Understanding Smoothness Specifications

Why Pavement Smoothness Specifications?

Why should we care about pavement smoothness? Here are five basic reasons why:

Types of Smoothness Specifications

Every state has some sort of requirement for pavement smoothness, generally in the form of a straightedge requirement such as, “Correct deviations in excess of 1/4-inch when tested with a 10 ft straightedge.” Formal smoothness specifications, however, are designed to look at overall pavement smoothness to help ensure user satisfaction.

There are two basic types of smoothness specifications: those that consider measured smoothness, and those that consider actual ride quality.

Measured Smoothness

Specifications that consider measured smoothness use devices such as rolling straightedges and profilographs to develop roughness profile traces of the pavement surface (normally measured in the wheelpaths) to calculate overall pavement smoothness indices. These indices are then related back to established thresholds of what drivers consider “acceptable” and “objectionable” in terms of smoothness. Owner-agencies then set pay adjustments based on these thresholds. To determine “must correct” or “must grind” areas, most owner-agencies make use of a bump template which identifies unacceptable local deviation from a reference plane.

Ride Quality

Smoothness specifications that measure ride quality better account for actual user perception, or what drivers really feel. Devices which are able to best measure ride quality include “response-type” devices and inertial profilers. Response-type devices measure characteristics such as vehicle suspension travel, which can be related back to established thresholds for acceptable and objectionable ride quantity. Inertial profilers, on the other hand, are able to produce profile traces which show the actual shape of the pavement surface, which can be then used to determine ride quality. Using established guidelines for what most drivers consider acceptable and objectionable in terms of ride quality, pay adjustments schedules and localized roughness criteria are developed by each owner-agency.

The type of specification used depends on the state. More and more states are realizing the importance of measuring actual ride quality, and not just raw smoothness. Ride quality better relates to user perception, and ride quality measurement devices are less susceptible to certain measurement biases. However, ride quality specifications generally require more sophisticated equipment and intensive training in measurement and interpretation of results.

Components of Smoothness Specifications

Smoothness Index System

The first component of any smoothness specifications is the smoothness index system that will be used. This system will drive the specifications in terms of how to select and certify measurement equipment, how to specify smoothness measurement for projects, how to evaluate collected data, and how to determine pay factors based on the smoothness data collected.

The two most commonly used smoothness indexes are the International Roughness Index (IRI) and Profilograph Index (PrI). IRI can be determined using measurements from any valid profiler (inertial profiler, inclinometer-based device, rod-and-level, etc.) which generates a profile trace showing the “true” shape of the pavement surface. This pavement profile is fed into an algorithm that determines the IRI value for the pavement. IRI can also be crudely measured by response-type systems using correlation to a reference profiler.

PrI is generally measured with a profilograph (California-type or Rainhart), although some software programs can compute PrI from a profile trace produced by an inertial profiler. PrI is determined by counting the number of scallops in the profile trace that fall outside of a specified blanking band. 0.2-inch (5 mm) and 0-inch (0 mm) blanking bands are most commonly used in the U.S., although a few states use a 0.1-inch (2.5 mm) blanking band as well. PrI is sometimes called Profile Index (PI) but the former is more specific.

Both IRI and PrI are reported in units of inches/mile or meters/kilometer. However, these measurements are not directly correlated and can not be directly interchanged. In general, profile traces are collected in either one or both of the wheelpaths within a pavement lane, although some states require “quarter-point” measurement instead. Some states average the values from the two wheelpaths, while other states use only one wheelpath for smoothness assessment.

Two indices that are also commonly used and are derivatives of the IRI are the Mean Roughness Index (MRI) and Half-car Roughness Index (HRI). The MRI is simply the average of the wheelpath IRI values reported for a given segment or lot. The HRI is calculated by applying the IRI algorithm to the average of the wheelpath profiles (see the Little Book of Profiling for a more detailed explanation).

Some of the other smoothness indices used in the U.S. include:

Developed in early 1980’s under the National Cooperative Highway Research Program (NCHRP), the RN was later revised and standardized by the University of Michigan for the FHWA. This index provides a prediction of mean panel rating (MPR) from profile data that estimates user perception of ride comfort. RN is reported as a number between 0 (poor ride quality) and 5 (excellent ride quality). The computation of RN should be using profiles from both wheelpaths. Though RN can also be calculated from a profile with either of the wheelpaths, research under NCHRP and later by the University of Michigan Transportation Research Institute (UMTRI) showed that this results in a poor estimate of MPR. In FHWA's ProVAL software, RN computed using both wheelpaths versus a single wheelpath is termed “two-channel RN” and “RN”, respectively.

MRN is a smoothness index produced by response-type devices, such as the Mayes Ride Meter. MRN is reported as a measurement in inches/mile (meters/kilometer), but does not have any direct correlation to PrI values. There is, however, usually a very good correlation between MRN and IRI.

CSI is similar to the Profilograph Index, but is produced by a specific rolling straightedge device.

