Comprehensive evaluation of guardrail delineation systems : final report

Comprehensive Evaluation of Guardrail Delineation Systems
Final Report
Prepared for: Georgia Department of Transportation
Prepared by: A. Amekudzi, Ph.D., D. Folds, Ph.D., and A. Zureick, Ph.D.
Georgia Institute of Technology
April 2004

TABLE OF CONTENTS

List of Tables and Figures

3

Executive Summary

4

Introduction

5

Literature and Survey Findings

6

Performance Evaluation

7

Conclusions

34

References

35

Appendix A: Details of Delineator Installation Tasks

36

Appendix B: Details of Laboratory Conspicuity Study

39

Appendix C: Details of Field Conspicuity Study

42

Appendix D: Details of Delineator Field Inspection

44

2

LIST OF TABLES AND FIGURES

List of Tables

1: Dimensions of "as received" Guardrail Delineators

8

2: Temperature Data from National Climatic Data Center (Atlanta, Fulton County)

12

3: Dimensions of Test Specimens

14

4: Hardness Test Results

16

5: Flexural Test Results (span = 2 in)

17

6: Hardness Test Results - Stage 2

18

7: Installation Times (in Seconds) for Delineators on Metal Spacers

21

8: Installation Times (in Seconds) for Delineators on Wooden Spacers

22

9: Mean Distance until Response by Age Group and Delineator Type (Freeway)

26

10: Mean Distance until Response by Age Group and Delineator Type (Non-Freeway)

27

11: Number of Guardrails by Type and Spacing of Delineator

27

12: Mean Distance until Response by Age Group, Delineator Type, and Delineator Spacing

28

13: Distribution of Delineator Type and Percent of Guardrails Reported across the Range

of Reference Distances

30

14: Subjective Delineator Cleanliness Ratings

32

15: Wear Observations for Each Delineator Type

33

16: Unit Costs Estimates for Acquiring and Installing Delineator Systems

34

List of Figures

1: Traffic Markers, Type A, B and C

9

2. Traffic Markers Type D (white, yellow), E1, and E2

9

3: Traffic Markers Type F1

10

4: Traffic Markers Type F2

10

5. One Freezing and Thawing Cycle (ASTM C666)

13

6: Flexural Test Setup

14

7: Hardness Test in Progress

15

8: Flexural Test in Progress

15

3

EXECUTIVE SUMMARY This study evaluates selected guardrail delineators being considered for use by the Georgia Department of Transportation (GDOT) based on their durability, conspicuity, maintainability and costs. The purpose is to conduct a comparative assessment of candidate delineators to enable GDOT make appropriate informed decisions on the specifications and details for future guardrail delineation efforts. Samples of guardrail from six manufacturers (Astro Optics, AKT, Carsonite, Davidson Plastic, Duraflex and PSS) in various shapes (T, I and L), various dimensions and colors were tested for their relative durability, conspicuity and maintainability. The candidate delineators were also compared based on their acquisition and installation costs. The study found no significant differences among the samples based on laboratory tests for durability, field inspections, and laboratory and field conspicuity tests, although the PSS system was found to have the most problems with installation and wear. Based on purchase and installation costs alone, it appeared that the PSS-L, Duraflex, Astro Optics acrylic and Davidson Plastics were the lower-cost alternatives. The field conspicuity tests show that 100 foot spacing of delineators in some conditions will produce satisfactory conspicuity. Use of 100 foot spacing could lead to significant cost savings in purchase and installation costs. Further evaluations could clarify the range of roadway conditions for which 100 foot spacing is appropriate.
4

1. INTRODUCTION Guardrail (also known as guiderail) is used throughout the United States to improve safety along roadways and interchange ramps. Installation is appropriate at locations where the risk of injury and property damage from a roadside hazard such as a fixed object or non-traversable slope is greater than the risk imposed by the guardrail. However, even when used properly, guardrail does constitute a fixed roadside obstacle. The hazard created is especially significant at night because guardrail systems are typically relatively difficult to see. Fixed object accident studies have shown that crashes involving guardrail increase at night (Campi 1990).
In the 1980s and 1990s, there were a number of efforts to develop appropriate delineation methods to improve the nighttime visibility of guardrail systems. In addition, agencies that are interested in evaluating alternative systems for use will benefit from having an evaluation framework that incorporates important performance characteristics. Important measurable characteristics of candidate guardrail delineation systems are as follows:
Conspicuity and positive driver perception; Durability under field conditions; Ease of installation and maintenance and Cost effectiveness.
Objective The purpose of this study was to conduct a comparative assessment of candidate delineators to enable Georgia Department of Transportation (GDOT) make appropriate informed decisions on the specifications and details for future guardrail delineation efforts. Samples of candidate guardrail delineators were provided by GDOT for testing. The evaluation was conducted on samples of guardrail delineators from six manufacturers (Astro Optics, AKT, Carsonite, Davidson Plastic, Duraflex and PSS) in various shapes (T, I and L), various dimensions and colors.
Methodology A literature review and survey were conducted to establish the context for comparatively evaluating guardrail delineation systems. Conspicuity and positive driver perception tests were conducted to determine the relative visibility of the delineation systems from the driver viewpoint. Durability tests were conducted to establish the relative robustness of competing systems. Installation and maintenance data were obtained to evaluate the relative ease of installing and maintaining the systems. Finally, cost data was obtained to evaluate the cost-effectiveness of the different delineator systems.
Outline The following sections of the report present and discuss the results of the literature review, survey and tests, and develop a comprehensive assessment for the samples. Detailed reports on the field inspections and laboratory conspicuity tests are included in Appendices A, B and C.
5

2. LITERATURE AND SURVEY FINDINGS While relatively little exists in the literature on guardrail delineation, the few articles that were found reinforce one another.
A study conducted by Campi for the New Jersey Department of Transportation (NJDOT) in the late 1980s investigated the need to delineate guardrails and evaluated the performance of different types of delineators under various field conditions. Presented in the Transportation Research Record 1160 (1988) and in an NJDOT report (1990), the study evaluates candidate delineated guardrails based on the ease of installation, resistance to soil, durability and cost. Based on a field evaluation of more than 20 different types of guardrail delineators, the study identifies a need for guardrail delineation to increase the benefits of guardrail to nighttime drivers. It concludes that the improvement in the nighttime visibility of guardrail through delineation should result in a reduction in guardrail accidents, which would help to offset the initial cost of delineation. In addition, the study determines that a flexible panel and metal bracket system manufactured by the Carsonite Company was the most suitable device with regard to durability, soil accumulation and ease of installation.
An article by Roberts (NJDOT) in the June 1986 issue of the Transportation Research Circular states that visually effective and useful reflective treatments appear to be lacking in many guardrail situations; the most easily maintained and cost effective installations, devices and mounting methods are not well known and documented, and guardrail does appear to be an area worth research investment because improvements in this area stand to reduce the costs of guardrail crashes. Roberts indicates that reflectorization research is important and should be oriented toward the following:
1. Determining where and under what conditions there is a need for better guardrail visibility, especially at night;
2. Finding the best ways, where needed, to improve the visibility of guardrails for both clear and adverse weather conditions and night, and
3. Determining if there are any improvements in traffic flow or alternative safety measures that can be found and documented.
He suggests the most difficult and expensive research to perform is that which demonstrates improvements with alternative safety measures. Such research, however, would be most needed because it would provide the most highly quantified and least subjective evidence to justify the installation and maintenance of reflector treatments on guardrails. He concludes that the following questions need to be considered: (i) Where and when is guardrail reflectorization needed? (ii) How much does its use improve safety and annual repair costs? (iii) What materials and mounting methods should be used? (iv) What are the maintenance costs?
As part of the GDOT study, a survey was developed to help assess the state of the practice in guardrail delineation in the southeastern states. This survey was sent out to five states: North Carolina (NC), South Carolina (SC), Florida (FL), Alabama (AL) and Mississippi (MS). South Carolina, with 1,296 miles of guardrail, responded to the
6

survey, indicating that a relatively small percentage (2-24%) of their guardrail was delineated. South Carolina indicated that most of the top mounts are made by Carsonite. Others could be made by any number of manufacturers so long as they meet the DOT specification. South Carolina Department of Transportation (SCDOT) has standards and guidelines concerning guardrail delineation. There are guidelines on spacing, color, materials, dimension and mounting steps. All cable systems must be delineated. The DOT has written criteria about maintenance; however engineers make decisions on replacement based on both missing markers and visual inspection. The DOT was asked to identify the most important issues when choosing delineators based on the following criteria: higher conspicuity/reflectivity; higher flexibility/durability; ease of installation, maintenance and replacement; and cost effectiveness. The DOT was also given the opportunity to specify other important criteria. Maintainability and conspicuity were SCDOT's most important criteria when choosing delineators. The agency explained that they felt that being able to see the unit was the most important; the other criteria were not as important.
The performance evaluation criteria used in this study are reinforced by the criteria identified as important in the literature and the survey. The guardrail delineation samples provided by GDOT were evaluated for their durability, conspicuity, maintainability and relative costs. The following sections discuss the methodology and results for each of these evaluations and provide a synthesis of the results. More detailed results for the field inspections and laboratory conspicuity tests are included in Appendices A, B and C.
3. PERFORMANCE EVALUATION The candidate guardrail delineators provided by Georgia Department of Transportation (GDOT) were evaluated on their relative durability, conspicuity, ease of installation and purchase and installation costs. The sections below describe the methodology and results of these evaluations.
3.1 Durability Analysis Objective The durability study was conducted to examine the effects of environmental exposure on commercially available plastic materials used in the construction of guardrail delineators being considered for adoption by GDOT. Such plastic materials form the base upon which reflective sheeting is bonded to improve visibility for motorists, and thus safety. These delineators need to be selected so that they are not seriously degraded when exposed to various environmental conditions during their service life.
Guardrail Delineators All materials examined in this study were obtained from guardrail delineators supplied by GDOT. Table 1 presents information on all the delineator samples "as received" from GDOT, also shown in Figures 1 to 4.
7

Table 1: Dimensions of "as received" Guardrail Delineators

Marker Type

Manufacturer

Shape

Overall Dimension (in x in x in)

Reflector Dimension
(in x in)

