Traffic Data for Pavement Design and Evaluation

Traffic as Pavement Loading Input

Traffic data is the most critical variable in pavement structural design alongside subgrade support. The entire pavement structure — surface layer, base course, subbase, and subgrade — is proportioned to withstand the cumulative damage inflicted by repeated vehicle loads over the design life. Without accurate traffic loading characterization, a pavement is either under-designed (premature failure) or over-designed (excessive initial cost).

The AASHTO 1993 Design Guide (Part III, Chapter 5) defines traffic loading as the cumulative number of Equivalent Single Axle Load (ESAL) applications over the design period. The design equation uses the term W18 — the cumulative number of 18,000 lb (80 kN) single-axle load applications expected in the design lane over the pavement’s design life. This is computed as:

w18 = DD × DL × W18

Where DD is the directional distribution factor (typically 0.50 for two-way roads), DL is the lane distribution factor (varies from 1.00 for single-lane roads to 0.50–0.75 for roads with four or more lanes per direction), and W18 is the cumulative two-directional ESAL count.

The lane distribution factor accounts for the fact that not all traffic uses the design lane. AASHTO 1993 Appendix D provides standard lane distribution factors: for 1 lane in each direction: 100% of truck traffic uses that lane; for 2 lanes: 80–100%; for 3 lanes: 60–80%; for 4 or more lanes: 50–75%. These factors reflect that heavy trucks tend to concentrate in the rightmost (slow) lane on multi-lane highways.

Traffic data also governs pavement evaluation. During condition surveys and structural evaluations, comparison of actual accumulated traffic against design traffic provides the first diagnostic clue. If a pavement exhibits severe distress after only 40% of its design ESALs have accumulated, the cause likely lies in excessive loading (overload trucks), inadequate structural capacity, or material/construction deficiencies — not simply age.

The Federal Highway Administration’s Long-Term Pavement Performance (LTPP) program maintains the most comprehensive database linking traffic loading to pavement performance. LTPP data shows that traffic loading explains 40–60% of the variability in flexible pavement deterioration rates across the US and Canada.

Vehicle Classification — FHWA Classes 1 through 13

Vehicle classification is the foundation of traffic data collection because different vehicle types impose vastly different magnitudes of pavement damage. The FHWA 13-Category Classification System defined in the Traffic Monitoring Guide (2013 edition) categorizes motor vehicles based on the number of axles and axle arrangement.

The 13 FHWA classes are defined as follows:

Class 1 — Motorcycles: Two or three-wheeled motorized vehicles with two axles. Contribute negligible pavement structural damage. Typical axle spacing: 1.00–5.99 ft.

Class 2 — Passenger Cars: Sedans, coupes, station wagons, and minivans. Two, three, or four axles (including trailers). Contribute less than 0.001 ESAL per pass. Typical axle spacing: 6.00–10.10 ft.

Class 3 — Other Two-Axle Four-Tire Single-Unit Vehicles: Pickup trucks, sport utility vehicles, vans, campers, motor homes, ambulances, hearses, and minibuses with a single rear axle fitted with single (not dual) tires. Despite having the same axle count as Class 2, these vehicles frequently tow trailers, producing 3- or 4-axle configurations. Axle spacing: 10.11–23.09 ft.

Class 4 — Buses: Two- or three-axle traditional transit and school buses. Minimum gross weight threshold: 12,000 lb. Axle spacing: 23.10–40.00 ft.

Class 5 — Two-Axle Six-Tire Single-Unit Trucks: Trucks with two axles and dual rear wheels. These are the most common single-unit trucks in urban delivery fleets. The LTPP classification rules require Axle 1 minimum weight of 2.5 kips and gross vehicle weight minimum of 8.0 kips for this class.

Class 6 — Three-Axle Single-Unit Trucks: Trucks with three axles and no trailer. Axle 1 minimum: 3.5 kips. Gross vehicle weight minimum: 12.0 kips.

Class 7 — Four or More Axle Single-Unit Trucks: Single-unit trucks with four, five, six, or seven axles. Includes specialized dump trucks with lift axles.

