Data Methodology

BridgeReport.org aggregates data from multiple authoritative government sources to provide comprehensive bridge infrastructure information. Our primary data sources include:

National Bridge Inventory (NBI)

The National Bridge Inventory is our primary data source and serves as the foundation for all bridge information on this website. The NBI is maintained by the Federal Highway Administration (FHWA), a division of the U.S. Department of Transportation. It contains detailed records for every highway bridge in the United States with a span of more than 20 feet that is located on a public road.

The NBI includes information on approximately 623,000 bridges and is updated annually based on inspection data submitted by all 50 states, the District of Columbia, and Puerto Rico. Each record contains over 100 data fields covering bridge identification, location, physical characteristics, condition ratings, traffic data, and maintenance information.

Source URL: https://www.fhwa.dot.gov/bridge/nbi.cfm

U.S. Census Bureau

We supplement NBI data with geographic information from the U.S. Census Bureau. This includes official place names (cities, towns, Census Designated Places), FIPS codes for geographic identification, county boundaries, and population data. Census data helps us associate bridges with recognizable places and provide demographic context.

Source URL: https://www.census.gov/data.html

FHWA Recording and Coding Guide

The FHWA Recording and Coding Guide for the Structure Inventory and Appraisal of the Nation's Bridges provides the official definitions and codes used in the NBI. We use this guide to translate numeric codes into human-readable descriptions for materials, design types, ownership, and other categorical fields.

Source URL: https://www.fhwa.dot.gov/bridge/mtguide.cfm

Historical Background

The National Bridge Inventory was established following the tragic collapse of the Silver Bridge on December 15, 1967. The Silver Bridge, which connected Point Pleasant, West Virginia, to Gallipolis, Ohio, collapsed during rush hour, resulting in 46 deaths. The disaster revealed that many bridges across the nation had not been systematically inspected or maintained.

In response, Congress passed the Federal-Aid Highway Act of 1968, which mandated the creation of the National Bridge Inspection Standards (NBIS) and required all states to inventory and inspect their bridges on public roads. The first NBI data collection began in 1971, and the program has been continuously maintained and improved since then.

What's Included in the NBI

The NBI covers highway bridges that meet the following criteria:

• Located on a public road (federal, state, county, city, or other public authority)
• Has a span of at least 20 feet (measured along the centerline of the roadway)
• Carries vehicular traffic (not exclusively pedestrian or railroad, unless also carrying highway traffic)

This includes bridges over waterways, other roads, railroads, and other features. Culverts (box structures that carry roads over waterways) are included when they have a clear span of 20 feet or more.

What's NOT Included

Several types of bridges are not included in the NBI:

• Bridges with spans less than 20 feet (estimated at 300,000+ nationwide)
• Bridges on private roads not open to public traffic
• Pedestrian-only bridges that do not carry vehicular traffic
• Railroad bridges that do not also carry highway traffic
• Bridges on federal lands managed under separate inspection programs (some National Park bridges, military installations, etc.)

Inspection Requirements

Under the National Bridge Inspection Standards, each bridge in the NBI must be inspected at regular intervals by qualified inspectors:

• Routine Inspections: At least once every 24 months for most bridges
• Underwater Inspections: At least once every 60 months for bridges with underwater structural elements
• Fracture-Critical Inspections: At least once every 24 months for bridges with fracture-critical members
• Special Inspections: More frequent for bridges in poor condition or with known deficiencies

States may request extended inspection intervals (up to 48 months) for bridges that meet certain criteria, though this requires FHWA approval.

Bridge inspectors assign condition ratings to key structural components using a standardized 0-9 scale defined by the FHWA. These ratings assess the physical condition of bridge elements based on visual inspection and, when appropriate, testing.

The 0-9 Rating Scale

Components Rated

Inspectors rate the condition of four main structural components:

• Deck (NBI Item 58): The riding surface of the bridge that directly carries traffic. This includes the wearing surface, deck membrane, and structural deck.
• Superstructure (NBI Item 59): The structural components that support the deck and transfer loads to the substructure. This includes girders, beams, trusses, arches, and floor systems.
• Substructure (NBI Item 60): The components that support the superstructure and transfer loads to the foundation. This includes abutments, piers, caps, and footings.
• Culvert (NBI Item 62): For culvert-type structures (pipes or boxes), this single rating replaces deck, superstructure, and substructure ratings.