Reflector Color

A

Astro Optics

T 5.25 x 2.82 x 0.087 4.75 x 2.5 white

B

AKT

I

4.13 x 5.89 x 0.122 4.13 x 4

yellow

C

Carsonite

I

3.47 x 9.04 x 0.153

3 x 5

yellow

D

Davidson Plastic

I

2.98 x 6.06 x 0.101

3 x 3

D

Davidson Plastic

I

2.98 x 6.06 x 0.101

3 x 3

white yellow

E1

Duraflex

E2

Duraflex

F1

PSS

F1

PSS

F1

PSS

F2

PSS

F2

PSS

F2

PSS

I

3.12 x 6.03 x 0.115

3 x 5 yellow/red

T

3.12 x 4.72 x 0.115

3 x 5

yellow

I

3.01 x 11.42 x 0.125

3 x 4

yellow

I

3.01 x 11.42 x 0.125

3 x 4

white

I

3.01 x 11.42 x 0.125

3 x 4 yellow/red

L

6.54 x 11.42 x 0.125

6.5 x 4

yellow

L

6.54 x 11.42 x 0.125

6.5 x 4

white

L

6.54 x 11.42 x 0.125

6.5 x 4 yellow/red

8

Figure 1. Traffic Markers Type A, B, and C
Figure 2. Traffic Markers Type D (white, yellow), E1, and E2 9

Figure 3. Traffic Markers Type F1
Figure 4. Traffic Markers Type F2 10

Test Environmental Conditions To determine the temperature and temperature range that delineators will be subjected to during testing, the annual climatological data from 1992 to 2002 for the Atlanta-Fulton County area, State of Georgia, as published by the National Oceanic and Atmospheric Administration were consulted. Table 2 summarizes the data. Due to the absence of GDOT guidelines for material degradation conditioning procedure, it was deemed necessary to conduct the experimental program in two stages. Stage 1: Samples were conditioned in accordance with the procedure standardized by the construction industry as described in the ASTM C666 Standard Test Method for Resistance of Concrete to Freezing and Thawing. Stage 2: Samples were conditioned under a cyclic temperature guided by the climatological data for the State of Georgia. Such a conditioning procedure consisted of 195 cycles from 32oF to 60oF to simulate three years of freeze thaw cycling. The 65 cycles/year was derived from examining the average number of days experiencing freezing-thawing in each month, and the temperature range as listed in Table 2. This was then simplified as follows:
1. Fifteen cycles from 32oF to 60oF for the month of January. 2. Ten cycles from 37oF to 60oF for the month of February. 3. Ten cycles from 40oF to 66oF for month of March. 4. Thirty cycles from 32oF to 60oF for the remaining months. From practical point of views, the temperature range between 32oF to 60oF was considered for all different months. In addition to the freeze-thaw aging condition, aging of the delineator materials at high temperature (112oF, the maximum temperature recorded between 1892 and 2002) for a duration of 14 days was performed.
11

Table 2: Temperature Data from National Climatic Data Center (Atlanta, Fulton County)

Month

Mean Minimum Temperature from 1994 to 2002

Mean Maximum Temperature from 1994 to 2002

Average Number of Days
Experienced Freezing-Thawing
from 1994 to 2002

January

33.9oF (1.0oC)

54.9oF (12.7oC)

14.4

February

37.2oF (2.9oC)

59.9oF (15.5oC)

9.8

March

41.7oF (5.4oC)

66.2oF (19.0oC)

6.4

April

49.1oF (9.5oC)

74.5oF (23.6oC)

0.9

May

57.2oF (14.0oC)

81.1oF (27.3oC)

0

June

65.3oF (18.5oC)

85.9oF (30.0oC)

0

July

69.6oF (20.9oC)

89.3oF (31.8oC)

0

August

68.4oF (20.2oC)

88.6oF (31.4oC)

0

September

62.0oF (16.7oC)

82.6oF (28.1oC)

0

October

51.0oF (10.6oC)

74.4oF (23.6oC)

1.0

November

41.4oF (5.2oC)

64.8oF (18.2oC)

6.4

December

35.1oF (1.7oC)

56.4oF (13.6oC)

13.4

Total

52.4

Stage 1 Test Results To determine the effects of freeze-thaw on the delineator materials, samples were placed in an environmental chamber in which the temperature was alternately lowered from 40 F to 0 and raised from 0 to 40 F within five hours as shown in Figure 5. One group of delineator samples was subjected to normal room conditions as a control for the experiment. Flexural and hardness tests were conducted on samples that were subjected

12

to 33 freeze-thaw cycles in one case and to 68 freeze-thaw cycles in another case. In all flexural tests, coupons excised from the delineator samples were tested in accordance with ASTM D790 Standard Test Methods for Flexural Properties of Unreinforced and Reinforced Plastics and Insulating Materials. A schematic showing the test set-up is shown in Figure 6. The dimensions of the test coupons from each delineator type are given in Table 2. The hardness tests were conducted on the "as received samples" and also on those subjected to 33 freeze-thaw cycles and 68 freeze-thaw cycles following the ASTM D 2240 Standard Test Method for Rubber Property-Durometer Hardness. Figure 7 shows a hardness test in progress while Figure 8 shows a flexural test in progress. The Shore D Hardness test results of all test samples are summarized in Table 3 and appear to fall in the same range, indicating that there is no significant difference detected as a result of freeze-thaw aging in accordance with the ASTM C666 test conducting procedure. It was also concluded that the flexural strength values of delineator materials of all types were unaffected by the freeze-thaw cycling procedure described in ASTM C666.
Figure 5. One Freezing and Thawing Cycle (ASTM C666)
13

Figure 6. Flexural Test Setup

Table 3: Dimensions of Test Specimens

Specimen Type
A B C D E1 E2 F1 F2

Thickness (in)
0.087 0.122 0.153 0.101 0.115 0.115 0.125 0.125

Hardness Test Specimen

Flexural Test Specimen (in x in)

As received As received As received As received As received As received As received As received

2.392 x 0.5 2.952 x 0.5 3.448 x 0.5 2.616 x 0.5 2.840 x 0.5 2.840 x 0.5 3.000 x 0.5 3.000 x 0.5

14

Figure 7. Hardness Test in Progress
Figure 8. Flexural Test in Progress 15

Table 4: Hardness Test Results

Specimen

Hardness (Shore D) of "As Received" specimens (ASTM D2240)

Hardness (Shore D) of Specimens
After 33 Cycles of Freezing and Thawing

Hardness (Shore D) of Specimens
After 68 Cycles of Freezing and Thawing

No. of Readings

Average C.O.V

Hard-

.

ness

(%)

No. of Readings

Average C.O.V

Hard-

.

ness

(%)

No. of Readings

Average C.O.V

Hard-

.

ness

(%)

A-H-T1-1

10

75.1

0.9

10

74.9

1.3

10

75.6

1.4

A-H-T1-2

10

75.0

0.9

10

74.0

1.1

10

75.2

0.6

A-H-T1-3

10

74.6

0.7

10

73.4

0.3

10

74.9

0.8

B-H-T1-1

10

80.4

1.2

10

80.2

1.6

10

81.0

1.0

B-H-T1-2

10

80.0

0.7

10

79.8

0.7

10

80.8

0.7

B-H-T1-3

10

80.7

0.9

10

79.8

0.8

10

81.1

0.4

C-H-T1-1

10

76.8

1.1

10

77.2

1.1

10

78.5

1.3

C-H-T1-2

10

77.2

0.8

10

78.0

1.0

10

78.8

1.2

C-H-T1-3

10

77.7

1.1

10

78.0

1.3

10

78.8

1.2

D-H-T1-1

10

73.9

1.0

10

73.1

1.6

10

74.2

1.0

D-H-T1-2

10

74.0

0.8

10

72.9

1.6

10

73.6

1.0

D-H-T1-3

10

74.0

1.4

10

73.1

1.2

10

73.9

1.0

E1-H-T1-1 10

76.6

1.0

10

76.5

1.1

10

77.3

1.2

E1-H-T1-2 10

75.7

1.2

10

76.0

0.8

10

76.4

1.2

E1-H-T1-3 10

76.5

0.9

10

75.5

1.9

10

76.5

1.3

E2-H-T1-1 10

76.7

0.7

10

76.5

0.6

10

76.8

0.7

E2-H-T1-2 10

76.8

0.9

10

76.2

0.6

10

76.7

0.6

E2-H-T1-3 10

77.2

0.8

10

76.2

0.8

10

76.7

0.9

F1-H-T1-1 10

67.0

1.1

10

67.2

1.0

10

67.2

1.0

F1-H-T1-2 10

67.5

0.5

10

67.3

1.0

10

67.2

1.3

16

F1-H-T1-3 10

67.7

0.3

10

67.6

0.9

10

67.4

0.7

F2-H-T1-1 10

68.0

1.0

10

67.8

0.9

10

67.8

0.9

F2-H-T1-2 10

67.8

0.5

10

68.0

0.4

10

68.1

0.4

F2-H-T1-3 10

67.6

0.9

10

67.7

1.1

10

67.1

1.1

Table 5: Flexural Test Results (span = 2 in)

Specimen Type

Flexural Strength of "As Received" Specimens

Flexural Strength of Specimens
After 33 Cycles of Freezing and Thawing

Flexural Strength of Specimens
After 68 Cycles of Freezing and Thawing

No. of Tests

Average C.O.V

Strength .

(ksi)

(%)

No. of Tests

Average C.O.V Strength .
(ksi) (%)

No. of Tests

Average C.O.V

Strength .

(ksi)

(%)

A

N.A.

N.A. N.A. N.A.

N.A. N.A. N.A.

N.A. N.A.

B

5

10.2

1.0

5

9.8

10.0

5

11.2

4.4

C

5

13.3

0.5

5

13.3

0.7

5

13.9

1.2

D

5

8.9

4.7

5

8.5

4.8

5

9.3

5.4

E1

5

7.9

9.6

5

8.5

8.5

5

8.6

14.4

E2

5

7.9

9.5

5

8.4

18.0

5

9.3

5.2

F1

5

6.1

1.6

5

5.7

3.7

5

6.7

3.8

F2

5

6.0

3.1

5

6.0

4.7

5

6.8

1.9

17

Stage 2 Results Based on the test results obtained from Stage 1 and the limited budget allocated to the laboratory experiments, the flexural tests in accordance with ASTM D 2240 were eliminated and only hardness tests on conditioned samples were conducted. Test results from Stage 2 are given in Table 6 showing that the Shore D hardness values were unaffected by the environmental exposure used in this experimental program.

