Guide for Pavement Friction (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

D-1 APPENDIX D SUMMARY OF STATE AND INDUSTRY INTERVIEWS

D-1 INTRODUCTION This appendix provides a summary of the results of interviews with selected states and industry organizations, supplemented with pertinent information from the project literature. It provides additional information (in addition to the findings of the state friction survey) regarding current and near-future practices, ideas, and issues related to friction, including economic factors. This summary is based on information obtained from representatives of state departments of transportation (DOTs), industry organizations, and experts or practitioners in the pavement surface characteristics field. It is organized to address the following important aspects of pavement friction: • Friction management. • Friction testing. • Determining friction demand. • Pavement surface selection and design. • Pavement construction. • Economic considerations. • Noise-related issues. • Suggested improvements to friction practices and desired areas of friction guidance. One or more representatives were contacted and interviewed at the following state DOTs: Agencies • Arizona (ADOT). • California (CALTRANS). • Florida (FDOT). • Georgia (GDOT). • Illinois (IDOT). • Louisiana (LADOTD). • Michigan (MDOT). • New York (NYSDOT). • Texas (TXDOT). • Virginia (VDOT). • Washington State (WSDOT). Representatives of paving associations, truck manufacturers, tire manufacturers, equipment manufacturers, and others were also contacted and interviewed. These included representatives from the following organizations: • American Concrete Pavement Association (ACPA). • National Asphalt Pavement Association (NAPA). • California Chip Seal Association (CCSA).

D-2 • Rubber Manufacturers Association (RMA). • International Grinding and Grooving Association (IGGA). • Mack Trucks. • Kenworth Truck Company. • Volvo North America Group. • International Cybernetics Corporation. • Dynatest Consulting. • Texas Transportation Institute (TTI). • Transportation Research Center (TRC). • N.V. Robuco of Belgium. FRICTION MANAGEMENT PRACTICES The interviewed states currently take more of a monitoring approach to pavement friction than a design approach. In other words, they generally put more emphasis on routinely evaluating friction levels and crash rates and reacting to deficiencies than on designing pavements to satisfy friction demands. Although there may be several reasons for this tendency, the type and quality of aggregates available for use is often a factor. In some states, such as Georgia, New Hampshire, and Washington, where good quality aggregates are largely available, the need to design mixes for friction is less of a priority because of the good performance provided by the aggregates. In states such as Florida, Louisiana, and Texas, where lower quality aggregates are more common, friction design generally is an integral part of the overall friction management program. Each of the interviewed states, as required by law, maintains a crash database and makes use of the data in identifying pavements with potential friction deficiencies. Various methodologies are used in analyzing crash data, and the results of such analyses are used in different ways. For instance, in Arizona, both high crash rates and low friction numbers can serve as triggers for special investigation, whereas in Illinois, the identification of high wet-weather crash rates by a district is usually a prompt for friction testing by the central office. FRICTION TESTING PRACTICES As reported in appendix C, the NCHRP Project 1-43 survey of states established that nearly all agencies use an ASTM E 274 locked-wheel friction testing device for pavement friction management and evaluation. Fifty-six percent of reporting agencies use the ASTM E 501 ribbed tire exclusively, and 15 percent use the ASTM E 542 smooth tire exclusively. The remaining agencies use a combination of tires for management and crash investigation. The quality of ASTM E 274 data collection is in large part controlled by the maintenance of the equipment, the training and dedication of the operator, and the adequacy of calibration activities. Manufacturers of locked-wheel friction testers indicate that data quality is most affected by the training, experience, and attention to detail of the equipment operators (Olinoski, 2004; Beck, 2004). They note maintenance problems such as tire and brake pad

D-3 wear can affect data quality. Infrequent or improper calibration activities are considered as more typical inhibitors to good data quality. They emphasize daily checks as well as monthly and regular calibration at the Texas Transportation Institute (TTI) or the Transportation Research Center (TRC) calibration facilities. Because of reported nozzle clogging, TTI suggested monitoring water flow as a method to ensure adequate water supply and accurate friction numbers (Zimmer, 2004). ASTM E 274 equipment manufacturers are providing laser texture measuring equipment and software for requesting agencies. At least four ASTM E 274 trailers with texture measurement capability have been delivered to agencies in the U.S. by March 2004 (Olinoski, 2004; Beck, 2004). Calibration checks for the texture lasers are as critical as calibration checks for the locked wheel measuring systems. Currently, equipment manufacturers are recommending that texture lasers be calibrated in a manner similar to those used for standard smoothness profiling. However, they recognize a need for better confirmation of measurement accuracy and are considering additional calibration methods for mean profile depth (MPD) data collection. The design of friction testing equipment can also affect friction number (FN) accuracy and repeatability. Reported recent equipment upgrades intended to improve data quality include moving the signal digitization to the trailer from the driver’s position to reduce analog signal distortion from long cables. Optical distance measuring wheel encoders have also been installed to replace potentially incorrect servo tachometers. Manufacturers tend to avoid torque tubes, which have limited ability to account for vehicle dynamics in the vertical load. In addition to adding texture measurement capability, other changes are on the near horizon for ASTM E 274 equipment. Improvements being considered include adding the ability to measure locked-wheel friction in increments of distance as well as time. This is a topic of discussion for the ASTM E-17 committee and will allow better correlations between friction measurements at different speeds. Global positioning equipment may also be integrated. Manufacturers recommend daily and regular calibration of ASTM E 274 equipment to maintain data quality. Two national ASTM E 274 calibration centers also provide rigorous equipment calibration testing that agencies typically complete every 2 to 3 years. TTI and TRC provide ASTM E 274 locked-wheel tester calibration services for about $11,000 per unit (Zimmer, 2004; Lyon 2004). This service provides static and dynamic calibration testing according to ASTM E 1890 using 108 runs on three calibration pads. Comparison with the TTI standard equipment allows agencies to inter-correlate their equipment output and compare their friction numbers with those of other agencies. Neither TTI nor TRC has developed methods for checking the calibration of laser texture measuring equipment. PAVEMENT SURFACE SELECTION Selecting pavement types and materials to meet frictional and noise needs reasonably requires estimating the frictional “demand” of a pavement. This “demand” is difficult to define but can be estimated as a function of traffic levels, climatic conditions, required maneuvers (braking, turning, accelerating, steady state), vehicle types (percent trucks), and

