Analysis and Predictive Modeling of Road Sign Retroreflectivity Performance

The Louisiana Department of Transportation and Development (DOTD) has over 400,000 traffic signs in its inventory. These signs vary widely by their sheeting grade, age, color, orientation, and proximity to the roadway edge. In a typical year the DOTD replaces about 60,000 of these signs because they are beyond their service life or have been damaged. Currently, the DOTD does not have a systematic sign maintenance program. As a result, many warranteed signs that are not in compliance with its performance requirements go un-replaced. This study was conducted to analyze the performance and deterioration characteristics of retroreflective sheeting materials with the goal of targeting DOTD compliance testing efforts to sign groups that may likely be prematurely degraded. In addition to this, three other objectives were accomplished in the study including, an analysis of the interaction between various sign properties and environmental factors to determine the factors that affect the rate of traffic sign deterioration, an evaluation of cleaning on sign retroreflectivity, and the development of mathematical models to predict future sign performance. The analyses showed that, overall, the various sign sheeting was performing well relative to the specification criteria during and after the warrantee period. The analyses also showed no statistically significant links between key environmental factors such as proximity to the road or sign orientation to a pre-mature deterioration. However, the analyses results did support the concept that cleaning can significantly improve specification compliance. A set of performance models was also developed to assess the rates of sign deterioration for specific conditions to assist highway agencies in focusing maintenance resources on specific groups of signs. WORD COUNT: 4,404 words + 6 figures and tables (@250 word equivalents ea.) = 5,904 word equivalents INTRODUCTION Like all traffic control devices, traffic signs deteriorate over time. This deterioration is due to numerous environmental factors including natural elements (sun exposure, dirt, ice, wind, etc.), manmade contaminants (exhaust emissions and industrial pollution), and impacts from vehicles. A typical life-span for a traffic sign is about 10 years. After this time signs tend to lose their color tones and retroreflective properties, making them difficult to read, especially at night. The Louisiana Department of Transportation and Development (DOTD) has over 400,000 traffic signs in its statewide inventory. These signs vary widely in their age, color, orientation, and proximity to the roadway edge. Like all highway agencies the DOTD must replace a substantial portion of these signs each year. In a typical year sign replacements number about 60,000. Due to limited personnel and maintenance resources, these replacements nearly always occur as a result of collision damage. The DOTD does not have the manpower nor financial resources to undertake an annual, systematic, or comprehensive sign maintenance program. As a result, many warranteed signs that are not in compliance with the minimum DOTD performance requirements go un-replaced because of the impracticality of checking all signs every year. Wolshon, Degeyter, and Swargam Page 2 16 Biennial Symposium on Visibility and Simulation, June 2 4, 2002, Iowa City, IA, USA To assess the compliance of traffic signs to the performance requirements specified by the DOTD and to analyze the characteristics of traffic sign deterioration, a field study was undertaken to evaluate the performance of colored sign sheeting materials in the DOTD’s field inventory. The study incorporated a three-step approach of field data collection, data analysis, and predictive model formation. Data collection focused on factors thought to be critical to the performance of traffic signs. Various statistical analyses were used to identify and quantify the contribution of these key factors, both individually and in combination with others. Finally, mathematical models were developed to predict performance based on combinations of sign age, color, orientation, location, and type of sign sheeting material. Thus, the procedures and models presented in this investigation can be used to predict sign performance under a variety of field conditions to more effectively allocate sign maintenance resources. RETROREFLECTIVE SHEETING SPECIFICATIONS Retroreflective sheeting is generally classified into one of eight categories defined by the American Society for Testing and Materials (ASTM) (1). The cost, performance, and material characteristics of the sheeting material in each of these categories vary significantly. The Louisiana DOTD uses three different ASTM sheeting types for its