Class 8 — Four or Fewer Axle Single-Trailer Trucks: Two-axle truck or tractor pulling a one- or two-axle trailer. Total: three or four axles.

Class 9 — Five-Axle Single-Trailer Trucks: The classic “18-wheeler” or “3S2” configuration — a two-axle tractor pulling a three-axle semi-trailer. This is the dominant heavy vehicle type in US highway traffic and typically accounts for the largest share of total ESAL loading on interstate highways. Spacing between axles 1-2: 6.00–30.00 ft; axles 2-3: 2.50–6.29 ft; axles 3-4: 6.30–65.00 ft; axles 4-5: 2.50–11.99 ft. Gross weight minimum: 20.0 kips.

Class 10 — Six or More Axle Single-Trailer Trucks: Includes configurations with additional axles for higher gross weight capacity (e.g., six-axle trucks operating under special permit in states like Michigan where gross vehicle weights of 164,000 lb are allowed with more axles limiting axle loads to 13,000 lb).

Classes 11, 12, and 13 — Multi-Trailer Trucks: Vehicles pulling two or more trailers. Class 11: Five or fewer axles; Class 12: Six axles; Class 13: Seven or more axles. These configurations are common on dedicated freight corridors.

The LTPP classification rules (adopted March 2006 by the Traffic Expert Task Group) use four variables for automated classification: number of axles, axle spacing, weight of first axle, and gross vehicle weight. This is essential because axle-count-and-spacing classifiers alone cannot distinguish between Class 3 (single rear tires) and Class 5 (dual rear tires) vehicles, since both have two axles with similar spacing but vastly different pavement damage potential.

Practical significance: For pavement design, trucks in FHWA Classes 5 through 13 are the only vehicles that contribute meaningfully to structural damage. A single Class 9 five-axle truck loaded to 80,000 lb GVW generates approximately 2.5 to 3.0 ESALs per pass, while a Class 2 passenger car generates approximately 0.0004 ESALs. This means one heavy truck causes as much pavement damage as approximately 6,000 to 7,500 passenger cars.

Axle Load Spectra

Modern pavement design is moving away from the single-value ESAL approach toward axle load spectra — a detailed characterization of the distribution of axle loads by axle type (steering, single, tandem, tridem, quad) for each vehicle class. The AASHTOWare Pavement ME Design (Mechanistic-Empirical) system uses load spectra as its primary traffic input, not ESALs.

An axle load spectrum is typically presented as a histogram showing the percentage of total axle passes that fall within each load increment (usually 2,000 lb or 4.45 kN bins) for each axle configuration. For example, a Class 9 five-axle truck’s steering axle load distribution might peak at 10,000–12,000 lb, its drive tandem at 30,000–34,000 lb, and its trailer tandem at 28,000–32,000 lb.

The LTPP database contains axle load spectra from hundreds of WIM sites across North America, providing the foundation for the default load distributions in Pavement ME. These spectra vary significantly by:

• Truck type (refrigerated vans load differently from tankers or flatbeds)
• Geographic region (interstate freight corridors vs. local rural roads)
• Cargo type (bulk commodities vs. packaged goods)
• Season (agricultural harvest seasons produce heavier loads)

Axle load spectra capture the full distribution of loading rather than a single average value. Two sites might have the same total ESAL count but very different deterioration rates because one has a higher percentage of loads near the legal maximum. This is because the damage function is not linear — a 34,000 lb tandem axle causes significantly more than 34/30 times the damage of a 30,000 lb tandem.

Site-specific load spectra are recommended for major pavement projects. The FHWA’s Traffic Monitoring Guide provides guidance on developing site-specific spectra from at least 3 to 7 days of continuous WIM data, with seasonal adjustment factors to extrapolate to annual loading.

ESAL Concept and Calculation

The Equivalent Single Axle Load (ESAL) is the standard unit for expressing pavement damage from traffic. One ESAL represents the damage caused by one pass of an 80 kN (18,000 lb) single axle with dual tires. All other axle loads and configurations are converted to ESALs using Load Equivalency Factors (LEFs).