Table 6: Hardness Test Results - Stage 2

Specimen

Hardness (Shore D) of "As Received" specimens (ASTM D2240)

Hardness (Shore D) of Specimens
After 195 Cycles of Freezing and Thawing (Stage2)

Hardness (Shore D) of Specimens
After aging under 112 F for 14 Days

No. of Readings

Average C.O.V

Hard-

.

ness

(%)

No. of Readings

Average C.O.V

Hard-

.

ness

(%)

No. of Readings

Average C.O.V

Hard-

.

ness

(%)

A-1

10

75

2

10

72

2

10

74

4

A-2

10

75

2.2

10

73

4

10

75

3

A-3

10

75

2

10

73

3

10

75

3

B-1

10

81

1.5

10

81

3

10

79

4

B-2

12

82

3

10

81

2

10

79

5

B-3

12

82

2

10

81

3

10

81

2

C-1

12

77

2

10

76

2

10

76

3

C-2

12

77

3

10

78

4

10

76

2

C-3

12

79

1.5

10

78.0

2

10

78

2

D-1

12

75

2

10

73

2

10

74

3

D-2

12

74

2

10

72

2

10

73

4

D-3

10

74

2

10

73

2

10

72

5

E1-1

10

77

1.0

10

75

2

10

76

1

E1-2

10

76

2

10

74

2

10

77

3

E1-3

10

77

2

10

73

5

10

77

3

E2-1

10

76

2

10

74

2

10

77

2

18

E2-2

10

78

3

10

75

2

10

77

3

E2-3

10

78

2

10

76

2

10

77

3

F1-1

10

68

4

10

66

2

10

68

5

F1-2

10

68

3

10

67

2

10

65

4

F1-3

10

68

3

10

67

2

10

66

4

F2-1

10

68

4

10

67

2

10

66

4

F2-2

10

68

4

10

66

2

10

65

4

F2-3

10

68

5

10

66

2

10

65

5

Conclusion Based on the experimental studies conducted in the durability analysis, all material types used in the traffic delineators considered by the GDOT perform virtually in the same manner when subjected to the environmental conditions as described earlier. The selection of the best traffic delineator should be based on visibility and cost properties that are beyond the scope of the durability analysis task.
3.2 Delineator Installation Task Analysis Ease of installation must be assessed by several complementary measures. In this review, the number of steps (sub-tasks) required for installation, the relative difficulty of each subtask, and the time required for installation were studied. The steps involved are very simple, so mental workload is not an important consideration. A detailed breakout of the installation sub-tasks for each delineator installation that was observed is presented in Appendix A.
A review of the guardrail delineator installation procedures revealed that the majority of the installation tasks are common to all delineators for a particular type of emplacement (i.e., for wood or for metal spacers). Slight differences in technique were observed between contractors and in some cases we observed a "demonstration" installation instead of an installation performed by a typical road crew. In most instances, the work may be accomplished by a one-person crew if necessary, but crews of at least two members are required for large stretches of guardrail in order to move the equipment truck forward as the installation progresses.
The installations of delineators by GDOT crews on the I-20 field test section were not representative installations (the delineators were not permanently affixed to the guardrail,

19

but instead were only temporarily mounted for the study), and were therefore not considered as a part of this task analysis.
Metal Spacers For metal guardrail spacers, there were six products evaluated. They fell into two categories, top-mounted (split flange) or side-mounted (straight flange) types.
Although procedures may differ slightly from crew to crew, the sidemounted delineators require additional steps for preparation and/or to ensure adhesion. PSS delineators, which are affixed with adhesive tape, require the crew to prepare each delineator by placing a strip of adhesive tape on the back. Prior to installation, the backing is peeled off the tape strip, and then the delineator is pressed onto the web of the spacer. CARSONITE delineators, which are affixed with adhesive paste, required the additional application of clothespins during the drying process to ensure adhesion and alignment. The crew had to then return to the site to retrieve the clothespins the next day. Side-mounted delineators are also affected more by weather conditions such as dampness or cold temperatures. Below 35 degrees, the metal spacers had to be heated with an electric heat gun for 15-25 seconds to achieve good adhesion for the adhesive strips used by the PSS system. Most contractors reported that they attempted to adjust installation schedules to avoid cold (< 35 degrees F) or wet conditions.
The installation procedures for the top-mounted delineators were virtually the same for all products. Although there were notable differences in technique between crews, the basic procedure is to apply a bead of adhesive to the top of the spacer web and push the flange of the delineator onto the web. The delineator is held in place by the flange while the adhesive sets. No additional measures are necessary to ensure adhesion. Interestingly, DAVIDSON PLASTICS recommends placing the bead of adhesive into the spacer flange channel rather than onto the guardrail spacer. None of the contractors were observed using this method. Crews installing the DURAFLEX and AKT delineators used three beads of adhesive (top and both sides) and crews installing the DAVIDSON PLASTICS used only one bead of adhesive along the top. It is not known if this difference between installation procedures was related to actual physical differences in the delineators (e.g., tighter fitting flange), quality of adhesive (DP-1 vs. DAPCO 795A), or merely differences in individual technique. On two of the products (AKT and DAVIDSON PLASTICS), the contractor took four additional steps to spread the adhesive more evenly, but the relative affect on adhesion is unknown. This would appear to be merely a difference in technique and not an installation requirement. The crew installing the DURAFLEX delineator was notably slower accomplishing the task, primarily due to individual skill/speed. The following table summarizes the observations of delineator installation times for delineators on metal spacers. Table 7 shows the observed installation times for the delineators on metal spacers.
20

Table 7: Installation Times (in Seconds) for Delineators on Metal Spacers

Metal Spacers Type Prep

DURAFLEX DAVIDSON AKT CARSONITE PSS "L" & PSS "I"

Topmount Topmount Topmount Sidemount Sidemount

3.1**

Apply Adhesive 20.4

Position Confirm

Delineator Adhesion

3.1

0.9

Total 24.4

3.8

8.9

13.7

9.5

8.5

18.0

4.6

4.9

8.8*

18.3

6.0

4.3

4.3

17.7

* This does not include time required to return to the installation site to retrieve the clothespins used to hold the delineators in place.

** In cold weather, a heat gun is used to prepare the surface of the spacer for the adhesive tape. This took an additional 30 seconds on average.

Wooden and Rubber Spacers For wooden and rubber guardrail spacers, the delineators are again divided into two types, top-mounted and side-mounted. Wooden and rubber spacers are sometimes tilted and must be straightened prior to installation of either type of delineator. The installation procedure for the top mounted delineators is identical for both AKT (not observed) and ASTRO-OPTICS (sheeted and acrylic). Installation is straightforward and simple, with two nails being driven into pre-drilled holes in the flange of the delineator.
Alignment is easy to maintain as vertical alignment is assured as long as the spacer is straight. Orthogonal alignment is stable once the delineator is placed. While not directly observed, installation on rubber spacers is assumed to be identical.

Side-mounted delineators are slightly more difficult to install due to the requirement to hold the delineator in place while driving nails or screws. Maintaining vertical alignment while driving nails or screws presents a slight challenge; however, orthogonal alignment is assured by alignment with the face of the spacer. PSS delineators have a notched side, which allows the installer to quickly align the delineator on the spacer. In some cases, there were no pre-drilled holes, or the screw heads were smaller than the pre-drilled holes. In these instances, screws were driven
directly through the flange of the delineator (PSS), a slightly more difficult task while holding the delineator in place. Corporate product literature (DAVIDSON PLASTICS) recommends using nails with washers, although no contractors were observed following

21

this procedure. The following table summarizes the delineator installation times for delineators on wooden spacers. Table 8 shows the installation time for delineators on wooden spacers.

Table 8: Installation Times (in Seconds) for Delineators on Wooden Spacers

Wooden Spacers ASTROOPTICS S ASTROOPTICS A PSS "I" & PSS "L"

Type Top-mount Top-mount Side-mount

Position Delineator 1.1
1.0
2.0

Apply Fasteners 8.0
7.0
18.3

Total 9.1 8.0 20.3

Conclusions For metal spacers, the installation of top-mounted spacers requires fewer steps and takes less time in most installations, with vertical and orthogonal alignment almost guaranteed. Side-mounted delineators are slightly more difficult to align and adhesion can be problematic at times. AKT, DURAFLEX, and DAVIDSON PLASTICS products appear identical for ease of installation. For the DAVIDSON PLASTICS delineator, one contractor used only one bead of adhesive versus three beads of adhesive they used on other products. One installation crewmember cited superior adhesive as the probable reason for the difference, but the actual reason remains undetermined. Other factors such as maintainability and cost will determine the best delineator for this application.

For wooden or rubber spacers, the top-mounted delineators are slightly easier to install due to the simpler placement/alignment task and relative stability gained by resting on top of the horizontal surface of the spacer. Of the installations observed directly, the ASTRO-OPTICS products were the easiest to install, but other vendors (e.g. AKT) apparently have similar designs that would be identical in installation. Again, other factors such as maintainability and cost will determine the best delineator for this application.

