people travel. The model takes into consideration the real choices made by residents in our region. This information is collected from rigorous travel behavior surveys that provide data regarding people’s travel choices. RTC’s last survey, the 2009 Household Travel Survey, tracked 1,650 households to understand how factors such as age, income, children, car ownership, and transportation infrastructure characteristics affect the travel choices of Clark County residents.
Data input into the transportation model includes population and employment, both for existing conditions and the forecast year. Future population and employment assumptions are developed to be consistent with local comprehensive plans. Transportation networks, including roadways and transit routes, for existing conditions and the forecast year also serve as major data inputs.
In the model, our region is divided into over 1,000 discrete geographic areas called transportation analysis zones. Census data, land characteristics, economic factors, land capacity and land availability measurements are used to develop land use forecasts that project the number of future households and jobs located in each zone.
RTC uses a standard four-step modeling process for travel demand forecasting. This four-step process consists of the following parts:
“Do I want or need to take a trip?”
The first step in the modeling process forecasts the number and types of trips derived by activities from each transportation analysis zone. The projection is based on the number and demographic characteristics of households in each zone, as well as the amount and type of employment.
Households are classified into 64 household types stratified by household size, household income and the age of the head of household. Employment is categorized into nine types, ranging from service sector and retail, to finance and agriculture. Using behaviors identified in the Household Travel Behavior Study, the model forecasts the likelihood of households to make certain types of trips based on household type and employment mixes in each zone. Trip types are classified as work, shopping, recreation, college, school, and other.
“Where do I want to go?”
Next, the model predicts where the trips produced in the first step are destined. Each zone’s availability of attractions— work, shopping, recreation and other activities—and the accessibility (access to auto networks and transit) from the zones where trips are produced determines where trips are likely to go.
“How will I get there?”
As in the real world, travelers in the model have many transportation mode choices, including walking, biking, driving alone or with others, and walking or driving to transit. For the model to forecast travel demand with a reasonable degree of confidence, it must account for why people make mode choice decisions.
The model considers the following factors when determining mode choice:
- Cost - What are the expenses of operating and maintaining a car? Are there parking expenses? How much does transit cost? Are there tolls?
- Travel time - Is it faster to drive, take transit, walk or bike?
- Auto availability - Do I have access to a car?
- Transit access - Can I get to transit easily?
- Urban design - Am I in a high-density, mixed-use area where I’m more likely to walk or bike?
- Socio-economic relationships - What is my household income? Are there as many cars as employed people in my household?
“What route should I take?”
The model uses data from the previous three steps to simulate the way people will travel. For auto trips, the model assigns traffic to streets in specified time periods. The model assumes the availability of multiple routes between origins and destinations, accounting for congestion. To forecast the transit trips route, the model considers the time segments of the journey - including the time walking to a transit stop; the wait time at transit stops/transfers; and the time spent traveling in a transit vehicle.
The Regional Travel Demand Model produces a number of measures that are useful when evaluating transportation policy, plans, and projects, including: directional link volumes, vehicle miles traveled, vehicle hours traveled, vehicle hours of delay, volume/capacity ratios, speeds, travel time, transit boardings/ridership, travel mode split, Origin/Destination statistics and others.
Before the model is used to forecast future conditions of the transportation system it is first validated against actual traffic count and transit boarding data from the present in order to ensure that the model is performing well.