signs. Engineering Grade, or Type I, sheeting makes up about 70 percent of its inventory. Type I is classified by ASTM as a medium-intensity retroreflective sheeting and is typically constructed using an enclosed sheet of glass bead lenses. Nearly all of the remaining 30 percent is made up of High-Intensity Grade, or Type III, sheeting. Type III is much more retroreflective material typically made up of an encapsulated glass bead retroreflective or prismatic material. The DOTD has also recently started to use Diamond Grade (Type VIII) sheeting in new construction projects. However, very few signs are in use, and no Type VIII sheeting data was collected as part of this investigation Performance requirements for traffic sign sheeting in Louisiana are specified in the DOTD’s manual of specifications for the construction of roads and bridges (2). These specifications are based on ASTM criteria and use age and coefficient of retroreflection to define the warrantee criteria. A summary of the minimum performance requirements for Type I and Type III retroreflective sheeting is presented in Tables 1 & 2. Table 1. Type I Sheeting B Observation Angle Entrance Angle White Yellow Orange Green Red Blue Brown 0.2° -4.0° 70 50 25 9.0 14 4.0 1.0 0.2° +30.0° 30 22 7.0 3.5 6.0 1.7 0.3 0.5° -4.0° 30 25 13 4.5 7.5 2.0 0.3 0.5° +30.0° 15 13 4.0 2.2 3.0 0.8 0.2 A Table source: (2), B Minimum Coefficient of Retroreflection (RA) cd/fc/ft (cd/lx/m) Table 2. Type III Sheeting B Observation Angle Entrance Angle White Yellow Orange Green Red Blue Brown 0.1° -4.0° 300 200 120 54 54 24 14 0.1° +30.0° 180 120 72 32 32 14 10 0.2° -4.0° 250 170 100 45 45 20 12 0.2° +30.0° 150 100 60 25 25 11 8.5 0.5° -4.0° 95 62 30 15 15 7.5 5.0 0.5° +30.0° 65 45 25 10 10 5.0 3.5 A Table source: (2), B Minimum Coefficient of Retroreflection (RA) cd/fc/ft (cd/lx/m) C Values for 0.1° observation angle are supplementary requirements that shall apply only when specified by the purchaser in the contract order. These tables illustrate the minimum criteria for each sheeting color. The coefficients of retroreflection in these tables are given for several different entrance and observation angle values. These entrance and observation angles approximate the angles that exist in a highway environment under common driving conditions. It can be seen that the expected level of performance for lighter colored signs, like white and yellow, is higher than those of darker Wolshon, Degeyter, and Swargam Page 3 16 Biennial Symposium on Visibility and Simulation, June 2 4, 2002, Iowa City, IA, USA colored signs, like blue and brown. These values are consistent with the rates of light reflection and absorption for these colors. The DOTD specification values are based on the installation of new sheeting and indicate a minimum retroreflection measurement during a minimum of seven year performance period for Type I sheeting and for ten years for Type III sheeting. The performance periods are considerably shorter for orange signs, three years and five years respectively because construction zones signs are subjected to more abuse and adverse environmental conditions around construction work areas. The DOTD values differ slightly from the published ASTM and significantly from the end-of-service life criteria recommended by the Federal Highway Administrtaion (FHWA). FHWA end-of-service life criteria are used to define the recommended service limits for retrorflective sheeting (3). For example, FHWA end-of-service minimum retroreflectivity values for Type III sheeting are between 30-70 cd/lx/m for white and 3055 cd/lx/m for yellow sheeting. OBJECTIVES The overall goal of the study was to help maintenance personnel to develop sign testing, maintenance, and replacement schedules. To achieve this goal the following four objectives were set to quantify the rate of specification compliance and determine the characteristics of retroreflective sign sheeting deterioration: 1. Evaluate the compliance of traffic signs to the DOTD performance specifications 2. Determine the effect of sign cleaning on the retroreflective properties of signs 3. Analyze the interaction between various sign property and environmental factors to the determine factors that influence the rate of traffic sign deterioration 4. Develop mathematical models to predict future sign performance based on sign properties and field conditions. DATA COLLECTION Sign data were collected from traffic signs placed along Interstate and State Highway routes throughout the State of Louisiana. A Model 920 Field Retroreflectometer was used to make measurements of sign retroreflectivity. The Model 920 has a fixed measurement geometry of a -4° entrance angle and a 0.2° observation angle with a measurement diameter of approximately 1 inch. The retroreflectometer produces a digital display