Introduction to Travel Demand Forecasting

Comprehensive vocabulary flashcards covering the concepts, stages, mathematical models, and modern alternatives of travel demand forecasting.

Travel Demand Forecasting

The process used to predict travel behavior and resulting demand for a specific future time frame, based on assumptions dealing with land use, the number and character of trip makers, and the nature of the transportation system.

Traffic Analysis Zones (TAZ)

Designated geographical zones used in trip generation to estimate the number of person-trips that will begin or end within a given area.

Trip Ends

The classifications predicted for each zone by trip generation models, consisting of either origins (productions) or destinations (attractions).

Four-Step Forecasting Model

The traditional sequential transportation planning process comprising Trip Generation, Trip Distribution, Modal Split (Mode Choice), and Traffic Assignment.

Trip Generation

The first stage of the four-step model, which estimates how many person-trips will produce or attract and models the decision to travel for a specific purpose.

Trip Distribution

The second stage of the four-step model, which allocates trips generated in one zone to other zones across the study area by modeling the traveler's choice of destination.

Modal Split (Mode Choice)

The third stage of the four-step model, which determines which mode of transportation (such as automobile or transit) will be used to make a trip.

Traffic Assignment (Route Assignment)

The final stage of the four-step model, which determines the actual highway and transit routes used and predicts the volume of vehicles on each network segment.

Home-Based Trips

Trips that either start or end at the traveler's home.

Non-Home Based Trips

Trips that start and end at locations other than home, such as a trip from work to a shopping mall.

Time-Based Trips

Trips categorized according to the specific time of day they occur, such as during peak or off-peak hours.

Person Type-Based Trips

Trips categorized according to traveler characteristics, such as age, occupation, or income level.

Cross-Classification

A trip generation technique developed by the Federal Highway Administration (FHWA) to determine the number of trips that begin or end at the home through linked sub-models.

Income-Sub Model

A cross-classification sub-model that relates households to their various income categories.

Auto-Ownership Sub-Model

A cross-classification sub-model that establishes the relationship between household income and auto-ownership levels.

Trip Production Sub-Model

A cross-classification sub-model that establishes the direct relationship between trips made by each household and independent demographic variables.

Trip Purpose Sub-Model

A cross-classification sub-model developed from origin-destination travel surveys that relates trip purposes to income so that productions can be divided among different purposes.

Trip Productions

Trips generated at the household end of a journey.

Trip Attractions

Trips that terminate at non-home activity zones, such as work, school, or retail shops, based on purpose.

Trip-Rate Analysis

A trip generation method based on average trip generation rates per unit (e.g., household or employee type) that is specifically suited to estimating trip attractions.

Multiple Regression Analysis (Trip Generation)

A common trip generation approach that minimizes the squares of differences between observed and expected values when a dependent variable relates to two or more independent variables.

Trip Table

A two-dimensional origin-destination matrix produced during trip distribution that displays the number of trips traveling from each origin zone to each destination zone.

Gravity Model

The most widely documented trip distribution model; states that trips between two zones are directly proportional to destination attractions and inversely proportional to a function of travel time.

Singly-Constrained Model

A gravity model formulation where the sum of calculated productions in each zone is constrained to equal the number of productions specified in the problem statement.

Doubly-Constrained Model

A gravity model formulation where both the sum of productions and the sum of attractions in each zone equal the respective productions and attractions specified in the problem statement.

Growth Factor Model

A trip distribution method used when only origin-destination data is available; can only forecast traffic between zones where traffic already exists and cannot account for changes in travel time.

Fratar Method

The most popular growth factor model; a mathematical formulation proportioning future trip generation estimates between zones as a function of current trips and the attracting zones' growth factors.

Logit Model (Modal Split)

A discrete choice model that evaluates the relative utility of each travel mode as a summation of modal attributes, expressing mode choice as a probability distribution.

Modal Utility Function

The mathematical equation $U_x = \sum_{i=1}^n a_i X_i$ used in logit modeling, where $U_x$ is mode utility, $X_i$ is an attribute value, and $a_i$ is a negative coefficient reflecting disutility.

Dijkstra's Algorithm

A mathematical procedure used in the traffic assignment stage proven to identify the shortest network path having the absolute minimum travel time, distance, or generalized cost.

Activity-Based Models (ABM)

An advanced travel forecasting alternative that models individual and household decisions throughout the day, focusing on the underlying activities that generate travel and complex tour patterns.

Integrated Land Use and Transport Models

Advanced modeling frameworks that capture the dynamic feedback between transportation and urban development, representing long-term investment impacts and location choices of households and businesses.