Sensitivity Analysis of Sign Luminance as Function of Input Factors

Abstract

The luminance, or brightness, produced by a traffic sign at night is a function of several different factors including the performance of the retroreflective sheeting, the light output of the headlamp, the geometry between the headlamps and driver, and the position of the sign relative to the vehicle. In this study, the author calculated luminance for 27 different combinations of the factors affecting sign luminance. The result is eight figures comparing the sign luminance for the analyzed conditions to the luminance required for a dark rural or suburban roadway environment. The results indicated that there can be a wide range in luminance performance depending on the input variables, many of which are beyond the control of transportation agencies. The final comparison found that the difference between the best- and worst-case combinations of input variables resulted in a luminance difference of over 2,000%. Such a wide range of performance measurement means that traffic sign sheeting that is selected based on performance for a specific set of circumstances may perform in a vastly different manner because of other factors such as the type of vehicle, the headlamps on the vehicle, the roadway geometry, or the position of a sign relative to the roadway and vehicle.

Keywords

operations traffic control devices retroreflectivity sign sheeting signs traffic signs

Traffic signs use retroreflective sheeting to make them more visible to drivers at night. Our transportation system would not function in a safe and efficient manner in the dark without retroreflective sheeting. Over the last century, sign retroreflectivity has advanced more or less in the following order:

Individual retroreflective elements in the sign (cat’s eyes);

Beads sprinkled on paint;

Exposed bead retroreflective sheeting;

Enclosed bead retroreflective sheeting;

Encapsulated bead retroreflective sheeting; and

Microprismatic retroreflective sheeting.

Today, transportation agencies can choose from a wide range of retroreflective sheeting products in specifications for traffic sign fabrication. Knowing which product to select can be challenging, as the luminance provided by a specific product varies depending on the factors that affect sign luminance. If all of the following factors are known, it is a straightforward process to calculate the sign luminance:

The vehicle geometry between the headlamps and driver,

The roadway geometry between the sign and vehicle,

The luminous intensity of the headlamp at the specific vector between the headlamp and sign centroid, and

The retroreflectivity characteristics of the sign sheeting for the specific geometry between the sign and vehicle.

Although straightforward, sign luminance is rarely calculated owing to the challenges associated with the photometric characteristics of vehicle headlamps and retroreflective sign sheeting. Both of these require extensive measurements in a photometric laboratory to provide the ability to calculate sign luminance. Such photometric measurements are not readily available in the public domain, particularly for sign sheeting products. As a result, no body of research has evaluated the impact of a variety of factors on the luminance performance of sign sheeting. This paper uses some of the limited amount of information to indicate the range of sign luminance that results from different input factors.

Background

To fully evaluate sign luminance, two different luminance values are needed. The first is the luminance supplied by the sign, which is a function of the vehicle, headlamp, sign sheeting, and relational geometry. The supply luminance alone has limited value as the user does not know whether the calculated/measured luminance is adequate to meet driver needs. The luminance needed is referred to as the required luminance or demand luminance and is typically determined through research efforts.

Supply Luminance

Equations 1 and 2 are the basic equations needed to calculate sign luminance once the relational geometric values (measurement geometry) are known. The measurement geometry includes the vehicle geometry (headlight height, headlight separation, setback distance from headlamps to driver, driver eye height, and driver eye offset from center of vehicle) and the roadway geometry (distance between vehicle headlights and the sign, sign height relative to the vehicle, and sign offset relative to the vehicle). The geometry defines the observation axis, illumination axis, retroreflector axis, datum axis, angles between the various axes, and distances along selected axes,

L = (R_{A, left} \times E_{⊥, left} + R_{A, right} \times E_{⊥, right}) / \cos v

(1)

E = I / d^{2}

(2)

where

L = sign luminance (cd/m²),

R_A = coefficient of retroreflection for the left or right headlamp as indicated (cd/lx/m²) at the applicable viewing geometry,

E_⊥ = illuminance from left or right headlamp as indicated (lx),

v = viewing angle (angle between observation axis and retroreflector axis) using driver as observation point (deg),

I = luminous intensity of headlamp at specific vector (lux), and

D = distance between headlamp and sign (m).