The ESAL concept originated from the AASHO Road Test (1958–1960) conducted in Ottawa, Illinois — the most comprehensive full-scale pavement test ever undertaken. The test subjected over 200 pavement sections to controlled traffic loading with known axle loads and recorded the number of load repetitions to failure. From this data, researchers derived the empirical relationship between axle load and pavement damage that is still used today.

Load Equivalency Factors (LEFs) from AASHTO 1993 (assuming Terminal Serviceability Index pt = 2.5, Structural Number SN = 5 for flexible pavements, Slab Depth D = 9 inches for rigid pavements):

The fourth power rule is a useful approximation: the damage ratio equals (actual load / standard load) raised to the fourth power. For a 30,000 lb single axle: (30,000/18,000)⁴ = (1.667)⁴ = 7.72, closely matching the AASHTO LEF of 7.9. This means a single 30,000 lb axle causes approximately 8 times more damage than an 18,000 lb axle, and over 26,000 times more damage than a 2,000 lb axle.

ESAL calculation procedure (AASHTO 1993 Appendix D):

1. Determine traffic volume by vehicle class (AADT for each FHWA class)
2. Determine axle load distributions for each class (from WIM data or default values)
3. Apply Load Equivalency Factors to each axle load increment
4. Sum damage contributions across all axles and vehicle classes
5. Apply directional distribution (DD = typically 0.50)
6. Apply lane distribution (DL = varies by number of lanes)
7. Apply growth factor to project cumulative ESALs over design life

The Truck Factor is a shorthand approach: the number of ESALs per truck for a given vehicle class. For Class 9 trucks, the truck factor typically ranges from 1.0 to 3.0 ESALs per truck depending on loading conditions. Multiplying the truck factor by the number of trucks yields total ESALs.

A fully loaded large passenger van generates approximately 0.003 ESALs, while a fully loaded tractor-semi-trailer can generate up to approximately 3 ESALs. An 80 kN single axle causes over 3,000 times more damage than an 8 kN axle (1.000/0.0003 ≈ 3,333). A 133.3 kN single axle does about 67 times more damage than a 44.4 kN single axle (7.9/0.118 ≈ 67).

AASHTO 1993 recommends a multiplier of 1.5 to convert flexible ESALs to rigid ESALs (or 0.67 to convert rigid to flexible) when comparing equivalent traffic between pavement types.

Weigh-in-Motion (WIM) Systems

Weigh-in-Motion (WIM) is the technology for measuring the dynamic tire forces of a moving vehicle at highway speeds and estimating the static axle loads and gross vehicle weight. WIM systems are the gold standard for traffic data collection because they capture axle loads, vehicle classification, and traffic volume simultaneously.

ASTM E1318-09 — “Standard Specification for Highway Weigh-in-Motion (WIM) Systems with User Requirements and Test Methods” defines the performance requirements for WIM systems:

Data items produced by WIM systems (per ASTM E1318-94 Table 1): wheel load, axle load, axle-group load, gross vehicle weight, speed, center-to-center axle spacing, vehicle class, site identification, lane and direction of travel, date and time, sequential vehicle record number, wheelbase, ESAL, and violation code (for overweight detection).

WIM sensor types include:

• Bending plate sensors: Installed in Portland cement concrete (PCC) pavements only. Measure strain from axle passage.
• Load cell sensors: Installed in PCC only. Use hydraulic or strain-gauge load cells.
• Quartz piezo sensors: Installed in PCC or asphalt concrete (AC). Generate charge proportional to applied force.
• Polymer piezo sensors: Lower-cost option for AC pavements. Less accurate at temperature extremes.
• Strain gauge strip sensors: Installed in PCC or AC.