of the coefficient of retroreflection for each sign in candelas/foot candle/square foot (4). Although the DOTD uses eight colors of background sheeting (yellow, red, blue, green, brown, orange, black, and white) on its signs. However, only white, green, and yellow signs were used in the analyses. All other colors were excluded for various reasons including manufacturing irregularities and a scarcity of signs. Orange signs were not used because they are erected primarily for temporary construction and are not consistently exposed uniform field conditions. Black signs were not included because they are not retroreflective. The sample of 237 signs was evenly distributed between the two sheeting grades and three colors. The ages of signs in the sample varied widely, from nearly new to over 20 years old. Sign orientation was recorded from azimuth angle measurements to determine the direction the sign was facing. Measurements of sign height (from the ground) and the lateral distance from the curb or travel lane were also taken. Retroreflectivity readings for each sign were recorded under existing (unwiped) and cleaned (wiped) conditions. The existing retroreflectivity measurement was taken as the sign was found in the field. The cleaned reading was taken after the test area was wiped free of dirt and grime with soap and water. The wiped reading was taken for comparitive purposes, to determine the performance differences between clean and dirty signs. It was also used to record the "true" retroreflectivity, rather than one that was limited by dirt on the sign face. SPECIFCATION COMPLIANCE At times the DOTD has found itself in the position of needing to replace signs just out of warrantee. With advanced knowledge some of these signs could have been replaced at little or no cost to the Department. While it was impractical to expect maintenance crews to check every sign in the inventory, the DOTD felt that sign performance statistics could allow them to focus their compliance testing efforts. Thus, the first step in the study was to determine how signs were performing with respect to the specifications, both in terms of sheeting grade and color. Wolshon, Degeyter, and Swargam Page 4 16 Biennial Symposium on Visibility and Simulation, June 2 4, 2002, Iowa City, IA, USA As shown in Tables 1 and 2, the DOTD performance requirements for traffic signs differed on the basis of sheeting type and sign color. To distinguish between the various grades and colors, the field data was categorized into six separate subsets based on combinations of color and grade. Figures 1 and 2 illustrate the distribution of the “wiped” Type I and Type III sign retroreflectivity measurements as a function of sign age, with respect to the performance specification criteria. Figure 1. Sign Compliance Distribution – Type I Sheeting Figures 1 and 2 present the compliance data as four-quadrant maps. All data points in quadrant III (the upper left) include the signs that were meeting the specification requirements and were within the warrantee period. Signs in quadrant II (the lower right) do not meet the performance specification requirements, but were also out of warrantee. The signs in quadrant I (upper right) were the best performers. These were signs that were out of warrantee but continued to meet the performance specification criteria. The area of greatest interest to the DOTD was quadrant III (the lower left). Signs in this region were not meeting the performance specification while under warrantee. Thus, they could be replaced at no charge to the DOTD. Generally speaking, the data showed that the traffic signs in the sample were performing well. Of all signs covered under warrantee, about 92 percent were performing at or above the minimum specification criteria. Of the signs that were out of warrantee, about 43 percent continued to perform at or above the minimum levels of retroreflectivity. This was positive news to the DOTD. Using these initial results, further analyses were conducted to determine if any of the six sign groups was performing better or worse than others. This was important since it would focus compliance testing efforts on sign groups that showed tendencies toward rapid deterioration. The data presented in Figures 1 and 2 has been summarized in Table 3. Here the signs are categorized by color and sheeting type to show the number and percentage of signs in each group that were in compliance with the DOTD specification criteria before and after the warrantee period. The table showed that both sheeting grades performed well during the warrantee period. Of the 149 signs under warrantee in the sample, only 12 (8.1 percent) failed to meet the performance specification. 0 100 200 300 0 20 40 60 80 100 120 Age (months) C oe ffi ci en t o f R et ro fle ct iv ity (f t-c nd l/f t)