Fortunately, tools are available that simplify calculating sign luminance. One of these tools is a computer program called Exact Roadway Geometry Output (ERGO) ( 1 ). ERGO was originally developed by Stimsonite and was supported by Avery Dennison for several years after they bought Stimsonite. ERGO provides a means of determining the measurement geometry and calculating sign luminance based on the following input factors:

Vehicle dimensions: headlamp height and separation, driver eye height and position relative to headlamps;

Headlamp performance: an array of headlamp intensities for a range of horizontal and vertical angles;

Sign position: sign offset and height relative to the right edge of the road plus twist, lean, sign rotation, and sheeting rotation;

Sign sheeting: an array of retroreflectivity values for a specific sheeting product for a range of observation, entrance, rotation, and orientation angles. (If using the application system of angles, the user can also use the intrinsic system or International Commission on Illumination [CIE] goniometer system.); and

Roadway geometry: the lane width, road alignment (straight or curved with radius value), and distance between sign and vehicle.

The specific values used for the analysis in this paper these factors are described in more detail later in this paper. The 2001 release of ERGO included sign retroreflectivity files for several sheeting products, a selection of headlamp profiles, and dimensions for a few vehicles. All of these represented sheeting products, headlamps, or vehicles that existed near the turn of the century. Avery Dennison no longer supports ERGO and it is no longer available for download. Possible reasons for dropping support of ERGO include, but are not limited to, the following:

Proper use of the scientific principles for luminance calculations requires a level of understanding and expertise that is not a part of a traditional traffic engineering background, thus creating the potential for inappropriate or inaccurate use of the program.

The program would require continuous updating to provide and maintain retroreflectivity, headlamp, and vehicle files for an ever-increasing number of sign sheeting products and vehicle models.

The potential for the product to be used in a manner other than intended to cast certain retroreflective sheeting products in a negative manner.

Required Luminance

Determining the luminance required for a specific sign in a specific location is a complicated process. Published research indicates significant levels of variation in the luminance needs for various signs, environments, and driving populations. An unpublished report by CIE includes a chapter on determining the luminance requirements for traffic signs ( 2 ), which the author used as the source for the required luminance.

The luminance requirements are based on the sign viewing process generalized in Figure 1. As indicated in the figure, drivers observe signs in three actions. In the first, labeled LOOK1, the sign comes into view and is recognized as a sign by the driver. In the second action (LOOK2), the driver begins to recognize color and shape and assesses whether the sign is pertinent to the current driving task. The sign is actually legible during the third action, which is labeled LOOK3. This is when the driver obtains the specific information presented in the sign legend. The LOOK3 reading distance generally begins at the maximum legibility index (40 ft/in. was used for this study). This approach assumes that a driver stops reading a sign at about the minimum legibility index (20 ft/in. was used for this study). The distances shown associated with each action are generalized and not in proportion to each other.

Having defined the threshold legibility indexes for when a sign is read, the luminance requirements at this distance are needed. The CIE report deals with required luminance in detail. For a passenger car in a dark rural or suburban environment, the CIE report relied on a study by Carlson and Hawkins, with slight adjustments to the luminance values ( 3 ). The Carlson and Hawkins study determined luminance needs based on the legibility index for white legend signs on green, blue, and brown backgrounds using a sample of drivers that were 55 years or older. The resulting luminance requirements for various accommodation levels are shown in Table 1. The performance levels can loosely be interpreted as the percent of older drivers served by that level of luminance.

Table 1.

Required Luminance Levels for Older Drivers

Legibility index (ft/in.)	Required luminance (cd/m²) based on percent accommodation
	10%	25%	50%	75%	85%	95%
20	0.10	0.50	1.00	3.00	6.00	10.00
25	0.28	0.75	1.375	3.75	7.50	12.50
30	0.45	1.00	1.75	4.50	9.00	15.00
35	0.63	1.25	2.125	5.25	10.50	17.50
40	0.80	1.50	2.50	6.00	12.00	20.00

For use in the comparisons, the luminance levels for the 75th, 85th, and 95th levels were plotted in each figure based on a legend height of 6 in. This relates to the LOOK3 range starting at 240 ft from the sign and ending at 120 ft from the sign for a sign with a white legend or white background.