Site selection criteria (FHWA WIM Pocket Guide, FHWA-PL-18-015):

• Horizontal curvature: For 200 ft before and 100 ft beyond sensors, radius ≥ 5,700 ft
• Vertical grade: ≤ 2% for Types I, II, III; ≤ 1% for Type IV
• AC pavement thickness: minimum 4 inches; top layer high-performance mix of 1.5–2 inches preferred
• Continuously reinforced PCC slab length: Slab length (ft) = 2.93 × (Truck Speed in mph) + 150 ft, minimum 300–400 ft for highway speeds
• Future repaving should not be planned within 5 years

Automatic Vehicle Classifiers (AVC) use axle sensors (piezo strips or inductive loops) to measure the number of axles and axle spacing pattern for determining FHWA vehicle class. AVC systems are simpler and less expensive than WIM but cannot provide axle load data. The LTPP classification rules integrate axle weight thresholds to resolve classification ambiguities — for example, distinguishing an empty Class 5 truck (single-unit, dual tires) from a Class 3 pickup truck (single-unit, single tires) requires weight data, as both have two axles with similar spacing.

Traffic Growth Rate and Projection

Traffic volumes and truck loading seldom remain constant over a pavement’s design life. Traffic growth rate accounts for the increase in both traffic volume and truck loading over time. AASHTO 1993 Appendix D Table D20 provides multipliers for given growth rates and design periods.

Annual Average Daily Traffic (AADT) is the fundamental traffic volume measure — the total annual traffic volume divided by 365 days. Future AADT is computed as:

Future AADT = AADT_current × (1 + r)^n

Where r = annual growth rate (decimal) and n = number of years in the projection period.

Growth rates vary significantly by vehicle class. Passenger car traffic may grow at 1–3% annually in urban areas, while heavy truck traffic can grow at 3–6% on major freight corridors. Regional authorities determine appropriate growth rates through analysis of historical traffic data at continuous count stations.

Real-world growth example: Interstate 5 at Mile Post 176.35 in Washington State carried approximately 200,000 ESALs per year when built in 1965, increasing to approximately 1,000,000 ESALs per year by 1994 — a five-fold increase over 30 years, equivalent to an annual growth rate of approximately 6%.

Special factors affecting traffic projections (TxDOT guidance):

• Streets becoming major arterial routes for city or school buses
• Roadways serving newly developed distribution or freight centers
• Highways impacted by new connecting roads
• Routes experiencing traffic decreases from parallel bypass openings
• Traffic increases from oil/gas field drilling or wind generator permits

Design periods: AASHTO 1993 specifies that traffic projections should cover the full design period — typically 20 years for flexible pavements and 30 years for rigid pavements. The longer design period for rigid pavements reflects their higher initial cost and longer expected service life.

Traffic growth calculation is essential because simply multiplying the original traffic count by the design life in years grossly underestimates total ESALs. For a 30-year design period with 4% annual growth, the total traffic is 56 times the first-year traffic, not 30 times.

The AASHTO reliability concept accounts for uncertainties in traffic forecasts, material properties, and construction. For high-priority routes (Interstate highways), reliability levels of 90–99% are specified, requiring thicker pavement sections to hedge against the possibility that actual traffic exceeds projections.

Comparison of Actual vs Design Traffic

Comparing actual accumulated traffic against design traffic is a critical step in forensic pavement evaluation. During a pavement condition inspection, the inspector should determine:

1. The actual accumulated ESALs on the pavement since construction or last rehabilitation (from WIM data, traffic counts, or toll records)
2. The design ESALs — the cumulative traffic the pavement was designed to carry over its design life
3. The ratio of actual to design ESALs as a percentage

This comparison provides the first diagnostic evidence:

• If the pavement shows severe distress but actual traffic is far below design traffic (<50%), the cause is likely material-related (poor construction, stripping, durability issues), environmental (freeze-thaw, thermal cracking), or structural (inadequate thickness for actual subgrade conditions).
• If the pavement shows severe distress and actual traffic is near or above design traffic (≥100%), the pavement has reached its structural design life and needs rehabilitation.
• If the pavement shows distress consistent with normal aging but actual traffic significantly exceeds design traffic (well above 100%), the pavement has been under-designed for the actual loading — a common situation on routes where traffic growth exceeded projections.