D-4 other factors. Very few agencies reportedly have developed a methodology for estimating the friction “demand” of a pavement. Texas and Maryland are among the first. The first step in the Texas Wet Weather Accident Reduction Program (WWARP) is to determine the overall frictional demand on a road surface (Stampley, 2004). This is accomplished by rating the demand as low, medium, or high based on the factors and levels shown in table D-1. These factors are grouped as either (A) general attributes or (B) parameters set by the designer. Recommendations are also given regarding which factors are more critical; however, the overall rating remains slightly subjective. Table D-1. Texas WWARP friction demand classification. Factor Attribute Low Moderate High A Rainfall, in/yr < 20 >20 < 40 > 40 A Traffic, ADT < 5000 >5000 < 15,000 > 15,000 A Speed, mi/hr < 35 >35 < 60 > 60 A Percent Trucks < 8 >8 < 15 > 15 A Vertical grade, % < 2 >2 < 5 > 5 A Horizontal curve, deg. < 3 >3 < 7 > 7 A Driveways per mi < 5 >5 < 10 > 10 A Intersecting Roadway ADT < 500 >500 < 750 > 750 B Cross slope, in/ft 0.375 – 0.5 0.25 – 0.375 < 0.25 B Design life, yr < 3 >3 < 7 > 7 B Proposed macro-texture Coarse Medium Fine 1 in = 25.4 mm 1 mi = 1.61 km/hr 1 mi/hr = 1.61 km/hr 1 in/ft = 0.083 m/m The Maryland DOT differentiates friction demand on straight segments and curves. For straight segments, five demand categories are selected according to the descriptions in table D-2. Additional differentiation is given in regard to mean speed and the percentage of trucks expected on the roadway. Guidelines for side friction factor requirements for curves are ranked according to driving complexity and curve radius, as shown in table D-3. They provide additional differentiation in regard to average speed and super-elevation.

D-5 Table D-2. Maryland DOT straight segment friction demand classification. Site/Demand Category Site Description 1 (high) Approach railroad crossing, traffic lights, pedestrian crossing, stop and give way controlled intersections 2 (medium to high) Curves with radius < 820 ft (250 m), downhill gradient > 10 percent, and > 164 ft (50 m) highway on/off ramp 3 (medium) Approach to intersections, downhill gradient 5 to 10 percent 4 (low to medium) Undivided highways without any other geometrical constraints which influences friction demand 5 (low) Divided highways without any other geometrical constraints which influences friction demand Table D-3. Maryland DOT curved segment friction demand classification. Driving Complexity* Radius, ft (m) 328 (100) High (UR =1) 820 (250) 328 (100) Critical (UR =0.875) 820 (250) 328 (100) Considerable risk (UR =0.675) 820 (250) 328 (100) Very dangerous (UR =0.5) 820 (250) *High = high safety standard required, critical = critical driving maneuvers possible, considerable risk = considerable crash risk exists, very dangerous = very dangerous, high crash risk. The New York DOT is currently researching the relationship between vehicle energy input (maneuvers), friction needs, and aggregate properties for use in defining friction “demand” (Skerritt, 2004). Other highway agencies specify their aggregate and texture properties based on traffic levels. For example, Illinois requires turf drag and transverse tining at 20- mm (075-in) spacing for roads with posted speeds in excess of 65 km/hr (40 mph). For lower speed roads, the finish can be turf drag with or without transverse tining (Rowdan, 2004).

D-6 PAVEMENT SURFACE FRICTIONAL DESIGN Asphalt Concrete Design Designing asphalt concrete (AC) pavements to meet frictional “demand” requires selecting mix designs and aggregate types and properties that can adequately provide long-term friction. Commonly, agency frictional design programs attempt primarily to control the micro-texture properties of the course aggregate. Aggregate types are differentiated and selected by carbonate content (e.g., limestone), British Pendulum Test/Polish Value (AASHTO T 278), magnesium phosphate soundness (AASHTO T 104), LA Abrasion (ASTM C 131), crushed particle ratio (ASTM D 5821), and others. Some agencies with polish- resistant aggregate achieve good frictional performance regularly and have not been required to significantly test their aggregates for frictional properties (Geary, 2004; Pierce, 2004). Other agencies must use available limestone or dolomite aggregates or pay large shipping fees. These agencies, in particular, are working to implement aggregate tests that will ensure good frictional performance. The Texas DOT has developed an aggregate rating system that classifies coarse aggregate source materials into four categories (A, B, C, and D) to match their demand classifications. These ratings are updated semi-annually based on aggregate properties from approved resources (TXDOT 2004). Source aggregates are rated according to polish value, LA Abrasion, and magnesium phosphate soundness for hot mix asphalt concrete and surface treatment applications. Suggestions for blending are also provided. Testing of the aggregate used in construction is completed to verify the classification, and post- construction frictional testing is conducted to ensure adequate friction levels. For selecting AC paving materials to best correlate with field performance, agency laboratory testing must evaluate short- and long-term micro-texture and macro-texture. The Michigan DOT has attempted to measure the overall friction properties (micro-texture- and macro-texture-related) of their coarse aggregates by running full-size tires on a large- scale track with embedded samples of uniformly graded aggregate. They then apply a scale version of a towed friction trailer to the worn surface to measure an Average Wear Index (AWI) representative of changes in frictional resistance with polishing. Shortcomings of this approach are the inclusion of only the coarse aggregate and the cost of the equipment and testing. The first limitation is evident in the variability between AWI ratings and field friction numbers (McDaniel, 2004). Another more comprehensive approach recently developed by the National Center for Asphalt Testing (NCAT) includes a combination of a circular track polishing machine, a Dynamic Friction Tester (DFT), and a Circular Texture Meter (CTM). The NCAT polishing machine uses three tires on a circular track of the same diameter as the CTM and DFT 11.2 in [284 mm]) (Nippo Sangyo, 2004). The device allows for lowered costs and full measurement of the polished surface. Using the DFT and CTM results, the International Friction Index (IFI) measurement can be obtained. This approach has been reviewed by the McDaniel of the Institute for Safe, Quiet, and Durable Highways (ISQDH) and is proposed for use in full-scale field comparisons to be completed by December 2004. The planned experimental matrix includes three gradations, two nominal maximum aggregate sizes, five aggregates, and three friction aggregate contents (McDaniel, 2004).