Methodology

The purpose of this study was to identify how changes in some key input factors can affect sign luminance. To achieve this, sign luminance was calculated using ERGO for a variety of input variables, as indicated in the following subsections. Luminance was calculated at distances ranging from 500 to 40 ft from the sign. To simplify the analysis, only one factor was changed for each comparison while the other factors remained at the base defined for this analysis. The analysis was focused on three input variables that can change in the field and that are beyond the control of the agency. These include the vehicle, headlamp, and driver eye height.

Most prior research on sign retroreflectivity has focused on luminance and retroreflectivity performance on straight roads owing to the complications of addressing sign luminance within a curve. To provide a sense of the variation in sign luminance in a curved roadway, the analysis included road geometry as a fourth input variable.

To keep the comparisons simple, only one sheeting product was used for almost all of the comparisons. To give a sense of the range in luminance that results from a different sheeting product, one analysis comparison compared two different sheeting products.

As a final comparison, the best-/worst-case scenario for each factor were used for comparison. The overall methodological approach resulted in six comparisons, with each being presented in a separate plot and presented in the Results section of the paper.

Vehicle

There is a wide range of vehicles on our nation’s highways and agencies have little control over what vehicles are allowed to use their roadways. Light trucks and sport utility vehicles (SUVs) of various sizes make up a significant proportion of the vehicle fleet. Furthermore, luminance levels for SUVs are generally lower than for an equivalent-sized sedan because of the larger observation angle that results from the driver sitting higher above the headlamps. Therefore, the author chose various SUVs as the representative vehicles, selecting three sizes of SUVs from a single manufacturer that were readily available to the author for measurement and that represented vehicle dimensions associated with a 2018 model year. The equivalency with the 2018 model year was important because of using headlamp profiles for 2018 model year vehicles (see Headlamp subsection). The following vehicles were selected for the analysis:

2018 Ford Expedition: This vehicle represents a large SUV and light truck class of vehicle. It also has the same headlight geometry as a Ford F-150, as both are built on the same frame. The 2018 and 2019 Expedition model year vehicles have the same headlight geometry. The Ford F-150 has been the best-selling vehicle in the United States for over 40 years.

2019 Ford Edge: This vehicle represents a mid-sized SUV class of vehicle. The 2019 model has the same headlamp geometry as the 2018 model. As the mid-sized vehicle of the three used in this analysis, the Edge was used as the base case vehicle in all of the comparisons.

2017 Ford Escape: This vehicle represents a smaller class of SUV vehicle. The 2017 model year Escape has the same body style and headlamp geometry as a 2018 model year Escape.

For each of these vehicles, the author measured five headlamp geometry dimensions, as indicated below, and created each vehicle in ERGO for use in the analysis. Table 2 indicates the vehicle dimensions used for the analysis.

Table 2.

Vehicle Dimensions (Inches)

Dimension	Expedition	Edge	Escape
Distance between headlights	65.25	55.25	57
Headlight height above road	40.5	36.75	34.063
Eye height above road	63.5	55	52.75
Eye setback from headlight	80.75	75.5	65.25
Eye distance left of vehicle centerline	18.25	14.75	14.5

Headlamp height: Height of the center of the low beam headlamp above the pavement;

Headlamp separation: Distance between the centers of the low beam headlamps;

Driver eye setback from headlamps: Horizontal distance from the headlamp to the driver’s eyes;

Driver eye height: Height of the driver’s eyes above the pavement. For this analysis, the eye height was measured for a driver that was 5-ft, 9-in. tall sitting in a normal driving position within the vehicle; and

Driver eye offset from vehicle center: Distance from the center of the vehicle to the center of the steering wheel. The analysis assumes that the driver head is aligned with the center of the steering wheel.