The TxDOT pavement structural model describes pavement damage as cumulative and unrecoverable. Each individual load inflicts a certain amount of damage, and when the total reaches a maximum value, the pavement has reached the end of its useful service life. Comparing actual vs. design traffic quantifies where the pavement is on this damage curve.

For overload detection, the comparison extends beyond total ESALs to the distribution of axle loads. A site with 100% of design ESALs but where 30% of trucks are overweight (exceeding legal axle limits) will show substantially more distress than a site with the same ESAL count but 5% overweight. The inspector should examine WIM data for the proportion of legal vs. overloaded vehicles.

Traffic and Pavement Distress — Overload Accelerates Cracking and Rutting

The relationship between traffic loading and pavement distress is both qualitative and quantitative. Certain distress types are directly load-associated — their severity and extent correlate strongly with cumulative traffic loading.

Fatigue cracking (alligator cracking) is the quintessential traffic-induced distress. According to the LTPP Distress Identification Manual (FHWA-HRT-13-092, 5th Edition), fatigue cracking is defined as a series of interconnected cracks caused by fatigue failure of the asphalt concrete surface under repeated traffic loading. It begins as longitudinal cracks in the wheel path and progresses to an interconnected alligator pattern. The mechanisms: repeated traffic loading induces tensile strain at the bottom of the asphalt layer. Each load application produces micro-cracking that accumulates until visible cracks form. An 80 kN single axle causes over 3,000 times more fatigue damage than an 8 kN axle. A 44.4 kN single axle must be applied more than 12 times to inflict the same damage as one repetition of an 80 kN single axle.

Rutting is a longitudinal surface depression in the wheel path, typically caused by traffic compaction or consolidation of one or more pavement layers. The LTPP DIM identifies rutting as a load-associated distress. In flexible pavements, rutting occurs when the accumulated permanent deformation from repeated traffic loading exceeds tolerable limits. Overloaded trucks accelerate rutting disproportionately because permanent deformation in unbound granular layers and subgrade is also related to stress levels by a power function.

Block cracking is primarily caused by HMA shrinkage and thermal cycling rather than traffic loading. Transverse cracking in flexible pavements is primarily thermally induced (low-temperature cracking) rather than load-associated. Edge cracking is influenced by both traffic loading and poor edge support.

Legal axle load limits (Federal US): single axle — 20,000 lb; tandem axle — 34,000 lb; gross vehicle weight — 80,000 lb. The Bridge Formula (W = 500 × [L × N / (N-1) + 12N + 36]) limits axle group loads to prevent overstress of bridges. Vehicles exceeding these limits cause disproportionate pavement damage — a 30,000 lb single axle (50% over the 20,000 lb legal limit) causes approximately (30/20)⁴ = 5.1 times the damage of a legal 20,000 lb axle.

Michigan’s unique approach allows gross vehicle weights of 164,000 lb compared to the normal maximum of 80,000 lb in other states, but with more axles limiting maximum axle load to 13,000 lb on single axles vs. 18,000 lb elsewhere. This demonstrates that the number of axles is as important as gross weight — spreading the load over more axles reduces per-axle damage exponentially.

Airport Traffic — Aircraft Mix, Gear Types, and Passes

Airport pavement design uses fundamentally different traffic characterization than highway pavement design. Aircraft loading is characterized by passes (number of times an aircraft travels over a given point), gear configuration (single wheel, dual wheel, dual tandem, dual tandem in a 6-wheel bogie), tire pressure (affects surface course stresses), and wheel load (affects structural depth).

The FAA Advisory Circular 150/5320-6G (June 7, 2021) provides guidance for design and evaluation of civil airport pavements. The FAARFIELD (FAA Rigid and Flexible Iterative Elastic Layer Design) program uses layered elastic theory for flexible pavements and layered elastic theory combined with 3D finite element theory for rigid pavements, with failure curves calibrated at the National Airport Pavement Test Facility (NAPTF).