D-7 Portland Cement Concrete Design Designing portland cement concrete (PCC) pavements for good frictional characteristics primarily requires selecting adequate surface texturing methods and secondarily requires selecting adequate aggregate properties. The PCC texturing properties used by state agencies indicate a sustained reliance on tining for producing macro-texture in PCC pavement surfaces. Further review of available agency specifications indicates a significant use of longitudinal tining, as table D-4 shows. The wide variety of random transverse spacing designs indicates that agencies are not regularly following the recommendations of the Wisconsin randomization study, although the reason is not defined. By far, the most common method of PCC tining includes transverse tining using 0.5-in (13-mm) spacing. Georgia DOT indicates that they have not had pavement-tire noise problems using this method (Geary, 2004). Table D-4. Standard specified PCC surface tining methods. Tining method Tine spacing, in (mm) Agency 0.5 (13) AK, CN, DE, GA, MI, MO, MS, SC 0.75 (19) AK, IL, IA, MT, NY, VA Transverse uniform 1.0 (25) TX Max. 0.5 (13) ID Max. 0.75 (19) HI 0.19 – 1.25 (5 – 32) IN 0.3 – 1.0 (8 – 25) TN 0.375 – 1.5 (10 – 38) LA 0.375 – 1.625 (10 – 41) IA 0.375 – 1.75 (10 – 44) OH 0.5 – 0.75 (13 – 19) NC 0.5 – 1.0 (13 – 25) FL, OK 0.5 – 1.25 (12 – 32) OR, WA 0.625 – 0.875 (16 – 22) MD, UT Transverse random 0.875 – 1.25 (22 – 32) NJ Longitudinal uniform 0.75 (19) AZ, CA, CO, NE, NV, NY Longitudinal random 0.5 – 1.0 (13 – 25) FL Other PCC texturing methods, used regularly in Europe, include Exposed Aggregate Cement Concrete (EACC) and Enhanced Porosity Concrete (EPC). Recently a 10.3-mi (16.6-km) section of the E40/A10 freeway from Brussels to Belgium carrying 57,000 ADT was paved using EACC and a stringless paver (Gomaco, 2004). Mr. Romain Buys of NV Robuco in Belgium indicates that the Austrian Cement Association has 8 to 9 years of experience with large-scale two-layer porous PCC construction. The layers are placed wet using a single modified paver. Aggregates used in the lower layer are typically PCC recycled from the original pavement, and the upper 1.4 in (35 mm) layer is designed using open graded quartzite aggregate (Buys, 2004).

D-8 Other Surface Property Factors Other surface property factors that need to be accounted for in designing pavements are pavement-tire noise, splash and spray, hydroplaning, and rolling resistance. Tire wear and glare also can be considered. Noise issues are discussed in a later section. Splash and Spray Splash and spray thrown against windshields by passing vehicles are a potential hazard in that these impair visibility, especially at night. However, the safety hazard created by splash and spray has not been precisely defined. The percentage of wet weather crashes directly attributed to splash and spray ranges from 1 to 10 percent (ISPA, 1977; Sabey, 1973). Splash is defined as the large droplets of water that are thrown off the tire or squeezed out from the pavement–tire contact area. Splash is associated with "large" water depths or low vehicle speeds. Spray is the mist that is carried alongside and thrown behind a vehicle by the turbulent airflow created by a moving vehicle. It is associated with shallow water depths or high vehicle speeds. Increasing macro-texture decreases splash and spray intensity and duration, thus improving safety through better visibility (Pilkington, 1982). As a vehicle travels through water on the pavement, the water is splashed both outward and inward from the rolling tires. It is also thrown forcefully backward by the front tires into the following tires and the vehicle's surfaces, where it is broken into smaller droplets. These smaller droplets are more easily affected by air turbulence and wind. The droplets, together with water blown and vacuumed off the road and vehicle surfaces by fast-moving vehicles, contribute to spray (WHI, 1973; Wambold et al., 1984). As water falls or runs onto a pavement, a certain initial amount is required to fill the pavement surface texture before runoff occurs. This is known as depression storage. When the surface voids are filled, runoff begins and increases to a constant value. The thin sheet of water on the surface at this time, excluding depression storage, is known as surface detention (Galloway, 1975). Besides reducing traction, surface retention contributes to splash. After runoff ceases, depression storage produces spray. This effect is longer lasting and causes poor visibility because of windshield splatter. Various techniques have been used for measuring splash and spray. Quantitative measuring instruments include densitometers, photometers, spraymeters, and spray collectors (Ritter, 1974). Photographs of a test vehicle, as it travels through a wet test course, can be front, side, oblique, or rear views, and the spray density can be evaluated subjectively by examining the pictures. The principal factors contributing to splash and spray are water, air, the driver, the vehicle, speed, the highway geometry, and the texture (WHI, 1973; Wambold et al., 1984). The water factor includes rain, snow, slush, mud, and muddy water. The form of the moisture, the amount, and its location on the highway are also important.

D-9 To reduce pavement water depths and drainage times, the following design, construction, and maintenance standards have been proposed (WHI, 1973; Wambold et al., 1984): • Increase cross slopes. • Reduce the number and length of zero grades by alternating slight plus and minus gradients. • Eliminate shoulders that tend to prevent water drainage. • Construct lateral grooving rather than longitudinal grooving. • Slope multilane divided highways away from the median. • Avoid excessively smooth surfaces, especially those with flat cross slopes. • Keep shoulders and outside pavement edges free of ice, snow, and slush. Macro-texture is more important than micro-texture in minimizing splash and spray, because it promotes water drainage (Bonds et al., 1974). Open-graded asphalt pavements generally are better than either PCC or dense-graded AC pavements in allowing surface water to drain because of their more desirable macro-texture. Pavements designed to reduce hydroplaning will also reduce splash and spray. Procedures have been published for both U.S. (ISPA, 1977; Gallaway et al., 1975; FHWA, 1973a; FHWA 1973b; NCHRP, 1974; NCHRP, 1978) and European (Cram, 1975; Sorenson et al., 1974) open-graded pavement designs that optimize water- drainage characteristics. Although open-graded AC pavements have a macro-texture that permits surface water to drain easily, they may also have some drawbacks (WHI, 1973; Wambold et al., 1984): • Load-carrying ability is less. • Dirt and oil can clog the voids and reduce drainage ability. • Voids can be sealed off by bleeding due to compaction under load. • Bases and sub-bases can become contaminated, thereby structural stability. • Subgrades can swell, heave, and creep, thus deforming breaking up pavement and base structures. • Water freezing in any structural layer can result in the entire roadway structure. Hydroplaning An important non-friction-related effect of pavement surface texture is its role in the prevention of hydroplaning. There are two highway-related forms of hydroplaning: dynamic hydroplaning and viscous hydroplaning. Pavement surface macro-texture texture plays an important role in the prevention of each. Dynamic hydroplaning occurs when the vehicle tire loses contact with a flooded pavement and rides on a layer of water. Macro-texture influences dynamic hydroplaning in two ways: it has a direct effect on the critical hydroplaning speed because it provides a pathway for water to escape from the pavement–tire interface, and it has an indirect effect on the critical hydroplaning speed through its effect on the water depth on the pavements (the larger the texture, the deeper the water must be to cover it). In both cases, increases in macro-texture depth tend to increase the critical hydroplaning speed. However, the road must also have the proper micro-texture to develop friction.