3.3 Conspicuity and Positive Driver Perception

Description of the Experiment The conspicuity and positive driver perception experiment consisted of two parts, a lab study and a field study. In each study, participants were presented with segments of roadway (in the form of prerecorded video in the lab portion, and live, as a passenger in a vehicle, in the field portion) and were tasked with indicating to the experimenter each object that they perceived from a defined set. The set of objects included guardrails, various types of signs, and disabled vehicles. (The non-guardrail objects were included to increase the workload of the participants to simulate actual driving, and to prevent them from solely focusing on spotting guardrails.) All prerecorded video was recorded at

22

night, and all field data collection was performed at night, so that light reflected by the delineators was the primary means of perceiving the guardrail segments.
Participants took part in the experiment individually under the direction of an experimenter, first in the lab and then in the field. Participants were trained to look for objects of interest on the right shoulder or overhead, to press a button as soon as they had confidently identified the presence of an object of interest, and to announce verbally to the experimenter the type of object identified. Button presses were automatically recorded with time and position stamps, and the participants' verbal announcements were logged by the experimenter.
Participants Thirteen participants, seven females and six males, took part in the study. The participants were recruited from both within and outside of Georgia Tech. Seven of the participants were under the age of 55 (primarily college students and people in their early 30s), and the other six were people over the age of 55. Two additional participants served as pilot participants; their data are not included in the analyses reported below.
Apparatus and Stimulus Materials The lab study was conducted in the Driver Perspective Simulator (DPSIM) at GTRI. The DPSIM consists of a large TV monitor (on which out-the-windshield video is presented), with computer-controlled video presentation. A second computer (the Experimenter Control Station) controls the DPSIM. Participant responses are automatically timestamped and recorded.
The video for the lab portion of the study was collected by mounting a video camera in a vehicle to record video from the driver's perspective. The video was recorded while the vehicle was driven at approximately 60 MPH in the rightmost through lane in areas where GDOT contractors had previously installed the various types of guardrail delineators being studied. The specific locations that were recorded are listed in Appendix B. GPS position data was collected simultaneously. The video was then edited into brief (typically one to two minute) segments containing multiple guardrails and other objects of interest. Each video segment was categorized as "dense" or "sparse" based on the number of objects of interest contained in the segment. Video segments were selected for inclusion in the experiment based on the validity of GPS data (without which, participant responses could not be scored), quality of the video, and presence of guardrail segments where the beginning of the segment was not obscured by traffic or other objects present when the video was collected. Some video segments without valid GPS data were used as demonstration trials.
Stimulus materials for the field study were the delineators installed on the test site along I-20E in Atlanta. GDOT personnel installed the delineators in accordance with our plan for balancing delineator type and delineator spacing along the eastbound and westbound segments of the test site. Eleven of the guardrails used in the field study had delineators installed at the regular spacing of 75 feet between delineators, and ten of the guardrails had delineators installed with an alternate spacing of 100 feet between delineators. This
23

manipulation was done to allow assessment of the conspicuity of the wider spacing in an environment that permitted direct comparison with the standard spacing. More details about the delineator installation for the field test sections are provided in Appendix C.
Procedure The lab study began with the experimenter describing the study and obtaining informed consent from the participants. Participants were then trained on the task to be performed. The participant was then presented with a series of trials, each consisting of a single video segment. The first set of trials was all freeway segments. The first trial was a demo trial illustrating the appearance of the objects of interest, which for the freeway trials included guardrails, green exit-related signs, orange construction-related signs, and disabled vehicles. The next two trials were practice trials, and then a series of 9 freeway testing trials was presented. The second set of trials was all non-freeway segments. Again, the first trial was a demo trial illustrating the appearance of the objects of interest, which for the non-freeway trials included guardrails, yellow intersection-related signs, orange construction-related signs, and disabled vehicles. The next two trials were practice trials, and then a series of 11 non-freeway testing trials was presented. Only data from the testing trials was used in the data analysis.
The testing trials alternated between sparse and dense segments, and also among the various types of delineators. Groups 1 and 3 viewed the testing trials in the base order, and groups 2 and 4 viewed the testing trials in reverse order, to offset any effects caused by the order of presentation of the trials. More details about the laboratory trials are available in Appendix B.
After the final lab trial, participants were taken in a Georgia Tech van for the field study. A driver drove the van at approximately 60 MPH in the rightmost through lane of the test section, which was located on I-20 east of I-75/85 in downtown Atlanta. The participant rode in the passenger seat, while the experimenter rode in the back seat to collect data. The participant's task in the field study was identical to the task performed during the freeway portion of the lab study. As in the freeway portion of the lab study, the set of objects of interest included guardrails, green exit-related signs, orange constructionrelated signs, and disabled vehicles.
During the field portion, driver perspective video and GPS position data was recorded so that the field data could be scored in the same way as the lab data. Audio was also recorded so that the participants' responses could be reconstructed if there was a loss of data. One participant in the lab study was unable to take part in the field study.
Results The first step in the laboratory data analysis procedure was to score the participants' responses. Only guardrail reports were scored; reports of the other objects of interest were not analyzed. A unique identifier was assigned to each guardrail, and the GPS data that was collected when the video was shot was used to determine the actual location of each guardrail and the earliest location from which each guardrail could be seen. (The earliest location from which each guardrail could be seen was empirically derived by
24

watching each video and identifying the first frame at which each guardrail could possibly be seen, as limited by headlight distance and physical sight obstructions.)
Each guardrail was also characterized based on the amount of visual clutter present when the guardrail first becomes visible. A three point scale was used. A clutter score of zero indicates that nothing is visible around the guardrail that would confuse the participant's perception of the guardrail or delay the participant's report of the guardrail. A clutter score of one indicates that there is some visual clutter, but nothing that would be likely to confuse the participant's perception of the guardrail or delay the participant's report of the guardrail. A clutter score of two indicates that there is some visual clutter that is likely to confuse the participant's perception of the guardrail delay or the participant's report of the guardrail.
Each participant's guardrail-related button presses were manually mapped to specific guardrail IDs, and the GPS position data associated with the button presses was used to calculate both the absolute distance from the location of the guardrail at which the participant reported perceiving the guardrail, and also the percentage of the maximum sight distance at which the participant reported perceiving the guardrail. The intent was that the percentage calculation would normalize for variations in the maximum possible sight distance among various guardrail segments.
Data analysis for the field study was similar to the analysis of the laboratory study data. The primary difference was that the participants' responses were scored against a reference detection distance rather than a maximum detection distance. The reference detection distance was determined by designating the video recorded during one participant's field trials as the reference video, and estimating the theoretical maximum viewing distance for each guardrail based on the reference video. The reference detection distances for certain guardrails were adjusted to compensate for sight distance obstructions in the reference video. Data for all participants were scored against these reference detection distances.
The main dependent variable used was distance until response. This variable represents the distance from the point where the guardrail was first visible, to the point where the guardrail was reported. Thus, a lower (smaller) distance until response indicates quicker response or higher performance. Participant responses that exceeded the determined maximum viewable distance were removed from analysis and are not reflected in the participant's mean performance score. Likewise, responses that occurred after the vehicle had passed the start of the guardrail were also not included in the distance until response dependent variable. Thus, analyses using the distance until response variable reflect performance on trials where the participant successfully reported the guardrail before reaching it. Given a constant velocity, the distance until response variable would provide the same results as a "time to report" variable.
A second dependent variable, percent reported, reflects the proportion of guardrails reported before the guardrail was reached. For this variable, guardrails that were reported
25

before the determined maximum viewable distance were still included as reported guardrails and did not count against performance.

A small number of guardrails coincided with relatively large amounts of "clutter". As these instances were not equally distributed across levels of the main independent variable of interest (delineator), they were removed prior to analysis so as not to skew the data analysis.

Lab Study Freeway Results There were nine freeway test trials, four with acrylic delineators, four with diamond I delineators, and one with diamond L delineators. [The term "acrylic" refers to the Astro Optics acrylic delineators. The terms "diamond I" and "diamond L" refer generically to all delineators using 3M Diamond Grade conspicuity sheeting, regardless of manufacturer.] Because there was only one freeway trial that used the diamond L delineator type, it was not included in any of the analyses. The level of performance on this trial was good, but without additional trials it cannot be reasonably assumed that this performance was due to the delineator type rather than some other aspect of the trial.

A computer software problem resulted in loss of lab study data for one of the participants, a young male. The analyses that follow below are based on the remaining twelve participants (six younger adults and six older adults). Table 9 shows the means and standard deviations of distance until response by age group and reflector type for freeway trials. The means were slightly lower for acrylic delineators compared to the diamond I delineators. The means were lower for younger adults compared to older adults.

Table 9: Mean Distance until Response by Age Group and Delineator Type (Standard Deviation) for Freeway Trials

Acrylic

Diamond I

Younger Adults 186.6 (56.4) 215.4 (54.3)

Older Adults 222.4 (89.7) 236.7 (48.0)

Note: Table entries are means and (standard deviations) in feet

A 2 (age) x 2 (order) x 2 (delineator) mixed ANOVA was run on distance until response, with delineator as a repeated measure. The main effects for delineator type and order were not significant. The main effect for age group approached significance F(1,8)=4.015, p=.080, with younger adults reporting guardrails somewhat earlier than older adults. No interaction effects were significant.

A similar analysis was run on the percent reported variable to look for any group differences in what was not reported. There was no effect of any kind found in the analysis, indicating that the small number of missed opportunities do not seem to be result of any of the variables measured.

26

Lab Study Non-Freeway Results Seven non-freeway test trials were analyzed, two with diamond I delineators and five with diamond L delineators. Table 10 summarizes the means and standard deviations for the non-freeway trials, broken out by delineator type and age group. The diamond L delineators tended to be detected sooner than the diamond I delineators. As with the freeway trials, younger adults tended to report the guardrails sooner than older adults.

Table 10: Mean Distance until Response by Age Group and Delineator Type for Non-Freeway Trials

Diamond I

Diamond L

Younger Adults 300.7 (64.0) 269.1 (28.7)

Older Adults 335.7 (76.3) 297.5 (25.8)

Note: Table entries are means and (standard deviations) in feet

A 2 (age) x 2 (order) x 2 (delineator) mixed ANOVA was run on distance until response, with delineator as a repeated measure. The main effects for delineator type and age group were not significant. The main effect for order approached significance F(1,8)=3.965, p=.082, with participants in the reverse order group reporting guardrails somewhat earlier than participants in the base order group. No interaction effects were significant.

The follow-up analysis on the percent reported variable showed a significant delineator x order interaction F(1,8)= 9.224, p=.016. Participants in the base order group missed a greater proportion of the diamond I guardrails than the diamond L guardrails, reporting 85.7% and 94.4%, respectively. Consequently, the main effect for delineator was also significant F(1,8)= 13.717, p=.006, with participants reporting a greater proportion of diamond L guardrails than diamond I guardrails. The reason participants failed to report more diamond I guardrails in the baseline order of trials versus the reversed order of trials is unknown.

Field Study Results

Field data are available for twelve participants (seven younger adults and five older adults; one participant in the lab study, an older adult, was unable to participate in the field study.) Field data were collected on 47 guardrails, 21 of which were included for analysis (see Table 11):

Table 11: Number of Guardrails by Type and Spacing of Delineator

Spacing 75 Feet 100 Feet

Acrylic 3 2

Diamond I 5 6

Diamond L 3 2

The remaining 26 guardrails were either not delineated, or were delineated by the older, "butterfly" style reflectors already in use. The present analysis focused only on

27

guardrails delineated by the three types of reflectors being studied (acrylic, diamond I, and diamond L).