Vehicle Headlamps

Whereas the vehicle headlamp geometry is specific to individual models of vehicles, the information needed to calculate headlamp performance is a weighted average across multiple vehicles from a single model year. The headlamp data used in this analysis were obtained by the University of Michigan Transportation Research Institute (UMTRI) and represents photometric data, measured by automotive manufacturers, for 62 tungsten-halogen (TH) headlamps (in 31 left–right pairs) and 48 light-emitting diode (LED) headlamps (in 24 left–right pairs) ( 4 ). Collectively, the lamps are used on 45 vehicle models from three manufacturers, which accounted for 36% of U.S. light-vehicle sales in 2018. The report lists the 45 vehicle models that comprised the headlamp sample but does not indicate the basis for selecting the specific models included. The photometric information for each lamp was weighted by the sales figure for the corresponding vehicle. The photometric results are presented for the 25th-, 50th- (median), and 75th-percentile luminous intensities. The measurement ranges are ±25° horizontally and from 7° up to 5° down vertically.

The LED 50th percentile distribution headlamp served as the base case for all of the comparisons. Altogether, there were six different headlamp profiles used in the headlamp comparison (LED 25th percentile, LED 50th percentile, LED 75th percentile, TH 25th percentile, TH 50th percentile, and TH 75th percentile).

Headlamps Over Time

Headlamp design has changed over time. To provide a sense of performance differences from different headlamps over time, one of the comparisons varied only the headlamp, using the 50th percentile headlamp distributions from the 2018, 2010, and 2000 model years. Previous UMTRI headlamp profiles were published in 2011 ( 5 ) and 2001 ( 6 ).

Driver Eye Height

Drivers come in a wide range of sizes. Vehicles are designed to accommodate this range by allowing drivers to adjust several aspects of the vehicle’s interior dimensions, such as the distance between the seat and steering wheel, the distance between the driving pedals and driver, the height of the seat, and the angle of the steering wheel. The range in driver height combined with the range in interior adjustments creates the potential for a range of driver eye heights. This analysis used five different eye heights for comparison. The base driver eye height was determined in the vehicle measurements using a male driver that was 5-ft, 9-in. tall. This is the average height of a male in the United States ( 7 ). The same source indicates that the average height of a female in the United States is 5 ft, 3.5 in. Eye height was then increased and decreased by 3 in. and 6 in.

Sign Position

Almost all of the sign luminance comparisons were based on a post-mounted sign located on the right side of a two-lane highway with no shoulders. The sign centroid was 9 ft above the pavement and 14 ft to the right of the edge line. This represents a sign that is 4 ft wide and 4 ft tall. This is also roughly equivalent to a 36- × 36-in. diamond sign. Other sign placement factors were set at zero to simplify the analysis, including sign twist, -lean, -rotation, and sheeting rotation. It is worth noting that some agencies twist and/or lean signs to reduce the potential for specular reflection and the resulting glare to drivers. Such practices, and the actual adjustment level, can vary, making the selection of appropriate input factors challenging.

One comparison evaluated the impact of sign position on luminance. The comparison evaluated the sign positions listed below with all other input variables representing the base case for each variable (straight road, Edge, LED 50th, no eye height adjustment).

Two-lane: The two-lane road without a shoulder represented the base case that was used in all of the comparisons using the sign placement dimensions mentioned in the preceding paragraph.

Four-lane: This variable represented a four-lane roadway with no shoulders and the vehicle in the left lane of the two lanes in each direction. The sign height remained at 9 ft but the sign was offset 26 ft to the right of the lane the vehicle was occupying.

Left: The sign was located on the left side of a two-lane highway with no shoulders. The sign height remained at 9 ft but the sign was offset 26 ft to the left of the lane the vehicle was occupying.

Overhead: The sign was centered over the lane the vehicle was occupying with a sign centroid height of 22 ft. This height resulted from an overhead clearance of 18 ft and a sign height of 8 ft.

Road Geometry

A straight (tangent) road served as the base case for calculating sign luminance in this analysis, which is consistent with how most previous studies of sign retroreflectivity have addressed road geometry. The following four different curved roadway geometries were included in the analysis for comparative purposes:

Curve right, 643 ft radius (represents 45 mph curve),

Curve left, 643 ft radius (represents 45 mph curve),

Curve right, 1,333 ft radius (represents 60 mph curve), and

Curve left, 1,333 ft radius (represents 60 mph curve).

