Office of Operations
21st Century Operations Using 21st Century Technologies
Table of Contents

Traffic Analysis Toolbox Volume XIV: Guidebook on the Utilization of Dynamic Traffic Assignment in Modeling

1.0 Introduction

This guidebook on the utilization of Dynamic Traffic Assignment (DTA) complements and enhances other existing guidebooks on traffic modeling by providing guidance on DTA. Since DTA modeling is a new and emerging technique, basic DTA modeling methods and techniques are discussed in this guidebook. This guidebook explains how to approach the development and application of a transportation model with DTA for alternatives analysis.

Dynamic Traffic Assignment is an evolving technique in transportation modeling. As of late 2012 when this Guide was developed, both DTA modeling process and software capabilities are rapidly changing. Thus, this guidebook does not expand into great detail on DTA methods of application due to the evolving nature of DTA. Over time FHWA will expand upon and provide more detailed information on DTA methods and approach.

1.1 DTA Modeling Considerations

DTA is a modeling method that can be applied to models of different sizes and resolutions and to different contexts for analysis time frames. These three considerations are shown along different axes in Figure 1.1.

Figure 1.1 DTA Modeling Considerations

Figure 1.1 depicts three different considerations for a Dynamic Traffic Assignment model. The considerations are Time Frames, Model type, and Model Size. The time frames range from real time, near term, interim, and long-range plan. The model types range from micro, meso, and macro. The model size range from facility, corridor, sub region, and region.

Source: Cambridge Systematics, Inc.

Model Size

The size of model networks can vary greatly when DTA is applied. Since route choice is a major element of DTA, the size of the model network at a minimum should include alternative routes to allow route choice to occur.

Improving software and computing capabilities are making it possible to apply DTA at different scales. Figure 1.2 is an illustration of how different scales of DTA could be applied.

Figure 1.2 Model Scales

Figure 1.2 depicts three different model scales in a sample urban area and is intended to illustrate that Dynamic Traffic Assignment models can be developed at different scales. The left most map shows a large regional area. The middle map shows a sub-regional area that is within the regional area. The right most map shows more detail of the corridor level scale.

Source: Cambridge Systematics, Inc.

Model Timeframes and Analysis Contexts

DTA can be applied for various time periods and time intervals. DTA also can be applied to near-term and future long-range plans, to fine tune travel demand estimates, and to conduct operational analysis on design improvements.

Model Resolution

DTA may be incorporated into macroscopic, mesoscopic, and microscopic models.

  • Macroscopic models. For the purpose of this document, macroscopic models refer to regional travel demand models for both traditional trip-based models and Activity-Based Models (ABM).
  • Mesoscopic models. These models use aggregated flow relationships and include more precise representation of traffic operations than travel demand models. DTA applications are strongly associated with mesoscopic type modeling software.
  • Microscopic models. These models simulate individual vehicle-to-vehicle interactions and traffic control strategies. DTA applications in microscopic models provide the most complex analysis of all the model types.

Traffic modeling often involves a combination of some or all of these model types. This concept is referred to as Multi-Resolution Modeling (MRM).

1.2 When to Use DTA?

DTA modeling methods provide the practitioner with extensive capabilities in transportation modeling. The capabilities of changing start times and using alternative route choices based on congestion and other information within the model allow for the testing of a multitude of transportation conditions. Applying DTA methods may require more effort than other static transportation modeling techniques, however; therefore, the need for DTA methods should be considered carefully, taking into account data needs, model building time, and calibration.

Project evaluations for which DTA would be an appropriate tool include the following:

  • Bottleneck removal studies;
  • Active Transportation and Demand Management (ATDM) strategies;
  • Integrated Corridor Management (ICM) strategies;
  • Intelligent Transportation Systems (ITS) strategies;
  • Operational strategies;
  • Demand management strategies;
  • Additional capacity;
  • Incident management response scenarios;
  • Special events;
  • Work zone impacts and construction diversion; and
  • Managed lanes and tolling projects.

1.3 Basic DTA Requirements

The requirements for preparing and applying DTA consist of traffic modeling software capable of handling DTA, adequate and sufficient data for the development and calibration of a model, and the knowledge and skills to apply the tools. The following is a brief overview of the key requirements for applying DTA.