A 2 (age) x 2 (delineator type) x 2 (delineator spacing) mixed ANOVA was run on distance until response, with reflector type and reflector spacing as repeated measures. Descriptive data is contained in Table 12.

Table 12: Mean Distance until Response by Age Group, Delineator Type, and Delineator Spacing

Acrylic

Diamond I

Diamond L

75 Foot Spacing

Older Adults

766.7 (204.9) 536.9 (103.2) 395.4 (146.2)

Younger Adults 483.5 (173.9) 325.9 (255.5) 261.5 (77.6)

100 Foot Spacing

Older Adults

556.9 (91.2) 487.1 (117.8) 129.8 (85.0)

Younger Adults 381.6 (111.9) 321.1 (106.0) 93.2 (56.4)

Note: Table entries are means and (standard deviations) in feet

The reflector type x age group interaction was significant F(2,20)=3.962, p=.036. Although older participants were slower to report guardrails than younger participants F(1,210)=6.561, p=.028, performance in the older age group was much closer to their younger counterparts when reporting diamond L delineated guardrails as opposed to those with acrylic or diamond I reflectors. The mean difference in performance (youngerolder) was 97.5 feet, 234.8 feet, and 199.0 feet, respectively. There was also an interaction of reflector type and reflector spacing, F(2,20)=8.619, p=.002. Overall, performance was better when reflectors were spaced 100 feet apart than when spaced 75 feet apart F(2,20)=78.226, p<.001, though the difference in performance was much smaller for guardrails with diamond I reflectors (33 feet) than acrylic (161.2 feet) or diamond L (207.2 feet) reflectors. The main effect of reflector type was also significant, F(2,20)=78.226, p<.001. Performance was best when reporting guardrails with diamond L reflectors, with Bonferroni-adjusted t-tests indicating a significant difference in performance over both the diamond I (p<.001), and acrylic (p<.001) reflectors. Performance was also significantly better on diamond I reflectors than acrylic reflectors (p=.001).

A 2 (age) x 2 (reflector type) x 2 (reflector spacing) mixed ANOVA was also run on the reports variable to examine differences in the proportion of guardrails reported. Guardrails with reflectors spaced at 75 foot intervals were more likely to be missed (M=90.3%, SD=.08) than guardrails with reflectors spaced at 100 foot intervals (M=96.4%, SD=.07), F(21,10)=6.690, p=.027. Combining this result with the previous analysis, it seems that performance was not only slower with 75 foot spacing, but participants were more likely to miss the guardrail altogether.

A main effect of reflector type was also found, F(2,20)=5.646, p=.011. Participants were more likely to report guardrails delineated by acrylic (M=97.6%, SD=.08) or diamond L

28

(M=95.5%, SD=.12) reflectors than those delineated by diamond I (M=86.9%, SD=.09) reflectors. Bonferroni-adjusted t-tests indicated that only the difference between the acrylic and diamond I reflectors was significant (p=.004). With regard to the previous analysis, participants were more likely to report guardrails delineated by acrylic reflectors than those delineated by diamond I reflectors, but of those reported, performance was better on the diamond I than the acrylic. The effect of age was not significant. Follow-up analyses were conducted to explore and clarify the apparent advantage for the diamond L delineators (in terms of distance until response) and for acrylic delineators (in terms of overall proportion detected). It was noted that the acrylic delineators were installed in locations that allowed long reference distances, that is, the delineators first became within view at a relatively long distance, whereas the diamond L delineators were installed in locations that did not allow long reference distances. Thus, the differences could be due to the roadway geometry rather than to fundamental differences between the delineator types. To explore this possibility, the data were divided into two reference length categories (see Table 13). The short reference category was comprised of ten guardrails with a reference distance of less than 850 feet. The long reference category was comprised of eleven guardrails with a reference distance of greater than 850 feet. Although diamond I delineated guardrails were evenly distributed across the two reference length categories, diamond L delineated guardrails were only contained in the short category, and acrylic delineated guardrails were only contained in the long category. Because reflector type was not evenly distributed across the range of the reference distance variable, separate repeated measures ANOVA analyses were run for each reference length category to test for detection differences due to reflector type while minimizing any effect of scaling. The main effect of reflector type in the short category (diamond I vs. diamond L) F(1,10)=2.781, p=.126, and the long category (diamond I vs. acrylic) F(1,10)=1.924, p=.196 were both not significant, indicating no detectible difference due to reflector type.
29

Table 13: Distribution of Delineator Type and Percent of

Guardrails Reported across the Range of Reference Distances.

Reference

Reference

Delineator Percent

Length Category Distance (Feet)

Type

Reported

Short

298

Diamond I

0.58

322

Diamond L

1.00

368

Diamond L

1.00

587

Diamond L

0.92

609

Diamond L

0.92

650

Diamond I

0.67

729

Diamond I

1.00

742

Diamond I

0.83

786

Diamond I

1.00

841

Diamond L

0.92

Long

863

Acrylic

1.00

948

Diamond I

0.83

950

Diamond I

0.75

964

Diamond I

0.92

975

Acrylic

0.83

1029

Diamond I

1.00

1065

Acrylic

1.00

1223

Acrylic

1.00

1232

Diamond I

1.00

1250

Diamond I

1.00

1536

Acrylic

1.00

Discussion Conspicuity results for all delineator types were essentially equal in the lab study. This result is not surprising, in that all methods of mounting the delineators result in a roughly equivalent orientation of the reflective material to oncoming traffic.

Younger adults tended to report delineators slightly sooner than older adults, but the differences were not significant. It is possible that the differences would be significant if a larger sample was employed.

Some differences did emerge in the field study. It appears, however, that the differences were primarily due to differences in the roadway geometry at the various installation sites rather than to any fundamental differences between the delineators. The relatively shorter distance until response for diamond L is most likely due to the fact that the roadway geometry associated with those installations did not permit long viewing distances to the diamond L delineators, thus, participants were much closer to the guardrail when they first had opportunity to detect it. Similarly, the relatively longer distance to response for acrylic is most likely due to the fact that the roadway geometries associated with these delineators afforded long viewing distances, thus, participants were much farther away from these delineators when they first had opportunity to detect it.

30

The roadway geometries for the diamond I delineators were distributed relatively evenly across the range of reference detection distances (from approximately 300 feet to 1500 feet). Statistical comparison of diamond I delineators versus diamond L delineators at reference distances of under 850 feet did not show an effect of delineator type. Similarly, statistical comparison of diamond I versus acrylic at reference distances of greater than 850 feet did not show an effect of delineator type.
It is noteworthy that participants were able to detect delineators at 100 foot spacing quite well, both in terms of overall proportion of guardrails detected and the distance until response. There is no inherent reason that delineators spaced at 100 feet would be more easily detected than delineators spaced at 75 feet; the question of interest is whether 100 foot spacing would result in satisfactory conspicuity. The present results show that the conspicuity for 100 foot spacing is quite satisfactory in the conditions studied. The apparent advantage of the 100 foot spacing over 75 foot spacing probably reflects differences in the distribution of viewing distances. The conditions included in the field study were limited to relatively benign stretches of illuminated interstate highway, and it may be inappropriate to generalize to all types of roadway. In the conditions studied, however, the 100 foot spacing produced satisfactory conspicuity, and this was especially true for the diamond L delineators.
3.4 Delineator Field Inspection The installation sites of the various guardrail delineators being studied were revisited one to two years after the initial installation was completed. The general procedure followed for the field inspections involved selecting a representative segment of guardrail for detailed inspection, and conducting a less detailed inspection of other segments of guardrail within a few miles of the representative segment. The delineators were inspected for installation failures (delineators that were missing or misaligned), reflectivity (based on the amount of accumulated road grime on the delineators), and wear (physical degradation of the delineators). Ten installation sites were inspected. Detailed field inspection results are provided in Appendix D.
The types of delineators at the sites that were inspected can be divided into four general types of installation:
Type 1: delineators attached with adhesive strips to the side of metal spacers Type 2: delineators attached with adhesive paste to the top of metal spacers Type 3: delineators attached with adhesive paste to the side of metal spacers Type 4: delineators attached with nails to top or side of wooden spacers
Two installations of Type 1 were inspected, PSS "L" delineators and PSS "I" delineators. Four installations of Type 2 were inspected, Duraflex delineators, Astro Optics Rail Rider delineators, AKT delineators, and Davidson delineators. One installation of Type 3 was inspected, Carsonite delineators. Three installations of Type 4 were inspected, Astro Optics Acrylic delineators, Astro Optics Sheeted delineators, and AKT delineators.
31

Installation Failures Installation failures are defined as cases where there was a significant loss of adhesion, allowing the delineator to shift in orientation or fall off the spacer completely. The adhesive strips used for Type 1 installations performed poorly in the installations inspected, with average failure rates around 10%, and failure rates approaching 20 to 25% on some segments. The PSS "L" delineators mostly failed by loosening and changing orientation, while the PSS "I" delineators mostly failed by loosening and falling off entirely.

One failure was observed for the Type 3 Carsonite delineator; the delineator had lost adhesion and fallen off the spacer. No failures were observed for Type 2 or 4 installations.

Reflectivity Reflectivity was assessed by making a subjective estimation of the cleanliness of typical delineators using a new delineator of the same type as a reference. The cleanliness of the delineators was described as "extremely clean," "moderately clean," "slightly dirty," "moderately dirty," or "extremely dirty." The results are shown in Table 14.

Table 14: Subjective Delineator Cleanliness Ratings

Delineator Type AKT (Type 2 Installation) AKT (Type 4 Installation) Astro Optics Acrylic Astro Optics Rail Rider Astro Optics Sheeted Carsonite Davidson Duraflex PSS "I" PSS "L"

Cleanliness Rating Slightly Dirty Moderately Clean Slightly Dirty Extremely Dirty Slightly Dirty Extremely Clean Moderately Dirty Moderately Dirty Moderately Dirty Moderately Dirty

In summary, one delineator type was rated "extremely clean," one delineator type was rated "moderately clean," three delineator types were rated "slightly dirty," four delineator types were rated "moderately dirty," and one delineator type was rated "extremely dirty." The majority of the delineator types used the same reflective sheeting, and their cleanliness ratings varied from the best to the worst possible rating with no particular pattern. The one delineator that used a different reflective material, the Astro Optics Acrylic delineator, fell in the middle of the range. There does not appear to be any significant difference in the ability of the various delineator types to resist the buildup of dirt and road grime.