The curve radii were calculated using Equation 3 for speeds of 45 and 60 mph with a maximum superelevation rate of 6% and side friction values of 0.15 and 0.12 for 45 and 60 mph, respectively. The author intended to include a curve with a radius representing a 30 mph roadway (231 ft radius) but found that the measurement geometry exceeded the values in the headlamp profiles. As a result, it was not possible to calculate luminance for that radius.

R = V^{2} / [15 (e \pm f)]

(3)

where

R = radius (ft),

V = speed (mph),

e = superelevation, and

f = side friction.

Retroreflective Sheeting

The greatest challenge in calculating sign luminance is having detailed information on the retroreflectivity performance of sign sheeting products. Although sign sheeting specifications may stipulate retroreflective performance at four to six measurement geometries, the sheeting specification retroreflectivity values do not sufficiently describe a material’s retroreflective performance. Detailed luminance analysis of retroreflective sign sheeting requires extensive photometric analysis of multiple samples of the same sheeting product. At present, such information is not available in the public domain for most retroreflective sheeting products.

The 2001 release of ERGO included sign sheeting files for several products that were measured in 1998. Each of these files has over half a million individual retroreflectivity measurement points, as indicated in Table 3. Several of these products are no longer available and most of the rest are not widely used on higher speed roadways. An additional retroreflectivity file became available in 2005 for a newer microprismatic sheeting product that is currently in wide use. This new material represents an ASTM Type XI ( 8 ) and an AASHTO Type D ( 9 ) product. This sheeting product served as the base case for all of the comparisons. Given the lack of retroreflectivity information on sheeting products currently in wide use, there were limited options for making sheeting comparisons. The author ultimately selected one of the 1998 retroreflectivity files that represents an ASTM Type III and an AASHTO Type A product. The 1998 Type III/A product is no longer produced but the comparison does provide a sense of the potential range in luminance performance of retroreflectivity sheeting. The retroreflectivity measurements for both sheeting products were made on white sheeting. The retroreflectivity values would be lower for sheeting of the same type in different colors.

Table 3.

Range of Retroreflectivity Measurements for Exact Roadway Geometry Output (ERGO) Materials

Alpha (observation)			Beta (entrance)			Epsilon (rotation)			Omega (orientation)			No. of Msmt
Min.	Max.	Incr.	Min.	Max.	Incr.	Min.	Max.	Incr.	Min.	Max.	Incr.	No. of Msmt
0.2	4.0	0.05	0	48	4	–180	+180	30	–180	+180	10	506,475

Note: Min. = minimum; Max. = maximum; Incr. = increment; No. = number; Msmt = measurements.

Results

Having defined all of the alternatives for input variables, the author calculated the sign luminance for 27 different scenarios in ERGO and created comparisons for the key input variables including vehicle size, headlamp, driver eye height, road geometry, sheeting type, and best/worst case. The comparisons are presented in the following subsections. Each subsection includes a plot of the sign luminance associated with the variables evaluated in that comparison. For all of the plots, the red (solid) line represents the base case, which is an Edge with LED 50th percentile headlamps, average driver eye height, and Type XI/D sheeting product on a straight road.

Impact of Vehicle Size on Sign Luminance

This analysis compared sign luminance on a straight road in which vehicle size was the only factor that changed. The vehicle comparison used the 50th percentile LED headlamp and the base driver eye height for all vehicles. Figure 2 shows the sign luminance for each of the three vehicles. Owing to the measurement geometry, the author could not calculate luminance at distances less than 60 ft because of the horizontal angle for the left headlamp exceeding the values in the headlamp array. The difference between the minimum and maximum luminance was 11.6 cd/m² at a distance of 240 ft (the maximum legibility distance for a 6 in. legend). At this distance, the luminance produced by the Edge was about 37% greater than that produced by the Expedition. The luminance for the Edge represents the base case in this comparison and for all of the other comparisons. An interesting point about this comparison is that although the Escape is a smaller vehicle than the Edge, the luminance associated with the Escape is slightly less than that produced by the Edge, probably as a result of the larger separation of headlamps and/or a lower headlamp height in the Escape. This shows that relative vehicle size alone is not a good predictor of which vehicle will provide a greater luminance.