  • Regional Travel Demand Model. The development of a DTA model requires Origin-Destination (O-D) inputs. Regional travel demand models are based on trips of different purposes with both an origin and destination. Building a DTA model without this type of information is very difficult. Ideally, the travel demand model for a region should be stratified into peak periods or individual hours. Daily regional models (24-hour assignments) are too coarse for a DTA modeling approach that examines congestion and traffic assignments at a time increment of less than 1 hour.
  • Robust Data Collection. In order to build and calibrate a DTA model, sufficient amounts of data collected in the field should be available so that the variation of traffic and congestion is understood and quantified. The types of temporal-based data to be collected include traffic counts speeds, congestion, and geographic data such as base mapping and lane geometry. Chapter 5 in this guidebook provides a comprehensive discussion of data needs.
  • Transportation Modeling Software with DTA Capability. The analysis software used must have DTA capabilities. Chapter 3 of this guidebook discusses the DTA capabilities required in traffic modeling software packages.
  • Transportation Modeling Skills. The application of DTA is another layer on top of existing transportation modeling techniques. Having fundamental knowledge of travel demand modeling and micro/mesoscopic simulation modeling techniques and working knowledge of model calibration and statistical analysis is needed to apply DTA successfully.

1.4 Purpose of This Guide

This guide demonstrates how to successfully apply DTA methods. These methods include recommended approaches to model building, calibration, and alternatives analysis. The intended audiences for this guidebook include the following:

  • Practitioners: Applying DTA requires skills and knowledge in many types of transportation modeling. An understanding of travel demand models and meso- and microsimulation is required in addition to an understanding of DTA. This guidebook assumes that practitioners have some background in these areas.
  • Program Managers: The intent of this guidebook is to provide direction on how to apply DTA. Research on the level of effort to apply DTA was not part of this effort; however, a program manager reading this document may gain a better appreciation for the effort that goes into DTA modeling, the types and amount of data that should be collected, and the types of modeling software tools that should be used.
  • Software Developers: The capabilities of the traffic modeling software are essential. The intent of this guidebook is to help bridge knowledge and methodologies used by practitioners and software developers. Chapter 3 of this guide discusses features that practitioners should look for in a software application.

Relationship to DTA Primer

A primer and FHWA DTA course that explain the fundamentals and concepts of DTA have previously been developed. The primer, which is a useful precursor to this guidebook, is available. A brief review of the fundamentals of DTA is presented in Chapter 2 of this guide.

1.5 Organization of this Guide

This guide is organized to provide a modeling framework for applying DTA. Chapters 2 and 3 provide background material on DTA modeling and traffic software. Chapters 4 through 9 provide how-to guidance on applying DTA.

  • 2.0 DTA Fundamentals. The major concepts of DTA are discussed in Chapter 2, including the differences between using Dynamic User Equilibrium (DUE) and “one-shot” non-DUE assignment procedures.
  • 3.0 Modeling Capabilities. Chapter 3 explains what software capabilities are available and what should be required in a software package if a DTA approach is used.
  • 4.0 DTA Modeling Framework. Chapter 4 discusses the overall approach to applying DTA. It covers the scoping of a DTA modeling project and provides an overview of each of the steps in the modeling process.
  • 5.0 Data Requirements. Chapter 5 discusses the data required to develop, calibrate, validate, and apply a DTA model.
  • 6.0 Base Model Development. Building a base model is similar whether it uses a static or dynamic approach. The major focus of Chapter 6 is on methods for developing O-D inputs.
  • 7.0 Error and Model Validity Checking. Before a base model can be calibrated or a future model can be used in an analysis, it is important to ensure that the model is free from errors and coding mistakes. In addition to basic error checking, it is important that the DTA model be “stress tested” to ensure the underlying behaviors and responsiveness of the model is adequate for the purposes intended. Chapter 7 provides a number of techniques for conducting these “validity” checks.
  • 8.0 Calibration and Validation. Calibration is one of the most important model activities. Chapter 8 outlines a systematic process that considers the objectives of the project, statistical measures, and a strategy for adjusting the model to achieve calibration. Chapter 8 also suggests a validation step; whereby, the calibrated model is tested using a different data set.
  • 9.0 Alternatives Analysis. The purpose of modeling is to test alternatives. With DTA there is a range of treatments and approaches that can be tested. Chapter 9 discusses where DUE and non-DUE testing could occur.
  • Appendix A provides summaries of several DTA implementations, including analysis objectives, study location, publication name and links to these publications.

You may need the Adobe® Reader® to view the PDFs on this page.