D-10 Viscous hydroplaning occurs when a thin film of water remains between the tire and the pavement with insufficient micro-texture to break through the film. Micro-texture prevents viscous hydroplaning because the small asperities penetrate the water film and allow semidry contact of the pavement with the tire. Excessive amount of water on the pavement surface causing hydroplaning can be reduced through design the pavement surface characteristics. Several factors influence the water film thickness on a pavement surface and thus the potential for hydroplaning. Identified factors include: • Pavement (micro-texture, macro-texture, cross-slope, grade, width, curvature, and longitudinal depressions). • Environmental (rainfall intensity and duration). • Driver factors (speed, acceleration, braking, and steering). • Vehicle (tire tread wear, ratio of tire load to inflation pressure, vehicle type). Rolling Resistance and Fuel Consumption Pavement surface texture has a small but potentially important influence on the fuel consumption of vehicles on the highway. Pavements with high levels of micro-texture and macro-texture cause some increase in fuel consumption. Fuel consumption can increase by as much as 10 percent, although the typical difference is much smaller (less than 2.0 percent). The exact relationship between pavement surface texture and fuel consumption has not been determined. However, an estimate of the influence can be inferred from independent relationships between pavement surface texture and tire rolling resistance and between tire rolling resistance and fuel consumption. Members of the trucking industry are particularly concerned that pavement designs optimize friction and rolling resistance properties (Yeakel, 2004). Tire and Pavement Wear Pavement–tire interactions cause wear of both the tire and the pavement. Wear rates of each are influenced by the texture of the pavement; however, wear is very slow and is affected by many other factors. For this reason, only minimal data have been obtained that document the texture effect; the conclusions drawn are largely qualitative. Traction on dry pavement is a function of grip and friction, both of which improve with an increased density of asperities in the micro-texture range. Generally, this same texture also leads to more rapid tire wear. Macro-texture is a relatively unimportant factor with regard to tire wear or traction, although increased macro-texture can cause more wear through rubber reversion (during skidding) and can result in reduced grip, both effects being disadvantageous. In general, dense-graded pavements (low macro-texture) appear to be more wear-resistant than open-graded pavements (high macro-texture). Uniformly macro-textured surfaces, such as those created by fluted float, rotating drum, or grooving processes, wear at a slower

D-11 rate than irregularly macro-textured surfaces created by such processes as burlap dragging and brooming. Glare, Light Reflection, and Night Visibility Work in the 1950s on the reflection characteristics of dry pavements determined that highly bright and specular (mirror-like) surfaces were best for high angle lighting. However, these pavements were unacceptable because of their slipperiness. Lighting under wet conditions is complex, and only very limited information exists regarding these properties. There has been some European research; however, and extensive work was reported in a Danish document (Sorenson, 1974). Increasing texture reduces pavement luminance, which in turn reduces the visibility of lighted roads. Increasing texture also provides better water drainage, thereby reducing glare for wet pavements. Retro reflection, light from headlights reflected back toward the driver, increases with macro-texture depth until the macro-texture peaks shadow one another. Night visibility of painted pavement markings is enhanced by macro-texture depth because the paint in the voids is protected from wear. Diamond Grinding Diamond grinding has been used in new and rehabilitation designs for PCC pavements to improve ride quality and reduce pavement-tire noise. Properly designing PCC grinding operations also requires research and understanding of aggregate properties and grinding techniques. On-going and previous research indicates that blade spacing of PCC grinding equipment is critical to both noise and smoothness (Roberts, 2004; Buys, 2004). Typically, narrower grinder blade spacing gives smoother pavements and faster wear, but wider spacing results in less pavement-tire noise and extended skid resistance. Practically, harder aggregates require more narrow spacing, due to the “unbroken aggregate fins” that form with wider spacing. Design blade spacing should be selected according to aggregate hardness to optimize “fin” breakoff. A Belgian contractor uses a device that evenly breaks off “fins,” allowing for wider spacing and less pavement-tire noise (Buys, 2004). Reportedly, Iowa has a specification for blade spacing as a function of aggregate hardness. South Dakota is doing research to develop such a specification, and a Belgian contractor has developed a similar relationship for European aggregates (Roberts, 2004; Buys, 2004). Microsurfacing Agencies describe using properly designed microsurfacings as fast, but sometimes expensive, methods for restoring pavement friction. The CCSA recommends setting limits on aggregate types for chip seals to avoid early wear, using only crushed aggregate, and avoiding sandstone. They also recommend allowing no more than 25 percent loss on the LA Abrasion test for chip seal and microsurface aggregate (Metcalf, 2004).

D-12 PAVEMENT CONSTRUCTION The methods and equipment used in constructing and rehabilitating AC and PCC pavements can also affect a pavement’s short- and long-term friction and noise properties. As an example, field experience has shown that higher friction, on the order of 5 friction points, can be achieved by operating the hot mix asphalt (HMA) laydown machine in the direction of vehicle traffic. Similarly, the type of compaction equipment and rolling patterns can influence the surface friction (i.e., rubber-tired rolling versus steel-wheeled rolling). The manner of construction of friction restoration treatments, such as chip seals, slurry seals, microsurfacing, and proprietary surfaces (e.g., NovaChip®, ItalGrip), are all susceptible to providing less than expected friction if poor construction practices are employed. Spacing of the diamond blades in grinding machines many times can be optimized during construction to achieve the best noise, smoothness, and long-term friction properties. ECONOMIC CONSIDERATIONS There are several economic aspects or considerations associated with improving pavement friction and friction-related characteristics. These considerations center around the costs and benefits of (a) managing pavement friction for safety, and (b) selecting, designing, and constructing surfacings for new pavement structures and restoration treatments for existing pavements. The economic considerations of friction and friction-related items can take the form of direct agency costs/benefits or indirect costs/benefits accrued by highway users. Moreover, economic impacts can take place at the network level or the project level. Detailed discussions of these considerations, based on the latest available literature and recent interviews with selected state agencies and industry organizations, are presented below. Pavement Friction Management As indicated earlier, state agencies take varied approaches to ensuring the adequacy of friction on their highways. The approaches range from a simple evaluation of crash rates with no or limited consideration of friction in design to an elaborate process involving network-level friction testing, detailed crash analysis, and friction performance-based design strategies. Naturally, program costs vary widely, as they depend on the amount of personnel and resources (computer hardware and software, laboratory equipment) needed to develop, implement, and operate each component of the program. As an example, the Florida DOT has estimated costs of $250,000 to develop and implement their friction database, $2,000/year for database software maintenance, and over $1 million/year for data collection and input (Brady, 2004). Costs for California’s friction data collection and processing have been roughly estimated at $350,000/year, derived using assumptions regarding the size of the highway inventory tested, the number of friction test operators and their labor rates, testing productivity, and equipment maintenance and depreciation costs (Vacura, 2004).