32

Wear The term "wear" was used to encompass any other changes in the appearance of the installed delineators from their appearance when new. The wear observations for each delineator type are summarized in Table 15.

Table 15: Wear Observations for Each Delineator Type

Delineator Type AKT (Type 2 Installation)
AKT (Type 4 Installation)
Astro Optics Acrylic Astro Optics Rail Rider
Astro Optics Sheeted Carsonite Davidson Duraflex
PSS "I"
PSS "L"

Wear Observations Slight bending at the flexible joint of the delineator Slight bending at the flexible joint of the delineator Rusty nails Slight bending at the flexible joint of the delineator Rusty nails No wear observed No wear observed Slight bending at the flexible joint of the delineator Some curling of delineators in the long dimension Some curling of delineators in both long and short dimensions

The most serious wear observations were for the PSS "I" and PSS "L" delineators, which curled an appreciable amount. The Carsonite delineators, which were similar to the PSS "I" delineators, did not appear to have the same curling problem.

Several of the delineators of the "rail rider" style showed some slight bending, either forward or backward, in the flexible joint between the reflective surface and the mounting surface. The AKT delineators (both Type 2 and Type 4 installations), Astro Optics Rail Rider delineators, and Duraflex delineators all showed this sort of wear. Only the Davidson delineators of this style did not show appreciable bending.

The nails used to attach two of the three types of delineators that required nails showed significant rusting. While this does not affect the performance of the delineators, the use of nails and screws that are rust-resistant should be considered.

33

Maintainability The relative maintainability of the delineators can be inferred from the installation failures, cleanliness ratings and wear observations of the delineators in the field inspection tests. With the exception of PSS-I and PSS-L, which were found to have the most problems with installation and wear, no significant differences were observed among the delineators. It is important to note however that re-installation or replacement tests were not conducted and could provide more information to revise this finding.

3.5 Cost Analysis Unit cost estimates for acquiring and installing the delineator systems were obtained from GDOT. Table 16 shows the data provided by GDOT.

Table 16: Unit Costs Estimates for Acquiring and Installing Delineator Systems

Delineator System

Unit Cost Estimate ($/unit)

AKT

14.00

Astro Optics acrylic

13.28

Astro Optics hinged

13.79

Carsonite

14.00

Davidson Plastics

13.50

Duraflex

13.25

Plastic Safety System (PSS) I

14.48

Plastic Safety System (PSS) L

12.81

Safe Hit Delineator Posts

20.83

It was assumed that these data were collected over similar frames of time for delineators of comparable sizes and expected useful lives. Based on purchase and installation costs alone, it appears that the PSS-L, Duraflex, Astro Optics acrylic and Davidson Plastics are the lower cost alternatives. The cost data however do not include maintenance costs for the different systems over their lifecycles, as well as their respective expected life cycles. Such data is necessary to develop a full cost effectiveness analysis for the different alternatives.

CONCLUSIONS In general, there were no major differences among the delineator candidates based on the durability tests, installation task analysis, field inspections to observe installation failures, reflectivity and wear, laboratory and field conspicuity tests, and cost comparisons. There were minor issues worth noting however. First, the delineator installation task analysis revealed that below 35 degrees, the metal spacers had to be heated with an electric heat gun for 15-25 seconds to achieve good adhesion for the adhesive strips used by the PSS system. For the laboratory conspicuity tests on freeway sections using diamond and acrylic delineators, acrylic was found to be slightly better but not significantly. For the laboratory conspicuity tests on the non-freeway sections, where only diamond delineators were available for observation, the diamond L (larger) outperformed the diamond I (smaller) approaching significance. The diamond L delineators outperformed the other types in the field testing, but the differences were primarily due to details of the roadway geometries approaching the guardrail rather than fundamental differences between the

34

delineator types. In the field inspection tests, PSS-I and PSS-L (diamond delineators) were found to have the most problems with installation and wear. The cost comparisons, based on unit purchase and installation costs for the candidate delineator systems, indicate that the PSS-L, Duraflex, Astro Optics acrylic and Davidson Plastics have lower acquisition and installation costs. Based on the results of this study, there seem to be no major factors to distinguish among the delineators based on durability, conspicuity, maintainability, and purchase and installation cost comparisons, with the exception that the PSS system posed problems both with installation and wear. Cost comparisons based on the full lifecycle costs of the delineators and their expected useful lives will be more helpful in evaluating the alternatives based on their relative cost-effectiveness. The field conspicuity tests do show that 100 foot spacing of delineators performed quite well in the test conditions studied, and are worth further evaluation. Adopting 100 foot delineator spacing for appropriate guardrails throughout the state could result in significant cost savings without any negative impact on conspicuity. Further evaluations should be performed to determine the range of roadway conditions for which 100 foot spacing is appropriate. REFERENCES Campi Jr., John S. (1990). Guiderail Delineation. Final Report prepared for the New Jersey Department of Transportation, Division of Research and Demonstrations, DOT-T90-18. Campi, JS, Jr. (1988). Guiderail Delineation. Transportation Research Record 1160, Transportation Research Board, Washington D.C., pp. 90-96. Roberts, Arthur W (1996) Guiderail Delineation, Transportation Research Circular, No. 306, June 1986, Transportation Research Board, Washington, D.C., pp. 7-8.
35

APPENDIX A DETAILS OF DELINEATOR INSTALLATION TASKS
The tables in this appendix detail the installation tasks that were recorded for each type of delineator installation that was observed.

Duraflex on Metal

Davidson on Metal

Carsonite on Metal AKT on Metal

Move to install site

Move to install site

Move to install site

Move to install site

Place bucket with

Place bucket with

Place delineators on

Place bucket with

delineators on ground delineators on ground or spacers (one crew

delineators on ground

on guardrail

member)

Apply glue to top of spacer web

Apply adhesive to top of Retrieve delineator from Apply glue to top of

spacer web

spacer (second crew

spacer web

member)

Apply glue to side of spacer

Retrieve delineator from Apply adhesive to

bucket

delineator

Apply glue to side of spacer

Apply glue to other side Press delineator on spacer Press delineator onto side Apply glue to other side

of spacer

of spacer

of spacer

Retrieve delineator from Remove delineator

Adjust alignment of

Retrieve delineator from

bucket

delineator

bucket

Place delineator on web of Spread glue on side of

spacer

spacer web with

delineator

Retrieve clothespins from Press delineator on spacer apron pouch

Test adhesion

Spread glue on other side Apply clothes pin to

Remove delineator

of spacer web with

junction of delineator and

delineator

spacer

Retrieve bucket

Re-apply delineator

Apply clothes pin to

Spread glue on side of

junction of delineator and spacer web with

spacer

delineator

Move to next install site Test adhesion

Move to next install site Spread glue on other side

of spacer web with delineator

Retrieve bucket

Return to installation site Re-apply delineator in 24-48 hours

Move to next install site Remove clothespins

Test adhesion

Retrieve bucket

Move to next install site

36

PSS "L" on Metal

PSS "I" on Metal

Astro Optics Sheeted Astro Optics Acrylic

Secure adhesive strip

Secure adhesive strip

Move to install site

Move to install site

Remove backing
Apply adhesive strip to delineator Move to install site
Place delineators on ground (If < 35 deg) Heat spacer with heat gun Wipe spacer with cloth
Retrieve delineator

Remove backing
Apply adhesive strip to delineator Move to install site
Place delineators on ground (If < 35 deg) Heat spacer with heat gun Wipe spacer with cloth
Retrieve delineator

Place bucket with

Place bucket with

delineators on ground delineators on ground

Straighten wooden spacer Straighten wooden spacer

if required

if required

Retrieve delineator from Retrieve delineator from

bucket

bucket

Place delineator on top of Place delineator on top of

wooden spacer

wooden spacer

Retrieve nails from apron Retrieve nails from apron

pouch

pouch

Align nail in delineator Align nail in delineator

hole

hole

Drive nail

Drive nail

Remove adhesive strip Remove adhesive strip Align nail in delineator

backing

backing

hole

Apply delineator to web of Apply delineator to web of Drive nail

spacer

spacer

Retrieve hammer

Retrieve hammer

Retrieve bucket

Align nail in delineator hole Drive nail
Retrieve bucket

Tap/rub delineator to set Tap/rub delineator to set Move to next install site

tape

tape

Test adhesion by wiggling Test adhesion by wiggling

delineator

delineator

Retrieve equipment

Retrieve equipment

Move to next install site

Move to next install site Move to next install site

37

PSS "L" on Wood

PSS "I" on Wood

Move to install site

Move to install site

Place delineators on

Place delineators on

ground

ground

Straighten wooden spacer Straighten wooden spacer

if required

if required

Retrieve screws

Retrieve screws

Retrieve drill

Retrieve drill

Retrieve delineator

Retrieve delineator

Place delineator on spacer Place delineator on spacer

Align delineator on spacer Align delineator on spacer

Place screw on drill bit Place screw on drill bit

Align screw on delineator Align screw on delineator

Install screw with drill Install screw with drill

Place screw on drill bit Place screw on drill bit

Align screw on delineator Align screw on delineator

Install screw with drill Install screw with drill

Retrieve equipment

Retrieve equipment

Move to next install site Move to next install site

38

APPENDIX B DETAILS OF LABORATORY CONSPICUITY STUDY

This appendix contains detailed information about the laboratory conspicuity study.