BridgeReport.org uses the FHWA's official Good/Fair/Poor classification system, which was introduced in 2016 as part of performance-based planning requirements under MAP-21 and the FAST Act. This system provides a simplified, consistent way to communicate bridge conditions to the public and policymakers.

Classification Method

The overall condition of a bridge is determined by the lowest rating among its primary structural components:

• For conventional bridges: The lowest of deck, superstructure, or substructure ratings
• For culverts: The culvert rating

This lowest rating is then classified as follows:

Important Clarification: "Poor" Does Not Mean "Unsafe"

"Structurally Deficient" is a historical classification that was the primary metric for bridge conditions before the Good/Fair/Poor system was introduced in 2016. We include this metric for continuity with historical reporting and because some government reports and media sources still use this term.

Current Definition (Post-2018)

Under current FHWA standards, a bridge is classified as Structurally Deficient if it has a condition rating of 4 or below for any of the following:

• Deck (NBI Item 58)
• Superstructure (NBI Item 59)
• Substructure (NBI Item 60)
• Culvert (NBI Item 62)

This definition is essentially equivalent to the "Poor" classification in the Good/Fair/Poor system.

Historical Definition (Pre-2018)

Prior to 2018, the Structurally Deficient classification also considered the Structural Evaluation and Deck Geometry ratings. The 2018 change narrowed the definition to focus solely on physical condition ratings, which is why you may see historical reports showing higher numbers of Structurally Deficient bridges than current data.

Our Implementation

BridgeReport.org uses the current (post-2018) definition consistently across all data, including historical years. This provides apples-to-apples comparisons across years, though it means our historical Structurally Deficient counts may differ from reports that used the pre-2018 definition.

Our data processing pipeline transforms raw NBI data into the structured, accessible information presented on BridgeReport.org. This section describes each step in the process.

Step 1: Data Acquisition

We download the complete National Bridge Inventory dataset from the FHWA website when new data is released, typically in the spring of each year for the previous calendar year's inspection data. The raw data is provided in delimited text format with fixed-width fields.

Step 2: Data Parsing

We parse the raw NBI files according to the FHWA's Recording and Coding Guide specifications. Each field is extracted, validated against expected formats, and converted to appropriate data types (integers, decimals, strings, dates).

Step 3: Data Validation

We perform validation checks on the parsed data:

• Geographic coordinates are checked for validity (within expected bounds)
• Condition ratings are verified to be within the 0-9 scale or valid N/A codes
• Year built dates are checked for reasonableness
• Required fields are checked for presence

Records that fail validation are flagged but not removed, as anomalies may reflect genuine data issues that should be visible to users.

Step 4: Code Translation

Numeric codes are translated to human-readable descriptions using lookup tables based on the Recording and Coding Guide:

• Material codes (1, 2, 3...) become "Concrete", "Concrete Continuous", "Steel", etc.
• Design type codes become "Slab", "Stringer/Multi-beam", "Girder", etc.
• Owner codes become "State Highway Agency", "County Highway Agency", etc.
• Historical significance codes become meaningful descriptions

Step 5: Derived Calculations

We calculate derived fields not present in the raw data:

• Lowest Rating: The minimum of deck, superstructure, and substructure ratings (or culvert rating)
• Condition Category: Good/Fair/Poor classification based on lowest rating
• Structurally Deficient: Boolean flag based on condition ratings
• Unit Conversions: Metric measurements converted to imperial (feet) for U.S. audience
• Age: Calculated from year built

Step 6: Aggregation

We calculate aggregate statistics at multiple geographic levels:

• National totals and averages
• State-level summaries
• County-level summaries
• City-level summaries

Aggregations include counts (total, good, fair, poor), percentages, averages (age, length, traffic), and distributions (by material, design, owner, decade built).

Step 7: Output Generation

Processed data is output in optimized JSON format for web delivery. We generate separate files for each geographic unit (state, county, city) and for rankings/lists. This preprocessing approach allows fast page loads without database queries for most requests.