D-13 The benefits of such programs can be examined in terms of the number of crashes and/or crash-related costs (fatalities, injuries, vehicle and property damage), or in terms of the friction-related litigation costs experienced by the agency. By comparing the overall benefits and costs associated with expanding or cutting back certain components of a friction program, agencies can make more informed decisions regarding the appropriate size and scope of their program. A further extension of this type of analysis involves the use of risk management principles and procedures. Selection, Design, and Construction of Pavement Surfacings and Restoration Treatments The process of selecting, designing, and constructing surfacings for new pavements and restoration treatments for existing pavements provides many opportunities for economizing highway pavement projects. This is because there are numerous materials, mixes, and construction techniques that form an array of pavement textures that generate a wide variety of surface friction, noise, drainage properties (i.e., splash/spray, hydroplaning). Moreover, under the loading applications of traffic and environment over time, these surfacings and treatments exhibit different texture wear characteristics, which impact their ability to meet the required or desired levels of friction, noise, and drainage. Because the mix/material types have varying costs and may perform differently over time, their overall life cycle costs can be substantially different. These agency cost differences can be further impacted by the inclusion of user costs associated with normal operating conditions (i.e., crash, time delay, and vehicle operating costs resulting from deficient friction, noise, and/or drainage) and the timing of future work zones (i.e., if one mix/material type fails functionally before another). Provided in the sections below are summaries of how economics factor into the friction design and management of asphalt and concrete pavements, respectively. The summaries are based on detailed reviews of compiled literature, as well as informative discussions with various state agency representatives. Asphalt Surfacing Mixes and Restoration Treatments The economic considerations for asphalt surfacings and treatments center around the type, quality, and gradation of aggregate used in the mix/material (coarse and fine dense-graded AC, OGFC, SMA, microsurfacing, chip seals, specialty surface treatments [Novachip®, Italgrip]) and, to a lesser extent, the type of asphalt binder (conventional, SuperPave, PMA, RA) used. The economic consequence of using alternative mixes/materials requires the acquisition and analysis of up-to-date unit bid prices and pertinent forms of time-series performance data (e.g., distress, ride, friction, noise). Information concerning construction time requirements and other construction-related issues (e.g., production, placement, finishing, lab and field testing) is also important. Although the compiled literature contained documents focusing on the performance characteristics of different asphalt mixes, only a few included information on the associated costs and/or economic impacts of the mixes. Fewer still focused specifically on friction and friction-related performance.

D-14 Studies done by the Oregon DOT (Hunt, 2002) and Florida DOT (Choubane et al., 1999), for example, looked at the performance and costs of different asphalt surface mixes (in Oregon, wet- and dry-processed rubber-modified mixes [dense- and open-graded]; in Florida, conventional and rubber-modified dense- and open-graded mixes). However, while friction performance was evaluated in both studies, structural and ride performance were the primary bases for determining the cost-effectiveness of the mixes. Perhaps the most pertinent study found in the literature was one done by the Maryland SHA (Chelliah, 2003). As part of a research effort begun in 2002 to develop a design policy to improve pavement surface characteristics, the Department performed and illustrated an example benefit/cost analysis comparing pavements designed and built at two different friction levels (FN=45 and 35). Using their own crash prediction model and crash cost statistics (for fatalities, injuries, and property damage), as well as estimates of performance and construction costs, a benefit/cost ratio of nearly 7 was determined, showing the economic advantage of using a pavement surface with higher friction. The study noted that using mixes with aggregates having polish values of 7 or higher may require importing aggregates from outside the region. However, when the cost of wet crashes is compared to the initial cost outlay of importing aggregates, the latter will be a much more economical option. Maryland’s research into friction design resulted in the adoption of an economically based design policy to minimize future wet-surface crashes. The policy entails the following: • Check if wet-weather problem exists. • Select target design friction level. • Predict potential reduction in wet crashes. • Calculate benefit/cost ratio of design target. • Evaluate effectiveness of design. • Select surface mix with adequate polish value. Interviews with selected state agencies did not reveal any specific studies involving the economic impacts of AC friction design and management. However, several points of interest were brought up in the discussions. Illinois, for example, described how its recent move to higher blend aggregates had an economic component to it, in that more locally available aggregates are now used, which provide comparable friction performance at a lower cost (Rowden, 2004). Michigan discussed its need last year to revise its asphalt specifications to include aggregate wear index testing for Superpave mixes with small stones (Hynes, 2004). This resulted in significant costs to change requirements on several in-service contracts, and it was expected to increase bid prices for those mixes corresponding to the higher quality aggregate needed to meet friction requirements. Other points made by state representatives were related to the bid prices and construction considerations of HMA mixes having different additives (e.g., recycled glass, recycled rubber, slag, fly ash) and aggregates with different polish/wear characteristics.

