1.1  Background

The Federal Highway Administration (FHWA), in cooperation with other U.S. Department of Transportation (DOT) modal administrations having freight responsibilities, developed the first Freight Analysis Framework (FAF, or FAF¹) commodity-based freight flow origin-destination (O-D) data with coverage that includes the geospatial center of each United States county as a potential freight origin and terminus in 2002.  The FAF O-D data were derived from 1997 Commodity Flow Survey (CFS) data and other public and private or proprietary databases.  Since its inception, the FAF freight database has been used in a variety of transportation analyses including highway capacity and bottleneck assessments, truck size and weight studies, evaluations of the benefits of strategic investments in transportation infrastructure, impacts of changes in road-pricing policies, multimodal freight policy analysis, impacts of toll proposals on shipper choice decisions, and the impact on national freight movement of natural and manmade disasters (e.g., the I-40 bridge collapse in Oklahoma in 2002; the I-95 bridge at Bridgeport, Connecticut in 2004; the impact on freight movement due to Katrina in 2005).  In addition, state Departments of Transportation (DOTs) and metropolitan planning organizations (MPOs) are also using FAF network flow data to understand regional and state freight needs and initiatives.

However, the proprietary nature of much private freight data has limited the use of the FAF to FHWA.  In addition, the non-transparent nature of the data makes it impossible to quantify the proportional distribution of the public and private data components as well as the dynamic updating capability of the FAF data when other public data sources are available.  As the demand for national freight data similar to FAF increases among State DOTs and metropolitan planning organizations (MPOs), there is a need to share or exchange freight data among FHWA, State DOTs, MPOs and other transportation agencies.  Non-proprietary national freight data can be a valuable source for State DOTs and MPOs to understand the freight activity outside their jurisdictions as well as identify the freight flow that does not terminate or origin within the state boundary.  During the course of the FAF data development, including forecasting, data disaggregating, tons to truck conversion, and network calibration, a number of techniques were developed.  However, the data disclosure issues restrict the sharing of the output results of these techniques (including disaggregating and matrix expansion) to the state-to-state level.

1.2  Objectives of FAF² Project

To overcome the limitations of the original FAF, FHWA recently developed the next generation FAF² freight O-D database using the 2002 CFS and other public data sources.  Intended in part to address issues and lessons learned from FAF¹ project, some of the primary objectives of the development of the FAF² data were to:

• Provide data and analytical capability to support various Federal needs related to policy and legislative issues for the new planning horizon.
• Maximize freight data use by allowing FHWA to share data freely with State DOTs and MPOs with no proprietary or nondisclosure clauses
• Provide leadership to develop, maintain, and update data to meet the growing demand for freight data and minimize the gap among FHWA, State DOTs and MPOs
• Enable FAF² data to be updated immediately following a new release of public data sources (e.g., CFS).
• Make FAF² data more transparent to all public and private users outside the U.S. DOT.

For data quality reasons, FAF² freight flow O-D coverage is limited to 131 freight analysis zones that include 114 CFS freight O-D zones and 17 major ports, border crossings, and freight ports.  The FAF² commodity flow data are benchmarked to 2002 and are forecasted to 2035.  However, this aggregate level of FAF² data significantly hinders users who want to apply acceptable network assignments to quantify the freight demand on each highway transportation supply link, which means that many key issues cannot be addressed fully.  These key issues include highway capacity need assessments, freight bottleneck assessment, and freight diversion due to “what if” policy decisions on the network supply and improvements, among others.  Given the current geographical aggregation of FAF² data (some states will have only one O-D data point, e.g., Montana, Idaho, Wyoming, and few others), the uses of FAF² data for network assignment will be minimal without the disaggregation of O-Ds at the county or sub-county level.

1.3  Objectives of FAF² Freight Traffic Analysis

This study is directed at conducting a national highway freight analysis designed to estimate the base year 2002 and the 2035 forecasted FAF truck flow and assess the system-wide congestion related performance elements of the nation’s highway systems.  The overall objectives of the FAF² Freight Traffic Analysis are to prepare:

• Updated FAF² highway network coverage data with Highway Performance Monitoring System (HPMS) 2002 data elements essential for freight network assignment
• Truck payload equivalency factors (including empty trucks)
• FAF² equivalent truck trip O-Ds derived from FAF² tonnage O-D data
• A disaggregated FAF² truck O-D at county or sub-county levels for subsequent freight network assignment
• A database of freight truck flows (freight assignment) on the highway network for the base year and forecasted years.
• A database of highway congestion, travel time, and delay for each FAF² highway network segment
• FAF² maps depicting national freight flows and congestion for the years 2002 and 2035.

1.4  Overview of the Methodology

The overall methodology of the Freight Traffic Analysis research project covers five general areas:

1. FAF² network development and integration of the HPMS
2. Development of truck payload equivalence factor
3. FAF² O-D disaggregation at county or sub-county level for subsequent network assignment
4. Freight truck assignment and calibration
5. Development of segment-specific network performance.

These five building blocks are illustrated in Figure 1.1.

1.5  Organization of the Report

The purpose of this project is to estimate highway freight traffic flow and determine the capacity-related performance of the freight transportation highway network based on the supply and demand characteristics of freight traffic flows for 2002 and 2035.  Chapter 2 discusses methodologies and steps used to develop freight highway networks using the latest version of the National Highway Planning Network (NHPN), the HPMS database, and input from other State DOT agencies.  Chapter 3 describes the development of tonnage to truck conversion procedures. Chapter 4 discusses the methodology adopted to disaggregate the 131 FAF² freight analysis zones to 3,784 freight analysis centers including counties, ports, border crossings, and intermodal freight loading/unloading points.  This chapter also describes the subsequent disaggregation of the 131 by 131 FAF² tonnage O-D matrix to a 3,784 by 3784 O-D matrix.  Chapter 5 describes the freight assignment models and associated calibration procedures utilized for the development of base year 2002 and year 2035 network flows.  Chapter 6 discusses various performance measures estimated and illustrates the results in tabular and map formats.  Chapter 7 presents conclusions.

 

1. FAF2 Network Development Network connectivity and directionality Update FAF2 linear referencing system (LRS) schema with HPMS LRS schema Assign HPMS section data to FAF2 Network links Develop missing data lookup table and estimate attributes for non-HPMS links Network and Attribute Quality Assurance (QC) analysis

2. Development of Truck Payload Equivalency Factor Identify major vehicle group and select major body type Estimate and normalize percent distribution by vehicle group and body type Estimate mean payloads by vehicle group and body type Analyze regional variability on payloads Estimate truck equivalency factor for each vehicle group and body type

3. FAF2 Truck OD Data Disaggregation Convert freight tonnage to equivalent truck traffic Develop FAF2 truck O-D Matrix Develop O-D disaggregation techniques for domestic and international data Disaggregate FAF2 Truck O-D to county or sub-county level Associate/index each disaggregate O-D pairs with FAF2 network node IDs

4. Freight Truck Assignment and Calibration Select and apply network assignment procedures Calibrate network flow Finalize base 2002 and forecast 2035 flow assignment

5. Freight-Related Performance Measures and Maps Derivation of link-specific congestion (vehicle:capacity ratio), delay and travel time Network truck flow maps for the year 2002 and 2035 Congestion flow maps and tabulation by highway functional classes

Figure 1.1:  Building Blocks of FAF² Freight Traffic Analysis

 

 

previous | next