Both the Edge and the Escape vehicles met the 95th percentile demand luminance values between 120 and 204 ft. The luminance produced by the Expedition dropped below the 95th percentile requirement at about 143 ft and dropped below the 85th percentile requirement at about 123 ft.

Impact of Headlamp on Sign Luminance

This analysis compared the sign luminance that resulted from using the six different 2018 model year headlamp profiles on a straight road using the Edge and base driver eye height. Figure 3 plots the luminance values that resulted from the six headlamp profiles. It is worth noting that the luminance scale for this comparison represents a larger range than the other comparisons. The luminance resulting from the different headlamps had the greatest variation of all the comparisons. As can be seen in this figure, the 75th percentile market weighted headlamps had luminance values that were much higher than the other profiles. The luminance resulting from these headlamps was high because the luminous intensity was greater than that of the other profiles. At the maximum legibility distance of 240 ft (for a 6-in. legend), the difference between the maximum and minimum luminance values was almost 110 cd/m².

Figure 4 presents the same results as in Figure 3 except that the vertical scale has been reduced to be the same as that used for most of the other comparisons. This provides the ability to see that three of the profiles dropped below the 95th percentile luminance requirement between 120 and 145 ft, and that the TH 25th percentile headlamp dropped below the 85th percentile demand luminance between 120 and 125 ft. All of the profiles provided a luminance greater than the 75th percentile demand luminance.

Impact of Eye Height on Sign Luminance

Figure 5 illustrates the impact of eye height on sign luminance. Eye height was changed in 3-in. increments from 6 in. below the base case to 6 in. above. Eye height, relative to the vehicle headlamps, is an important factor as a driver closer to the headlamps has a smaller observation angle than a driver who is further away. The smaller the observation angle, the higher the sign luminance. As indicated in this figure, the sign luminance decreased as the driver eye height increased. The difference between the minimum and maximum luminance was 15.7 cd/m² at a distance of 240 ft (the maximum legibility distance for a 6-in. legend). At this distance, the luminance produced by the –6-in. eye height was about 45% greater than that produced by the +6-in. eye height.

The +3 in. eye height luminance fell below the 95th percentile requirement at about 123 ft from the sign whereas the +6 in. eye height luminance fell below the 95th percentile requirement at about 131 ft from the sign. All of the headlamp profiles remained above the 85th percentile luminance requirement throughout the critical legibility distance range.

Impact of Road Geometry on Sign Luminance

Sign luminance fluctuates to a greater degree on curved roads than on a straight road. This is demonstrated in Figure 6, which shows that the sign luminance for the curved roadways is less than that of a straight roadway at the longer legibility distances. At distances less than about 145 ft, the luminance for curves to the right was greater than that for a straight road. The larger fluctuation in sign luminance on curved roads may help to explain why most prior evaluations of sign retroreflectivity have focused on tangent roadway segments.

At the maximum legibility distance of 240 ft, there was a difference of 23.5 cd/m² between the highest and lowest luminance levels. The sign luminance associated with these roadway geometries was above the 95th percentile required luminance in all cases except for the curve to the left with a radius of 643 ft during the most distant 3 ft (237 to 240 ft) of the critical legibility distance range.

Impact of Sign Position on Sign Luminance

Signs are located in a variety of positions relative to a vehicle and the sign luminance will vary depending on where the sign is located. Figure 7 indicates the sign luminance that results from four different sign positions. At the maximum sign legibility distance, the luminance range was 14.2 cd/m². All of the post-mounted signs remained above the 85th percentile demand luminance. Owing to the larger legend size normally used with overhead signs, the required luminance values were applied over a longer distance as the specific luminance level was based on the legibility index (ft/in.). This figure includes the required luminance for an 85th percentile accommodation associated with a 16-in. legend. The luminance for the overhead sign was greater than the 85th percentile required luminance for this size of legend.