D-15 Concrete Surfacing Techniques and Restoration Treatments The economic considerations for concrete surfacings and treatments mostly center around the surface texturing (burlap or turf dragging, brooming, tining, exposed aggregate texturing, grooving, grinding, abrading, plastic brushing) or surface dressing (chip sprinkling), but also the properties of the aggregate and cement paste. Again, the economic consequence of using alternative methods/materials requires the acquisition and analysis of up-to-date unit bid prices and pertinent forms of time-series performance data (e.g., distress, ride, friction, noise). Information concerning construction time requirements and other construction-related issues is also important. As with asphalt surfacing mixes and restoration treatments, the compiled literature contained several documents focusing on the performance characteristics of different concrete texturing techniques and restoration treatments. Only a few of them included specific assessments of costs tied in with friction and friction-related performance. Studies by the Michigan DOT and FHWA (Buch et al., 2000), the Colorado DOT (Ardani and Outcalt, 2000), and the Oregon DOT (Hunt, 1999) looked at the performance and costs of different concrete texturing techniques (in Michigan, exposed aggregate texturing and transverse tining; in Colorado, uniform transverse tining [with and without Astroturf drag] and various longitudinal tining [with Astroturf drag]; in Oregon, millabrading and diamond grinding). The most pertinent concrete study is an on-going, FHWA-funded study initiated in 1998 and to be completed in 2005. In this study, two different forms of texturing—randomly spaced transverse tining and longitudinal diamond grinding—were implemented on a new PCC pavement on I-190 in Buffalo, New York (Burge, Travis, and Rado, 2001). Results of friction and macro-texture testing and analysis after 2 years, indicated somewhat higher levels of friction for the longitudinally ground surface as compared to the transverse tined surface, but a generally greater loss of macro-texture. In terms of noise, the longitudinally ground surface was shown to be 2 to 5 dBA quieter than the transverse tined surface. Although the construction cost and time associated with diamond grinding was found to be higher than transverse tining, it was expected that these costs would be partly offset by an extended service life. Interviews with selected state agencies did not reveal any specific studies involving the economic impacts of PCC friction design and management. However, a few of the states were very interested in the latest performance and cost-effectiveness information concerning different tining techniques (transverse or longitudinal, uniformly- or randomly- spaced), as well as other initial texturing options, such as the Astroturf drag used by Minnesota. As a case in point, the Michigan DOT noted of a recent project on I-275 near Detroit where the randomly spaced tines were not completed as required, resulting in a noisy pavement (Hynes, 2004). In response, the Department had the surface diamond ground, only to observe low friction numbers a few years later.

D-16 Closing As discussed previously, several of the documented studies found in the compiled literature involved assessments of friction, texture, and/or noise performance. Although costs were not examined directly in most of these studies, the implications of low friction and inadequate drainage (increased crashes) and of high noise (need for noise mitigation) from which costs can be derived were often discussed. Hence, the value of these documents in further defining the economic impacts of friction and friction-related properties is well recognized. PAVEMENT–TIRE NOISE Pavement surface texture is generally accepted as one of the major contributors to pavement–tire noise; however, the exact role of texture is not completely understood. Two types of tire noise measurements have been made to quantify the effects of texture: near- field and far-field measurements. Far-field noise is more relevant than near-field noise because it is a measure of community noise impact. Commonly, efforts are directed toward control of far-field noise through the use of barriers rather than through the mitigation of noise at the source. However, near-field noise is less difficult to measure or predict. Therefore, near-field noise will be a useful measure of tire/pavement noise when correlated to far-field noise. Efforts to achieve this correlation are underway. However, the effects of pavement surface texture on far-field pavement–tire noise are extremely difficult to predict. Actual noise can vary as a result of the reflecting surfaces at the test site, wind velocities, ground absorption, attenuation of the sound by foliage, and thermal gradients. Primary pavement-tire noise research activities are currently focused in Arizona, California, and Indiana. In April 2003, Arizona DOT received pilot status with FHWA to allow pavement surface type as an accepted noise mitigation strategy (Scofield 2003a). This status permits ADOT to receive a 4 dBA credit in pavement noise design for using asphalt rubber friction course (ARFC) materials. Final allowance will be considered at the conclusion of a 10-year, $1 million ADOT noise research effort that includes evaluating both AC and PCC pavements. PCC pavement textures being evaluated include standard 1-in (25-mm) transverse tining, random transverse tining, 1-in (25-mm) longitudinal tining, and several diamond grinding methods. AC pavement types include ARFC, permeable European mixture (PEM), stone matrix asphalt (SMA), neat-asphalt friction course, polymer modified friction course, and terminal blend asphalt friction course (Scofield, 2003b). Several pavement-tire and vehicle noise measurement methods are being used by ADOT. These include using a near-field Close Proximity (CPX) trailer designed by the NCAT and modified to include both sound pressure and noise intensity probes. Measurements have also been collected using a separate vehicle equipped with unshielded noise intensity probes. Far-field Statistical Pass-By noise measurements are also being collected regularly along with significant environmental data. Some Pass-by testing is being conducted on a PCC test site. Vehicle speed and measurement offset are also being evaluated. Most of the evaluations are scheduled to continue for at least 10 years.

D-17 Among the objectives of the research include validating the 4dBA reduction allowance for ARFC, quantifying the acoustic properties of ARFC over time, defining a correlation between near field and far field noise measurements, evaluating pavement material properties for acoustical performance, develop test procedures for evaluating the noise potential of AC mixtures, develop procedures for noise construction quality control, and others (Scofield, 2003a). CALTRANS has also been conducting pavement and vehicle noise research, and working closely with ADOT, has received similar pilot status. They are reportedly analyzing open graded asphalt friction course material as their primary noise reduction surface (APA, 2003). A side finding reported from this research is that pavement-tire noise on PCC pavements can be affected by the joint reservoir width and overfilling the joints with sealant (Roberts, 2004). The ISQDH at Purdue University has been conducting noise and materials analysis using their drum facility, laboratory, and other sound measuring systems. Their recent research has included analysis of the effect of pavement texture on pavement-tire noise (Bernhard et al., 2003), development of quiet and durable porous PCC materials (Olek et al., 2003), and developing porous AC mixes for noise control application (McDaniel et al., 2003). Reports on development of porous modified asphalt mixes for noise control applications and relating surface texture of rigid pavement with noise and skid resistance are expected to be released in summer 2004. An FHWA quiet pavement European scanning tour is also scheduled for the summer of 2004. SUGGESTED IMPROVEMENTS TO FRICTION PRACTICES AND DESIRED AREAS OF FRICTION GUIDANCE Agency personnel face differing challenges in designing and managing their pavements for friction, safety, and noise. Economic and other factors can limit their ability to adequately resolve these challenges. Critical to that process, however, is identifying the weaknesses and problems in a process. When agency personnel were asked about changes they would like to see implemented in their friction design and management programs, their answers were typically related to incorporating friction management into pavement management systems, developing friction specifications for new construction, and defining relationships between aggregate sources, properties, and performance. Among their desired changes were the following (Brady, 2004; Geary, 2004; Hynes, 2004; McGhee, 2004; Pierce, 2004; Rasoulian, 2004; Skerritt, 2004; Stampley, 2004; Vacura, 2004): • Network testing and collection focused on critical information. • Evaluate and implement construction quality assurance (QA), if deemed necessary. • Modify and update the pavement management system database to provide pavement section history, aggregate types, aggregate properties, friction history, and crash history. • Use wet weather crash data more effectively. • Incorporate friction data into the DOT pavement management database. • Break down friction data by construction project and integrate it into the PMS system. • Make friction testing part of the pavement management system.