Coordinates

Bridge locations in the NBI are specified as latitude and longitude coordinates. We display these coordinates as provided in the NBI without modification. However, users should be aware that:

• Some coordinates are approximate, especially for older bridges
• Some coordinates may place bridges slightly off their actual location
• A small percentage of bridges have coordinates of 0,0 or other obviously incorrect values

FIPS Codes

We use Federal Information Processing Standards (FIPS) codes for geographic identification:

• State FIPS (2 digits): Uniquely identifies each state (e.g., 06 = California)
• County FIPS (5 digits): State FIPS + 3-digit county code (e.g., 06037 = Los Angeles County, CA)
• Place FIPS (5 digits): Identifies cities, towns, and Census Designated Places within states

Place Name Processing

The NBI uses Census FIPS place codes to identify the cities/places where bridges are located. We translate these codes to place names using Census Bureau data. Some considerations:

• Census place names include type suffixes (e.g., "Los Angeles city", "Lakewood CDP")
• We clean these names for display while preserving the official Census designation
• Some bridges are in unincorporated areas not associated with any named place
• Census Designated Places (CDPs) are statistical areas, not incorporated municipalities

BridgeReport.org provides trend analysis comparing bridge conditions across multiple years of NBI data (2020-2024). This allows users to see how conditions are changing over time.

Bridge Matching

We track individual bridges across years using the Structure Number, a unique identifier assigned to each bridge. This identifier remains constant even if other bridge attributes change, allowing us to follow the same bridge over time.

Trend Classification

Bridge trends are classified as:

• Improving: Condition category improved (e.g., Poor → Fair, or Fair → Good)
• Declining: Condition category worsened (e.g., Good → Fair, or Fair → Poor)
• Stable: Condition category remained the same across the analysis period

Aggregate Trends

For states and the nation, we calculate aggregate trends showing:

• Change in total bridge count
• Change in percentage of bridges in each condition category
• Overall trend direction based on Poor bridge percentage changes

Counts and Percentages

We calculate the following counts and percentages at each geographic level:

• Total Bridges: Count of all bridges in the NBI for that area
• Good/Fair/Poor Counts: Number of bridges in each condition category
• Good/Fair/Poor Percentages: Each count divided by total, rounded to one decimal place
• Structurally Deficient: Count and percentage of SD bridges

Averages

We calculate arithmetic means for:

• Average Year Built: Mean of year built values (excluding nulls)
• Average Length: Mean bridge length in feet
• Average Daily Traffic (ADT): Mean of ADT values (excluding zeros and nulls)

Distributions

We calculate distributions (counts by category) for:

• Material type (concrete, steel, wood, etc.)
• Design type (slab, beam, truss, arch, etc.)
• Owner (state, county, city, federal, private)
• Decade built (pre-1900, 1900s, 1910s, ... 2020s)
• Condition rating (0-9)

Rankings

We generate various rankings including:

• States by Poor bridge percentage (best and worst)
• Longest bridges by total length
• Oldest bridges by year built
• Most trafficked bridges by ADT
• Historic bridges on the National Register

We believe in transparency about data limitations. Users should understand the following constraints when using BridgeReport.org:

Data Currency

• NBI data is typically released in spring for the previous calendar year
• Individual bridge inspections may have occurred up to 24 months before the data year
• Conditions may have changed significantly since the last inspection
• Bridges may have been repaired, rehabilitated, replaced, or closed since the data was collected

Scope Limitations

• Only bridges with spans ≥20 feet are included
• Only public road bridges are included
• An estimated 300,000+ smaller bridges are not in the NBI
• Some federal lands bridges may not be included

Data Quality

• Data quality varies by state and reporting year
• Some fields have significant missing data
• Geographic coordinates may be approximate
• Some coding errors exist in the source data

Interpretation Cautions

• Condition ratings are professional judgments, not absolute measurements
• The same condition may be rated differently by different inspectors
• Poor condition does not indicate imminent danger
• Comparisons between states may reflect different inspection practices

We update our data following this schedule:

• Annual Full Update: When new NBI data is released (typically April-May for previous year's data)
• Processing Time: Updates typically appear on the site within 2-4 weeks of FHWA release
• Data Year Display: Each page shows the NBI data year to indicate currency

We do not modify or update individual bridge records between annual releases, as this could create inconsistencies with the official NBI data.

Primary Sources

• FHWA National Bridge Inventory— Official NBI data downloads
• Recording and Coding Guide— Official field definitions and coding instructions
• National Bridge Inspection Standards— Inspection requirements and procedures

Additional Resources

• ASCE Infrastructure Report Card— Overall infrastructure condition assessment
• ARTBA Bridge Report— American Road & Transportation Builders Association analysis
• U.S. Census Bureau Data— Geographic and demographic data