Impact of Headlamp Changes Over Time on Sign Luminance

Both vehicles and headlamps change over time. Figure 8 indicates the change in sign luminance that resulted from only changing the model year of vehicle headlamp. The vertical scale on this figure has a higher maximum value than was used in most of the other comparisons in this paper. All other input variables used the same base case conditions as previously described. At the maximum legibility distance, the range in luminance values was 19.2 cd/m². The interesting aspect of this comparison was the relative sign luminance performance of these headlamps at distances beyond the maximum legibility distance. As can be seen in the plot, the sign luminance for the 2018 headlamp did not significantly increase as the distance increased.

The change in headlamp profiles is an important factor to be considered in sign sheeting selection. Although modern vehicle headlamps may seem brighter, the analysis indicated that sign luminance had decreased over time because of less light reaching the sign face. This is because modern headlamps direct more light below the horizon. As a result, sign sheeting that may have been adequate with headlamps typical of the vehicle fleet 10, 20, or 30 years ago may no longer meet the needs of drivers.

Impact of Sheeting on Sign Luminance

All of the prior comparisons used the same sheeting product as an input variable. The last comparison in this paper evaluates the performance of the microprismatic material (ASTM Type XI, AASHTO Type D) used in all the prior scenarios with an encapsulated bead product (ASTM Type III, AASHTO Type A). It is worth noting that the product used for the comparison is no longer sold by the manufacturer. Fully characterized retroreflectivity files are not available for other materials that are currently available. For this comparison, luminance was calculated for two vehicles (Edge and Expedition). Base case conditions were used for all other input variables. Figure 9 indicates the relative performance of the two sheeting products for each vehicle. The luminance values for both of the Type XI/D materials were the same as those shown in Figure 2; the luminance values for the Type III/A materials for each vehicle represent the new data. As can be seen, the beaded product for each vehicle was less than the 85th percentile demand luminance and at or slightly below the 75th percentile demand luminance. Despite the age and lack of product availability for the encapsulated bead product, the comparison did provide a sense of the difference in performance between a high-performing microprismatic material and a high-performing beaded product.

Comparison of Best and Worst Cases from Comparisons

The final comparison of sign luminance was created by combining the best- and worst-case values for each of the input factors. The best-case luminance resulted from using the following input factors: vehicle: Edge, headlamp: LED 75th percentile, eye height: −6 in., and geometry: straight. The worst-case luminance resulted from using the following input factors: vehicle: Expedition, headlamp: TH 25^th percentile, eye height: +6 in., and geometry: curve left with 643 ft radius. The results are shown in Figure 10. The luminance scale for this figure is the largest of all the comparative plots. At the maximum legibility distance of 240 ft, the difference in luminance between the best and worst case was 160 cd/m², which is a 2,250% difference. The best case was far above the 95th percentile required luminance whereas the worst case was between the values for the 75th and 85th percentile required luminance.

Conclusions

The calculations and comparisons presented in this paper demonstrate that real-world conditions can provide a wide range of sign luminance and that some of the scenarios evaluated as part of this study showed that the resulting sign luminance values fell below the luminance needs of drivers at either the 95th or 85th percentile levels for a least a portion of the distance during which signs are typically read.

In the author’s opinion, the most important takeaway from this paper is the lack of sign sheeting and headlamp performance photometric data to provide the ability to analyze sign performance for a variety of conditions. The lack of retroreflectivity performance criteria for sign sheeting products is especially critical as it prevents agencies from evaluating the relative performance of various sign sheeting products on their roadways.

The retroreflectivity file for the microprismatic material used in this evaluation was created over 15 years ago and the retroreflectivity values in that file may not accurately represent the retroreflective performance of the same product available today, although the assumption for this paper is that the current version of the product has the same retroreflective performance as it did in 2005. As previously mentioned, the additional product used in the sheeting comparison is no longer available.

The luminance calculations were based on retroreflectivity measurements that were performed on new material. At present, there is no information in the public domain that characterizes the retroreflective performance of sheeting that has been in place for a period of years. If sheeting products were removed from the field after 10 or 15 years and then the retroreflective performance were fully characterized, the resulting sign luminance analysis might show results that are substantially different from those presented in this paper.

