Estimated roadway segment traffic data by vehicle class for the United States: A machine learning approach
Data files
Apr 21, 2025 version files 2.08 GB
-
hpms_traffic_data_ml_v2.zip
2.08 GB
-
README.md
7.49 KB
Abstract
The Highway Performance Monitoring System, managed by the Federal Highway Administration, provides essential data on average annual daily traffic across U.S. roadways, but it has limited representation of medium- and heavy-duty vehicles on non-interstate roads. This gap limits research and policy analysis on the impacts of truck traffic, especially concerning air quality and public health. To address this, we use random forest regression to estimate medium- and heavy-duty vehicle traffic volumes in areas with sparse data. This results in a more comprehensive dataset, which enables the estimation of traffic density at the census block level as a proxy for traffic-related air pollution exposure. Our high-resolution spatial data products, rigorously validated, provide a more accurate representation of truck traffic and its environmental and health impacts. These datasets are valuable for transportation planning, public health research, and policy decisions aimed at mitigating the effects of truck traffic on vulnerable communities exposed to air pollution.
Dataset DOI: 10.5061/dryad.gmsbcc2zz
Description of the Data and Available Formats
1. HPMS Network with Road Link Average Annual Daily Traffic (AADT):
Description: This dataset includes estimates for light-, medium-, and heavy-duty vehicle traffic across U.S. roadways. The data is derived from the 2018 Highway Performance Monitoring System (HPMS), managed by the Federal Highway Administration (FHWA). The HPMS provides essential information on average annual daily traffic (AADT), but it has limited representation of medium- and heavy-duty vehicles on non-interstate roads. To address this limitation, we applied random forest regression to estimate AADT for medium-duty vehicle (MDV) and heavy-duty vehicle (HDV) traffic in regions with sparse data. Light-duty vehicle (LDV) AADT was then estimated by subtracting the sum of MDV AADT and HDV AADT from the total AADT for each roadway link.
Available formats:
.csvfile for direct data access.- ArcGIS feature class stored in an ArcGIS geodatabase for spatial analysis.
2. Census Block Traffic Density (VKT/km²):
Description: This dataset provides traffic density estimates (vehicle kilometers traveled per square kilometer, VKT/km²) at the census block level for the U.S. The data is based on the 2020 U.S. Census and the HPMS network with road link AADT for light-, medium-, and heavy-duty vehicle traffic.
Available format:
.csvfile for direct data access.
File Structure and Schema Information
1. HPMS Network with Road Link Average Annual Daily Traffic (AADT):
[hpms_aadt_estimated.csv]
This file contains the estimated traffic data for road links across the U.S., including AADT data for different vehicle classes (light-duty, medium-duty, and heavy-duty vehicles). Below are the columns present in this file:
| Column Name | Data Type | Description |
|---|---|---|
| FID_Link_Cnty_Intxn | Long | Unique identifier for the road segment. |
| STATEFP | Text | FIPS code for the state. |
| STATENAME | Text | Full name of the state. |
| COUNTYFP | Text | FIPS code for the county. |
| GEOID | Text | 2020 Census block GEOID (5 digits). |
| F_SYSTEM | Long | Functional classification of the roadway. Values include: |
| - 1: Interstate | ||
| - 2: Principal Arterial – Other Freeways and Expressways | ||
| - 3: Principal Arterial – Other | ||
| - 4: Minor Arterial | ||
| - 5: Major Collector | ||
| - 6: Urban Minor Collector | ||
| URBAN_CODE | Text | 2010 Urban Area Census Code (UACE) identifying the urban area classification. |
| URBAN | Long | Urban-rural classification field: |
| - 0: Rural | ||
| - 1: Urban | ||
| - 2: Small Urban | ||
| THROUGH_LANES | Double | Number of through lanes on the road segment. |
| LANE_KMS | Double | Length of lanes in kilometers. |
| LANE_MILES | Double | Length of lanes in miles. |
| VKT | Double | Vehicle kilometers traveled (VKT) across the road segment. |
| VMT | Double | Vehicle miles traveled (VMT) across the road segment. |
| AADT | Double | Average Annual Daily Traffic (AADT) across the road segment. |
| AADT_LDV | Double | AADT for light-duty vehicles (LDVs). |
| AADT_MDV | Double | AADT for medium-duty vehicles (MDVs). |
| AADT_HDV | Double | AADT for heavy-duty vehicles (HDVs). |
| Shape_Length | Double | Length of the road segment (in meters). |
2. Census Block Traffic Density (VKT/km²):
[census_block_traffic_density.csv]
This file contains traffic density estimates at the census block level, including total vehicle kilometers traveled (VKT), VKT for each vehicle class, and traffic density (TD) for each block. Below are the columns present in this file:
| Column Name | Data Type | Description |
|---|---|---|
| GEOID20 | Text | 2020 Census block GEOID (5 digits). |
| STATEFP | Text | FIPS code for the state. |
| COUNTYFP | Text | FIPS code for the county. |
| Area_Block | Double | Area of the census block (in square meters). |
| VKT | Double | Vehicle kilometers traveled (VKT) within the census block. |
| VKT_LDV | Double | VKT for light-duty vehicles (LDVs) within the census block. |
| VKT_MDV | Double | VKT for medium-duty vehicles (MDVs) within the census block. |
| VKT_HDV | Double | VKT for heavy-duty vehicles (HDVs) within the census block. |
| TD | Double | Traffic density (vehicle kilometers traveled per square kilometer, VKT/km²). |
| TD_LDV | Double | Traffic density for light-duty vehicles (LDVs) in (VKT/km²). |
| TD_MDV | Double | Traffic density for medium-duty vehicles (MDVs) in (VKT/km²). |
| TD_HDV | Double | Traffic density for heavy-duty vehicles (HDVs) in (VKT/km²). |