RQI, developed by the Michigan Department of State Highways in the late 1960’s, is an index mathematically calculated from a profile produced by an inertial profiler device, and is reported as a number between 0 (perfectly smooth) and 100 (extremely rough), although the theoretical maximum is 141.85 with no minimum. The RQI value is correlated to user perception of pavement ride quality by linking it to content at key wavelengths in profile power spectral density (PSD) function.

Equipment Types

The equipment type used for smoothness measurement is essentially dictated by the smoothness index system. For IRI/HRI/MRI systems, inertial profilers or inclinometer-based devices (walking profilers, Dipstick) must be used. Both high-speed and lightweight inertial profilers are available. High-speed inertial profilers can run at highway speeds and are appropriate for long sections of pavement that need to be tested under traffic. Lightweight profilers are ideal for testing new pavements that may have not reached “opening to traffic” strength. Lightweight profilers travel at lower speeds and may not require as much “lead-in” distance as high speed profilers, making them ideal for testing shorter segments that are closed to traffic.

For the PrI system, profilographs (with either automated or manual trace reduction) are normally used, but some states also permit inertial profilers to be used as long as the data collected by the device can be used to accurately simulate profilograph traces. No matter which device is used, device certification is required, particularly if the contractor is conducting the testing. Most states have separate Test Methods or Test Procedures for the smoothness measurement devices that are permitted.

Operation and Evaluation

Procedures for smoothness measurement equipment operation need to be specified to ensure that data collection quality meets the owner-agency requirements. Some states may even require operator certification, particularly for inertial profilers, since it is not a trivial task to operate and collect profile data correctly. States which require the contractor to conduct smoothness testing normally require both operator and profiler certification.

Evaluation or analysis of the smoothness data and the report elements/format are also often specified. For inertial profiler data collection, it is important to make sure that the profile data format is compatible with standard software such as ProVAL. The benefits of this requirement in the specifications will help to ensure that the owner-agency and contractor can readily exchange data and resolve discrepancies.

Report Segment Length

The report segment length (or “lot length”) for pavement smoothness is a very important component due to the averaging process of smoothness index computation. The most commonly used report segment length is 0.1 mile or 528 ft (160 meters). Occasionally, 0.05 mile or 264 ft (80 meters) lengths may also be used. Keep in mind, that in general, the longer the segment the lower the ride index values due to the amount of data that is averaged. Normally, pay adjustments are based on the smoothness index reported for each lot.

Localized Roughness

In addition to the overall smoothness index report, localized roughness is often identified and reported separately. Localized roughness (or “hot spots” are isolated areas of roughness, which by themselves can cause a significant increase in the overall reported smoothness index.

The main reason for identifying localized roughness is that it may be objectionable, and possibly hazardous, even if it appears on an otherwise smooth road. However, the hot spot may be “hidden” if no localized roughness provision is used and it is not quite severe enough to cause the overall smoothness index value to exceed a certain threshold. Therefore, most states specify localized roughness criteria separately from the overall smoothness index threshold so that these areas can be identified and corrected separately.

Incentives/Disincentives and Corrective Action

Most states are willing to “pay for smooth pavement” in the form of incentives in the smoothness specifications. Virtually all states require that the contractor either correct a pavement that doesn’t meet a specified smoothness level or accept a payment reduction. For states which do provide incentives and/or disincentives, pay adjustments generally take the form of either a lump-sum dollar amount for each lot, or a multiplier applied to the contract unit price paid for the paving material. Most specifications require corrective action only for localized roughness. Some states, however, permit the contractor to accept a lump-sum “penalty” for each localized roughness incident, in lieu of corrective action.

Many different methods have been used to determine incentive and disincentive levels. Each state must carefully evaluate what they are willing to accept and what they are willing to “pay for” in terms of pavement smoothness. If enough data is available, these incentives/disincentives should be based on life-cycle cost analyses that account for initial as-constructed smoothness, long-term smoothness, and overall pavement performance. The four basic smoothness index thresholds which will be found in any incentive/disincentive smoothness specification include:

Smoothness specifications should describe what forms of corrective action are permitted. Most states permit diamond grinding for correction of both PCC and HMA pavement surfaces. Some states require full removal and replacement or an additional overlay for correction of HMA pavement. The majority of states which provide incentives in their specifications will not pay incentives after corrective action has been completed unless the corrective action consisted of full removal and replacement.

Transition System

When states make a major change in smoothness specifications, it is often necessary to have a transitional period in order for the contractors and owner-agencies to become more comfortable with the specification. Such might be the case if a state is changing from a Profilograph Index specification to an IRI specification. This transition makes it easier for contractors to “digest” and adjust to the new specification, while also permitting the owner-agency to evaluate and make adjustments to the specification. During transitional periods a developmental or pilot specification may be used. Often, these developmental or pilot specifications limit or even eliminate the disincentives for “pilot projects” that are constructed under the new specification. The maximum incentive is also often limited. This reduces the risk to the contractor and allows them to evaluate how well their paving operations will perform under the new specification so that changes can be made.