Another challenge to agencies and individuals wishing to evaluate sign luminance is the lack of a program for calculating sign luminance. The ERGO program used in this analysis is no longer publicly available and there are no equivalent programs available to agencies and individuals. If detailed sign performance is to be evaluated in the future, such a program needs to be developed.

The comparisons presented in this analysis included the 75th, 85th, and 95th percentile required luminance levels for a sign with a 6-in. legend, to give the reader a sense of how well a specific sign product met driver needs for sign brightness in the described scenario. These luminance levels are based on the needs of older drivers, who have a tendency to self-regulate and avoid driving at night when their vision deteriorates to the point they have difficulty seeing. There are no set criteria for the level of luminance that a sign needs to have in a given scenario. Most roadway design criteria are based on something less than 100% accommodation of the driving population. The unanswered question for sign luminance is whether to base the required luminance on the 75th, 85th, or 95th percentile accommodation. The required luminance levels used in this analysis were for a white legend or a white background with a legend that was 6-in. tall. The required luminance level would be different for different colors and different legend sizes.

In summary, this paper identifies the wide range of sign luminance that can result from different combinations of vehicle, headlamp, driver, roadway, and sign sheeting. The profession has the ability to calculate sign luminance to better improve the selection of sign sheeting for specific applications, but lacks the information on sign retroreflectivity performance.

In conclusion, the author offers the following recommendations:

Agencies should require manufacturers to provide photometric information using the ranges indicated in Table 3 for retroreflective performance of the products they purchase or work cooperatively to develop sheeting retroreflectivity performance files;

Vehicle manufacturers should make headlamp geometry available on request;

Sign sheeting manufacturers or researchers should develop a tool for calculating sign luminance (similar to ERGO) and make that tool available to agencies and researchers; and

Agencies should work cooperatively to develop a fully characterized retroreflectivity database of sign sheeting products that have been in the field for 10 and 15 years. Information about each sample should include sheeting color, sign location, location elevation, location average annual temperature, exposure direction, and sign age when removed.

Footnotes

Author Contributions

The author confirms sole responsibility for the following: study conception and design, data collection, analysis and interpretation of results, and manuscript preparation.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Gene Hawkins

References

Uding

K. D.

Exact Road Geometry Output Program for Retroreflective Road Sign Performance. Transportation Research Record: Journal of Transportation Research Board, 1993. 1421: 61–68.

Carlson

P. J.

Performance Evaluation of Retroreflective Traffic Signs. Paper Number 15-2963. 94th Annual Meeting Compendium of Papers, Transportation Research Board, Washington, D.C., 2015.

Carlson

P. J.

Hawkins

H. G.

Updated Minimum Retro-Reflectivity Levels for Traffic Signs. Report FHWA-RD-03-081. FHWA, U.S. Department of Transportation, Washington, D.C., 2003.

Flannagan

M. J.

A Market-Weighted Description of Tungsten-Halogen and LED Low-Beam Headlighting Patterns in the U.S. UMTRI-2019-5. The University of Michigan Transportation Research Institute, Ann Arbor, 2019.

Schoettle

Flannagan

M. J.

A Market-Weighted Description of Low-Beam and High-Beam Headlighting Patterns in the U.S: 2011. UMTRI-2011-33. The University of Michigan Transportation Research Institute, Ann Arbor, 2011.

Schoettle

Sivak

Flannagan

M. J.

High-Beam and Low-Beam Headlighting Patterns in the U.S. and Europe at the Turn of the Millennium. Technical Report No. UMTRI-2001-19. The University of Michigan Transportation Research Institute, Ann Arbor, 2001.

Average Human Height by Country. https://en.wikipedia.org/wiki/Average_human_height_by_country.

ASTM D4956-19. Standard Specification for Retroreflective Sheeting for Traffic Control. ASTM International, West Conshohocken, PA, 2019.

AASHTO M 268. Standard Specification for Retroreflective Sheeting for Flat and Vertical Traffic Control Applications. American Association of State Highway and Transportation Officials, Washington, D.C., 2015.