D-18 • Develop performance-related specifications for SN40R on new construction projects. • Define the relationship between quarry sources, aggregate properties, as-built SN40R, and long-term SN40R. • Develop and implement method of testing for qualifying aggregate according to polishing resistance. • Incorporate texture (and IFI) into friction management/design/etc. • Develop standards for classifying aggregate based on performance and the need for dedicated funding for safety improvements. • Develop of texture-based measurement. • Identify aggregate sources that work (matching aggregate with performance). • Identify particular designs (AC mixes, PCC textures) that work. Based on these and other perceived needs, interviewed agency personnel also provided listings of what specific guidance they would like to see in the new Guide for Pavement Friction. Their responses indicated a particular desire for the Guide to include background information and assistance managing, designing, and constructing highway pavements with good long-term frictional properties. Specific items requested included the following: • Provide background information on friction management, design, construction, legal, and economic issues. • Provide performance information on aggregate types. • Provide information about methods for designing pavement to ensure friction. • Provide information about management methods and alternative frameworks, advantages and disadvantages. • Provide specific guidance on the viability and design limits (related to friction) of such additives as recycled glass, recycled rubber, slag, and fly ash. • Include specific guidance on aggregate analysis (including petrographic examination) and a better understanding of aggregate properties and methods for analyzing these properties. • Provide more information on the effect of asphalt modifiers on friction. • Document what other states are doing to address pavement frictional needs (i.e., how they select mix designs, crash location correction, etc.). • Provide guidance in liability issues. • Provide an understanding of texture (micro and macro), and its effect on friction, noise, and other factors. • Clarify the effect of anti-lock brake systems on friction measurement and design. • Provide good information about seasonal friction variability, long-term friction trends of different aggregates, temperature variation effects, and speed variation effects. • Provide guidance on how the impending peak friction number can be used. • Provide a concise pavement friction handbook that clearly describes related issues for engineers and aggregate producers. People need a big-picture understanding of what’s happening in aggregate wear, data collection, and testing. It should be written to include word pictures to allow the readers to visualize the concepts (e.g., the energy expended in hundreds of thousands of cars braking on a pavement surface). • Provide methods for surgically remediating critical sections with fast, high-quality fixes that will last 8 to 15 years (e.g., microsurfacing with expanded clays or other materials).

D-19 • Provide methods for defining the worst frictional demand conditions not particularly based on volume (e.g., energy input). • Provide detailed guidance or a case studies synthesis (what’s been done, what worked) on matching aggregates with frictional performance, matching AC mixes and PCC textures with frictional performance, and resolving PCC noise and skid problems. • Provide assistance with how to design mixes with sensitivity to friction, noise, splash and spray. • Provide tort liability protection methods. • Quantifying the benefits of friction management and design. • Provide guidance in how to transition from ribbed to the smooth tire as well as the IFI (micro-texture and macro-texture). The Guide should provide some flexibility, not hard-and-fast numbers. On the other hand, the Guide should not be so flexible as to allow every individual practice, as is currently the case with the AASHTO Guide on pavement smoothness testing. REFERENCES Ardani, A. and W. Outcalt. 2000. PCCP Texturing Methods, Final Report, Report No. CDOT-DTD-R-00-1, Colorado Department of Transportation Research Branch, Denver, Colorado. Asphalt Pavement Alliance (APA). 2003. “Big Noise over Noise Reduction,” Alliance in Action newsletter, March 10, 2003, website: http://www.asphaltalliance.com/ upload/e- Alliance%20in%20Action%20March%202003.pdf), Lexington, Kentucky. Beck, W and F. Holt. 2004. Notes from Feb 17, 2004 teleconference with Bill Beck and Frank Holt of Dynatest Consulting, Inc. Bernard, R.J., W.D. Thorton, and J Bauman. 2003. “The Effects of Varying the Tire Cap Ply, Sidewall Filler Height, and Pavement Surface Texture on Tire/Pavement Noise Generation,” Final Report, SQDH 2003-1, HL 2003-1, The Institute for Safe, Quiet and Durable Highways (ISQDH), Purdue University, West Lafayette, Indiana. Brady, P. 2004. Notes from February 19, 2004 teleconference with Mr. Patrick Brady, Florida Department Transportation State Traffic Safety Engineer. Brakey, B.A. 1973. “Design, Construction, and Performance of Plant Mix Seals.” FHWA Notice, Washington, D.C. Buch, N., R. Lyles, and L. Becker. 2000. “Cost Effectiveness of European Demonstration Project: I-75 Detroit,” Report No. RC-1381, Michigan Department of Transportation, Lansing, Michigan. Burge, P.L., K. Travis, and Z. Rado. 2001. “A Comparison of Transverse Tined and Longitudinally Diamond Ground Pavement Texturing for Newly Constructed Concrete