ICAO ACR-PCR protocol (Aircraft Classification Rating / Pavement Classification Rating) replaced the older ACN-PCN method. A PCR (Pavement Classification Rating) must be determined for all pavements intended for aircraft of mass greater than 5.7 tons. The PCR is reported on a scale from 0 to 1000.

Key variables for airport traffic characterization:

• Aircraft mix: The specific aircraft types that will use the pavement (Boeing 737, Airbus A320, Boeing 777, etc.), each with different weights, gear configurations, and tire pressures.
• Annual departures: The number of takeoff operations per aircraft type per year. The critical aircraft is the one requiring the greatest pavement thickness.
• Passes: The number of times an aircraft passes over a given point. For runways, the most critical zone is typically at the end where aircraft are stationary before takeoff.
• Gear type: Single wheel (small general aviation), dual wheel (regional jets, narrow-body), dual tandem (wide-body), dual tandem with 6-wheel bogie (Boeing 777) — gear configuration determines the stress distribution through the pavement structure.
• Tire pressure: Affects surface course requirements. Higher tire pressures in modern aircraft require higher-quality surface mixes.
• Wander: Aircraft do not track perfectly in a single path like highway vehicles. The lateral distribution of passes reduces the maximum damage compared to channelized traffic.

Historically, airport pavements have performed well for 20 years (DOT/FAA/AR-04/46). The FAA uses failure criteria calibrated to the NAPTF to determine allowable passes for a given pavement structure and aircraft loading.

The four pavement structure components identified in FAA AC 150/5320-6G: subgrade (naturally occurring soil), paving materials (surface layer, base, subbase), applied loads (weight, tire pressure, location, frequency), and climate (high/low temperatures, rainfall, freeze-thaw). Traffic loading interacts with all other components in determining pavement life.

Traffic Data in PCI Trend Analysis

The Pavement Condition Index (PCI) is a numerical rating from 0 (failed) to 100 (excellent) that quantifies pavement condition based on the type, severity, and quantity of distress. PCI trend analysis uses the relationship between PCI and traffic loading to predict future condition, plan maintenance, and diagnose structural issues.

ASTM D6433 defines the PCI calculation methodology. For a given pavement section, the PCI is computed by:

1. Measuring the density (extent) of each distress type at each severity level
2. Applying deduct values from established curves
3. Subtracting total deduct from 100

Traffic data enters PCI analysis in multiple ways:

• Segmentation: Pavements are divided into management sections with uniform traffic loading, construction history, and pavement type. Traffic volume (AADT and truck percentage) is a primary segmentation criterion.
• Deterioration modeling: PCI deterioration curves are developed for combinations of pavement type, climate zone, and traffic level. For example, an arterial road with 10,000 AADT and 15% trucks will deteriorate faster than a residential street with 500 AADT and 2% trucks.
• Performance prediction: The FHWA study FHWA-HRT-18-065 used the LTPP database to develop PCI prediction models. 942 asphalt road examples were analyzed with 14 attributes including traffic volume. Decision tree models predicted PCI with >70% accuracy, identifying traffic loading as one of the most significant attributes.
• Maintenance prioritization: Sections with high traffic loading and rapidly declining PCI are prioritized for intervention to maximize the benefit per dollar spent.

Iowa DOT PCI equations (developed through ISU research) use statistical regression analysis to relate PCI to distress measurements. Different attributes contribute to the PCI depending on the type and severity of distresses present. Traffic loading is used as an independent variable in these models.

Advanced PCI prediction using machine learning (ASCE studies) can predict PCI ratings over a 4-year span using pavement distress data and severity levels, traffic volume data, pavement age, and climate factors. These models enable proactive maintenance planning based on anticipated future condition under projected traffic loading.

PCI trend analysis for forensic evaluation: When a pavement section shows a PCI deterioration rate substantially steeper than the standard curve for its traffic category, it signals an anomaly — either traffic loading exceeds the design traffic, construction quality was deficient, or material durability problems exist. Comparing the actual PCI trajectory against the expected trajectory for the measured traffic loading provides forensic evidence for root cause analysis of premature pavement failure.