The following abbreviations are used in the Inventory column of tables B-1 and B-2: GR Guardrail GS Green Exit Sign (freeway only) OS Orange Construction Sign RV Roadside Vehicle YS Yellow Intersection Sign (non-freeway only)

Table B-1: Freeway Trial Description for Lab Study

Trial # 1 2 3 4 5 6 7 8 9 10 11 12

Video Location I-985 SB, vicinity of mile marker 12 I-75 NB, vicinity of mile marker 306 I-75 SB, vicinity of mile marker 306 I-985 SB, vicinity of mile marker 12 I-95 SB, vicinity of mile marker 100 I-75 SB, vicinity of mile marker 306 I-575 SB, vicinity of mile marker 29 I-75 SB, vicinity of mile marker 306 I-95 SB, vicinity of mile marker 100 I-75 NB, vicinity of mile marker 306 I-575 SB, vicinity of mile marker 17 I-75 SB, vicinity of mile marker 306

Inventory N/A 2GR 3GS 6GR 2OS 4GR 1GS 1GR 2GS 1RV 1GR 2GS 2GR 1GS 1OS 1GR 6GR 3GS 2GR 1RV 1GR 1GS 3GR

Table B-2: Non-Freeway Trial Description for Lab Study

Trial # 32 10 24 12 22 14 27 16 25 13 23 15 26 11

Video Name US 341 NB, vicinity of mile marker 11 in Dodge County US 341 NB, vicinity of mile marker 11 in Dodge County GA 316, vicinity of mile marker 14 in Gwinnett County US 341 NB, vicinity of mile marker 11 in Dodge County GA 316, vicinity of mile marker 14 in Gwinnett County US 341 NB, vicinity of mile marker 11 in Dodge County GA 316, vicinity of mile marker 14 in Gwinnett County US 341 NB, vicinity of mile marker 11 in Dodge County GA 316, vicinity of mile marker 14 in Gwinnett County US 341 NB, vicinity of mile marker 11 in Dodge County GA 316, vicinity of mile marker 14 in Gwinnett County US 341 NB, vicinity of mile marker 11 in Dodge County GA 316, vicinity of mile marker 14 in Gwinnett County US 341 NB, vicinity of mile marker 11 in Dodge County

Inventory N/A 1GR 3YS 6GR 1YS 2YS 6GR 2YS 3GR 5GR 1YS 2OS 2YS 5GR 2YS 2YS 6GR 4YS 3YS 4GR 2YS 5GR 1YS

39

Table B-3: Base Order Trial Reference for Lab Study

Trial Name
Freeway Demo Trial Freeway Training Trial 1 Freeway Training Trial 2 Freeway Test Trial 1 Freeway Test Trial 2 Freeway Test Trial 3 Freeway Test Trial 4 Freeway Test Trial 5 Freeway Test Trial 6 Freeway Test Trial 7 Freeway Test Trial 8 Freeway Test Trial 9 Non-Freeway Demo Trial Non-Freeway Training Trial 1 Non-Freeway Training Trial 2 Non-Freeway Test Trial 1 Non-Freeway Test Trial 2 Non-Freeway Test Trial 3 Non-Freeway Test Trial 4 Non-Freeway Test Trial 5 Non-Freeway Test Trial 6 Non-Freeway Test Trial 7 Non-Freeway Test Trial 8 Non-Freeway Test Trial 9 Non-Freeway Test Trial 10 Non-Freeway Test Trial 11

Trial Num
Trial 1 Trial 2 Trial 3 Trial 4 Trial 5 Trial 6 Trial 7 Trial 8 Trial 9 Trial 10 Trial 11 Trial 12 Trial 13 Trial 14 Trial 15 Trial 16 Trial 17 Trial 18 Trial 19 Trial 20 Trial 21 Trial 22 Trial 23 Trial 24 Trial 25 Trial 26

Density (# Targets, Trial Time)
N/A Dense (5, 1:34) Dense (8, 2:00) Dense (5, 1:30) Sparse (4, 0:51) Sparse (3, 2:17) Dense (4, 1:43) Sparse (1, 2:05) Dense (9, 1:52) Dense (3, 1:45) Sparse (2, 1:06) Dense (3, 1:44)
N/A Sparse (4, 2:07) Dense (7, 2:08) Sparse (4, 1:06) Dense (8, 2:13) Sparse (3, 0:37) Dense (6, 1:39) Sparse (4, 2:16) Dense (8, 2:11) Sparse (2, 1:12) Dense (10, 2:13) Sparse (3, 1:20) Dense (6, 2:25) Sparse (6, 2:02)

Delineator Type
Diamond L Acrylic Acrylic Diamond L Diamond I Acrylic Diamond I Acrylic Diamond I Acrylic Diamond I Acrylic Diamond I Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I

40

Table B-4: Reverse Order Trial Reference for Lab Study

Trial Name
Freeway Demo Trial Freeway Training Trial 1 Freeway Training Trial 2 Freeway Test Trial 1 Freeway Test Trial 2 Freeway Test Trial 3 Freeway Test Trial 4 Freeway Test Trial 5 Freeway Test Trial 6 Freeway Test Trial 7 Freeway Test Trial 8 Freeway Test Trial 9 Non-Freeway Demo Trial Non-Freeway Training Trial 1 Non-Freeway Training Trial 2 Non-Freeway Test Trial 1 Non-Freeway Test Trial 2 Non-Freeway Test Trial 3 Non-Freeway Test Trial 4 Non-Freeway Test Trial 5 Non-Freeway Test Trial 6 Non-Freeway Test Trial 7 Non-Freeway Test Trial 8 Non-Freeway Test Trial 9 Non-Freeway Test Trial 10 Non-Freeway Test Trial 11

Trial Num
Trial 1 Trial 2 Trial 3 Trial 12 Trial 11 Trial 10 Trial 9 Trial 8 Trial 7 Trial 6 Trial 5 Trial 4 Trial 13 Trial 14 Trial 15 Trial 26 Trial 25 Trial 24 Trial 23 Trial 22 Trial 21 Trial 20 Trial 19 Trial 18 Trial 17 Trial 16

Density (# Targets, Trial Time)
N/A Dense (5, 1:34) Dense (8, 2:00) Dense (3, 1:44) Sparse (2, 1:06) Dense (3, 1:45) Dense (9, 1:52) Sparse (1, 2:05) Dense (4, 1:43) Sparse (3, 2:17) Sparse (4, 0:51) Dense (5, 1:30)
N/A Sparse (4, 2:07) Dense (7, 2:08) Sparse (6, 2:02) Dense (6, 2:25) Sparse (3, 1:20) Dense (10, 2:13) Sparse (2, 1:12) Dense (8, 2:11) Sparse (4, 2:16) Dense (6, 1:39) Sparse (3, 0:37) Dense (8, 2:13) Sparse (4, 1:06)

Delineator Type
Diamond L Acrylic Acrylic Acrylic Diamond I Acrylic Diamond I Acrylic Diamond I Acrylic Diamond I Diamond L Diamond I Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I Diamond L Diamond I

41

APPENDIX C DETAILS OF FIELD CONSPICUITY STUDY

This appendix contains detailed information about the field conspicuity study.

The eastbound test segment (Trial 29 for identification purposes) extended from the Glenwood Ave. interchange (Exit 61B) to the Columbia Dr. interchange (Exit 66). The westbound test segment (Trial 30 for identification purposes) extended from the Columbia Dr. interchange (Exit 66) to the Glenwood Ave. interchange (Exit 61B).

The selection of the guardrails that were delineated was based on a number of factors, including roadway and guardrail geometry, and the available inventory of the various types of delineators.

The following tables give details on all of the right shoulder guardrails in the test segments.

Table C-1: I-20 Eastbound Guardrail Details

Guardrail ID
GR_29_01 GR_29_02 GR_29_03 GR_29_04 GR_29_05 GR_29_06 GR_29_07 GR_29_08 GR_29_09 GR_29_10 GR_29_11 GR_29_12 GR_29_13

Delineator Type
Diamond I Undelineated Undelineated Undelineated Undelineated Undelineated Undelineated Diamond L Acrylic Undelineated Diamond I Butterfly Diamond L

GR_29_14 GR_29_15 GR_29_16 GR_29_17 GR_29_18 GR_29_19 GR_29_20 GR_29_21 GR_29_22 GR_29_23 GR_29_24

Undelineated Diamond I Diamond L Acrylic Diamond I Butterfly Diamond I Diamond I Undelineated Undelineated Undelineated

Number and Spacing
8 @ 100' N/A N/A N/A N/A N/A N/A 10 @ 75' 14 @ 100' N/A 3 @ 75' N/A 3 @ 75'
N/A 9 @ 100' 9 @ 100' 2 @ 100' 12 @ 100' N/A 9 @ 75' 12 @ 75' N/A N/A N/A

Notes
Two of three delineators fell off, so this segment was excluded from analysis.

42

Table C-2: I-20 Westbound Guardrail Details

Guardrail ID
GR_30_01 GR_30_02 GR_30_03 GR_30_04 GR_30_05

Delineator Type
Diamond I Butterfly Undelineated Diamond I

GR_30_06 GR_30_07 GR_30_08 GR_30_09 GR_30_10 GR_30_11 GR_30_12 GR_30_13 GR_30_14 GR_30_15 GR_30_16 GR_30_17 GR_30_18 GR_30_19 GR_30_20 GR_30_21 GR_30_22 GR_30_23

Diamond I Undelineated Acrylic Butterfly Butterfly Undelineated Butterfly Undelineated Undelineated Acrylic Diamond I Diamond L Undelineated Undelineated Diamond L Acrylic Diamond I Undelineated

Number and Spacing
4 @ 100' N/A N/A 4 @ 75'
7 @ 75' N/A 2 @ 75' N/A N/A N/A N/A N/A N/A 3 @ 75' 8 @ 100' 3 @ 100' N/A N/A 6 @ 75' 11 @ 75' 9 @ 100' N/A

Notes
A group of reflectors was erroneously listed as a guardrail during initial analysis. The ID was left in as a placeholder to avoid renumbering subsequent guardrails.