D-20 Pavement,” Paper prepared for the 2002 Annual Meeting of the Transportation Research Board, Washington, D.C. Buys, R. 2004. Notes from March 11, 2004 teleconference with Romain Buys, General Manager with N.V. Robuco, Ltd in Buggenhout, Belgium. Chelliah, T.C., P. Stephanos, M.G. Shah, and T. Smith. 2003. “Developing a Design Policy to Improve Pavement Surface Characteristics,” Paper presented at 2003 Annual Meeting of the Transportation Research Board, Washington, D.C. Choubane, B., G.A. Sholar, J.A. Musselman, and G.C. Page. 1999. “A Ten-Year Performance Evaluation of Asphalt Rubber Surface Mixes,” Paper presented at the 78th Annual Meeting of the Transportation Research Board, Washington, D.C. Cram, I.A. 1975. “Asphalt Surface Treatments,” Highway and Road Construction, Vol. 43 No. 1782/3, pp. 18-20. Federal Highway Administration (FHWA). 1973. “Open Graded Plant Mix Seals” FHWA Pavement Design Branch, FHWA Notice, Washington, D.C. Galloway, B.M. et al. 1975. “Tentative Pavement and Geometric Design Criteria for Minimizing Hydroplaning,” Final Report, FHWA, Report No. FHWA-RD-75-11, Texas Transportation Institute, College Station, Texas. Geary, G. 2004. Notes from February 17, 2004 teleconference with Ms. Georgene Geary, State Materials and Research Engineer, Georgia Department of Transportation. Gomaco. 2004. “The World's Quietest Paver Slipforms A Quiet New Roadway,” Gomaco World 32.1, February 2004, Website: http://www.gomaco.com/Resources/ worldstories/world32_1/betonac.html, Ida Grove, Iowa. Hunt, E.A. 1999. “Millabrading Test Evaluation,” Report No. FHWA-OR-RD-00-02, Oregon Department of Transportation, Salem, Oregon. Hunt, E.A. 2002. “Crumb Rubber Modified Asphalt Concrete in Oregon,” Report No. FHWA-OR-RD-02-13, Oregon Department of Transportation, Salem, Oregon. Hynes, T. 2004. Notes from February 19, 2004 teleconference with Mr. Tom Hynes, Pavement Evaluation Engineer with Michigan Department of Transportation. International Symposium on Porous Asphalt (ISPA). 1977. Proceedings. SCW Record 2. Study for Road Construction, Amsterdam. Lyon, B. 2004. Notes from Feb 16, 2004 teleconference with Bill Lyon of the Transportation Research Center. McDaniel, R.S. and B.J. Coree. 2004. “Identification of Laboratory Techniques to Optimize Superpave HMA Surface Friction Characteristics,” Phase I, Final Report, SQDH 2003-6,

D-21 HL 2003-19, Institute for Safe, Quiet, and Durable Highways (ISQDH), Purdue University, West Lafayette, Indiana. McGhee, K. 2004. Notes from February 18, 2004 teleconference with Mr. Kevin McGhee of the Virginia Transportation Research Council. Metcalf, S. 2004. Notes from February 16, 2004 teleconference with Scott Metcalf, technical marketing engineer with Koch Pavement Solutions in Fontana, California. National Cooperative Highway Research Program (NCHRP). 1974. “Wear-Resistance and Skid Resistant Highway Pavement Surfaces.” NCHRP Research Results Digest 61. Highway Research Board, Washington, D.C. National Cooperative Highway Research Program (NCHRP). 1978. “Open-Graded Friction Courses for Highways.” NCHRP Synthesis of Highway Practice 49, Transportation Research Board, Washington, D.C. Nippo Sangyo. 2003. Nippo Sangyo Ltd. website, http://www.nippou.com/product3e.html, Tokyo, Japan. Olek, J, W.J. Weiss, and R. Garcia-Villarreal. 2004. “Relating Surface Texture of Rigid Pavement with Noise and Skid Resistance,” Final Report, SQDH 2004-1, HL 2004-1, The Institute for Safe, Quiet and Durable Highways (ISQDH), Purdue University, West Lafayette, Indiana. Olinoski, R and B. Olinoski. 2004. Notes from Feb 18, 2004 teleconference with Rob and Bob Olinoski of International Cybernetics Corporation. Pierce, L. 2004. Notes from February 18, 2004 teleconference with Ms. Linda Pierce, Pavement Design Engineer with Washington Department of Transportation. Pilkington, G.B. 1982. “Reduction of Truck-Induced Splash and Spay.” Public Roads, Volume 46, pp. 6-12, Washington, D.C. Public Roads. 2003. “Alternative Pavement Surfaces May Reduce Roadway Noise,” Public Roads, May/June, Federal Highway Administration, Washington, D.C. Rasoulian, M. 2004. Notes from February 19, 2004 teleconference with Mr. Masood Rasoulian, Louisiana Transportation Research Center. Ritter, T.E. 1974. “The Development of Techniques to Measure Vehicle Spray on Wet Roads,” SAE Transactions, Vol. 83, Paper No. 74026, Washington, D.C. Rowdan, L. 2004. Notes from February 19, 2004 teleconference with Ms. LaDonna Rowdan, Illinois Department of Transportation. Roberts, J. 2004. Notes from March 1, 2004 teleconference with John Roberts, Executive Vice President of the International Grooving and Grinding Association in New York.

D-22 Sabey, B.E. 1973. “Accidents: Their Cost and Relation to Surface Characteristics.” Symposium on Safety and the Concrete Road Surface-Design, Specification and Construction, Cement and Concrete Association, Birmingham, United Kingdom. Scofield, L. and P.R. Donavan. 2003. “Development of Arizona’s Quiet Pavement Research Program,” Arizona Department of Transportation, Phoenix, Arizona. Scofield, L. 2003. “SR202 PCCP Whisper Grinding Test Sections – Construction Report,” Arizona Department of Transportation, Phoenix, Arizona. Skerritt, W.H. 2004. Notes from February 19, 2004 teleconference with Mr. Bill Skerritt of the New York Department of Transportation, Engineering Geology section of the Materials Bureau. Sorensen, K. and B. Nielsen. 1974. “Road Surfaces in Traffic Lighting.” Danish Illuminating and Engineering Laboratory, Ministry of Public Works, Lyugby, Denmark. Stampley, B. 2004. Notes from February 13, 2004 teleconference with Mr. Bryan Stampley, Pavement Management Engineer with Texas Department of Transportation. Texas Department of Transportation (TXDOT). 2004. Concrete Rated Source Quality Catalog and Bituminous Rated Source Quality Catalog, Material Producer List, Texas DOT, http://www.dot.state.tx.us/business/materialproducerlist.htm, Austin, Texas. United Stated Department of Transportation. 1980. Skid Accident Reduction Program, Technical Advisory T 5040.17, Federal Highway Administration, Washington, D.C. Vacura, P. 2004. Notes from February 19, 2004 teleconference with Mr. Peter Vacura, California Department of Transportation. Western Highways Institute (WHI). 1973. “Splash and Spray Characteristics of Trucks and Truck Combinations,” Research Committee Report No. 5. Wambold, J.C., Henry, J.J., and R.R. Blackburn. 1984. “Pavement Surface Texture: Significance and Measurement.” FHWA contract DOT-FH-11-9091, PTI Report PTI 8312, College Station, Pennsylvania. Yeakel, S. 2004. Email correspondence on March 1, 2004 from Skip Yeakel, Principal Engineer at Volvo Trucks. Zimmer, D. 2004. Notes from Feb 17, 2004 teleconference with Dick Zimmer of the Texas Transportation Institute.