43

APPENDIX D DETAILS OF DELINEATOR FIELD INSPECTION
This appendix contains the detailed reports produced by Richard Gordon in August and September of 2003 for each delineator field inspection.
Field Inspection Report PSS "L" Delineators Hwy 316 East, vicinity mile 14 Installed: 13 November 2001 Inspected: 27 August 2003 Method: Adhesive strip to metal spacer.
General I selected one 920' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the Plastic Safety Systems "L" type delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were generally present and in working condition along the Hwy 316 installation site. Delineators showed evidence of exposure and road-grime, but for the most part appeared to be functional. The PSS "L" delineators were mounted using an adhesive strip, which in several cases showed signs of failure.
Installation Failures: On the particular segment I selected, the failure rate for the delineators was 25%, however, the failure rate in general appears to be much lower. I estimate that the failure rate in general approaches approximately 10%. For this type of mounting, the failure was due to the adhesive strip losing adhesion and allowing the delineator to "slide" downward (see Figures 1-2). In each case, the delineator was still attached, but in the wrong orientation. The adhesive strip was only sticking on a small percentage of the surface (see Fig. 3).
44

Fig. 1

Fig. 2

Fig. 3

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "moderately dirty". A grayish film and in several instances, road tar had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was degraded to some degree. In Figure 4, the left side of the delineator was cleaned with a damp cloth, the right side shows the level of accumulated grime. The delineators were also compared to a clean delineator of the same type (see Fig. 5).

Fig. 4

Fig. 5

Wear: All of the delineators inspected showed signs of exposure. The delineators tend to curl on their vertical and horizontal axes causing the reflective surface to cant downward approximately 3-5 (see Fig. 6-8). The observation angle relative to the surface of the delineator is changed slightly and would negatively affect perceived luminance.

45

Fig. 6

Fig. 7

Fig. 8

46

Field Inspection Report PSS "I" Delineators I-85 North, vicinity mile 117 Installed: 27 November 2001 Inspected: 03 September 2003 Method: Adhesive strip to metal spacer.
General I selected a 1200' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the Plastic Safety Systems "I" type delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were generally present and in working condition along the I-85 installation site. Delineators showed evidence of exposure and road-grime, but for the most part appeared to be functional. The PSS "I" delineators were mounted using an adhesive strip, which in several cases showed signs of failure.
Installation Failures: One the particular segment I selected, the failure rate for the delineators was 19%, however, the failure rate in general appears to be slightly lower. I estimate that the failure rate in general approaches approximately 10%. For this type of mounting, the failure was due to the adhesive strip losing adhesion and allowing the delineator to fall off (see Figures 1-2). In each case, the delineator had fallen completely off the metal spacer. The adhesive strip was only sticking on a small percentage of the surface (see Fig. 3).

Fig. 1

Fig. 2

Fig. 3

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "moderately dirty". A grayish film and in several instances, road tar had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was degraded to some degree. In Figure 4, the top half of the delineator was cleaned with a damp cloth, the

47

bottom half shows the level of accumulated grime. The delineators were also compared to a clean delineator of the same type (see Fig. 5).

Fig. 4

Fig. 5

Wear: There were minimal signs of exposure damage on most of the delineators at this site. Only a few of the delineators inspected showed signs of damage. The damaged delineators tend to curl on their horizontal axes causing the reflective surface to cant downward approximately 3-5 (see Fig. 6-8).

Fig. 6

Fig. 7

Fig. 8

48

Field Inspection Report Duraflex Delineators I-85 North, vicinity mile 148 Installed: 28 November 2001 Inspected: 03 September 2003 Method: Adhesive paste on top of metal spacer.
General I selected a 620' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the Duraflex type delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were all present and in working condition along the I-85 installation site. Delineators showed evidence of exposure and road-grime, but all appeared to be functional. The Duraflex delineators were mounted using three beads of adhesive glue along the top of the metal spacer.
Installation Failures: One the particular segment I selected, there were no failures (see Figures 1-2). I observed no failure cases within the installation area. In each case, the delineator was secure and upright.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "moderately dirty". A grayish film had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was degraded to some degree. In Figure 3, the top half of the delineator was cleaned with a damp cloth, the bottom half shows the level of accumulated grime. The delineators were also compared to a clean delineator of the same type (see Fig. 4).

49

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on most of the delineators at this site. Only a very few of the delineators inspected showed signs of damage. The damaged delineators tend to bend forward at the flexible joint causing the reflective surface to cant downward approximately 3-5 (see Fig. 5-6).

Fig. 5

Fig. 6

50

Field Inspection Report Astro Optics Acrylic Delineators I-75 North, vicinity mile 306 Installed: 11 April 2002 Inspected 08 September 2003 Method: Nails on wooden spacer.
General I selected a 1100' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the Astro Optics Acrylic type delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were all present and in working condition along the I-75 installation site. Delineators showed evidence of exposure and road-grime, but all appeared to be functional. The Astro Optics Acrylic delineators were mounted using two nails, one on either side at opposite corners onto the top of the wooden spacer.
Installation Failures: On the particular segment I selected, there were no failures (see Figures 1-2). I observed no failure cases within the installation area. In each case, the delineator was secure and upright.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "slightly dirty". A grayish film had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was degraded to some degree. In Figure 3, the left half of the delineator was cleaned with a damp cloth, the right half shows the level of accumulated grime. The delineators were also compared to a clean delineator of the same type (see Fig. 4).

51

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on most of the delineators at this site. Of note, the nails used to attach the delineators were showing signs of rust from exposure (see Fig. 5). None of the delineators inspected showed signs of damage, but several of the wooden spacers had become misaligned. The misaligned spacers tend to bend forward or backward causing the reflective surface to cant. (see Fig. 6).

Fig. 5

Fig. 6

52

Field Inspection Report Astro Optics "Sheeted" Delineators I-575 South, vicinity mile 28 Installed: 5 March 2002 Inspected 08 September 2003 Method: Nails on top of wooden spacer.
General I selected a 600' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the Astro Optics Sheeted type delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were all present and in working condition along the I-575 installation site. Delineators showed evidence of exposure and road-grime, but all appeared to be functional. The Astro Optics delineators were mounted using two nails, one on either side at opposite corners onto the top of the wooden spacer.
Installation Failures: On the particular segment I selected, there were no failures (see Figures 1-2). I observed no failure cases within the installation area. In each case, the delineator was secure and upright.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "slightly dirty". A grayish film had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was degraded to some degree. In Figure 3, the top half of the delineator was cleaned with a damp cloth, the bottom half shows the level of accumulated grime. The delineators were also compared to a clean delineator with the same type reflective sheeting (see Fig. 4).

53

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on most of the delineators at this site. Of note, the nails used to attach the delineators were showing signs of rust from exposure (see Fig. 5). None of the delineators inspected showed signs of damage, but several of the wooden spacers had become misaligned. The misaligned spacers tend to bend forward or backward causing the reflective surface to cant. (see Fig. 6).

Fig. 5

Fig. 6

54

Field Inspection Report Astro Optics "Rail rider" Delineators I-575 North, vicinity mile 7 Installed: Unknown (March 2002?) Inspected: 08 September 2003 Method: Adhesive paste on top of metal spacer.
General During inspection of other designated installation sites, I discovered several miles of guardrail delineated with an Astro Optics product not previously studied. I selected a 350' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the Astro Optics "Rail Rider" type delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were all present and in working condition along the I-575 installation site. Delineators showed evidence of exposure and road-grime, but all appeared to be functional. The Astro Optics delineators were mounted using adhesive glue along the top of the metal spacer.
Installation Failures: On the particular segment I selected, there were no failures (see Figures 1-2). I observed no failure cases within the installation area. In each case, the delineator was secure and upright.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "extremely dirty". A grayish film had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was degraded to some degree. In Figure 3, the top half of the delineator was cleaned with a damp cloth, the bottom half shows the level of accumulated

55

grime. The delineators were also compared to a clean delineator of the same type (see Fig. 4).

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on most of the delineators at this site. One of the delineators inspected showed signs of damage. The delineator bend rearward at the flexible joint causing the reflective surface to cant upward approximately 3-5 (see Fig. 5). In another case, the metal spacer had shifted, resulting in alignment problems similar to those found at other sites (see Fig. 6)

Fig. 5

Fig. 6

56

Field Inspection Report AKT Delineators I-95 North, vicinity mile 100 Installed: 12 December 2001 Inspected: 21 September 2003 Method: Adhesive paste on top of metal spacer.
General I selected an approximately 600' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the AKT delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were all present and in working condition along the I-95 installation site. Delineators showed evidence of exposure and road-grime, but all appeared to be functional. The AKT delineators were mounted using three beads of adhesive glue along the top and sides of the metal spacer.
Installation Failures: On the segment I selected, there were no failures (see Figures 1-2). I observed no failure cases within the installation area. In each case, the delineators were secure and upright.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "slightly dirty". A grayish film had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was degraded to some degree. In Figure 3, the top half of the delineator was cleaned with a damp cloth, the bottom half shows the level of accumulated grime. The delineators were also compared to a clean delineator of the same type (see Fig. 4).

57

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on most of the delineators at this site. A few of the delineators inspected showed signs of wear. The affected delineators bend backward at the flexible joint causing the reflective surface to cant upward approximately 2-3 (see Fig. 5 - 7).

Fig. 5

Fig. 6

Fig. 7

58

Field Inspection Report AKT Delineators I-16 East, vicinity mile 114 Installed: Unknown; Probable installation about December 2001 Inspected: 21 September 2003 Method: Nails on top of a wooden spacer.
General This delineated segment was not filmed during installation, but I felt the different method of installation warranted an inspection. I selected an approximately 250' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the AKT delineator. I also conducted a general visual inspection of other segments within an area of approximately ten miles of the chosen segment. In the I-16 installation area, delineated segments were generally very short and several miles apart.
Overall Assessment: Delineators were all present and in working condition along the I-16 installation site. Delineators showed slight evidence of exposure and road-grime, and all appeared to be functional. The AKT delineators were mounted using two nails through the double flange skirt into the side of a wooden spacer.
Installation Failures: On the segment I selected, there were no failures (see Figures 1-2). I observed no failure cases within the installation area. In each case, the delineator was secure and upright.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately clean", and "extremely dirty", these delineators are subjectively rated as "moderately clean". A grayish film had accumulated on the reflective surface. The reflective surface could still be easily seen, but reflectivity was slightly degraded. In Figure 3, the top half of the delineator was cleaned with a damp cloth, the bottom half shows the level of accumulated grime. The delineators were also compared to a clean delineator with the same retroreflective sheeting (see Fig. 4).

59

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on the delineators at this site. The mounting technique using the split-flange delineators caused the delineator's reflective surface to cant downward approximately 3-5 (see Fig. 5 - 6).

Fig. 5

Fig. 6

60

Field Inspection Report Carsonite Delineators GA 341 East, vicinity mile 111 Installed: 16 January 2002 Inspected: 21 September 2003 Method: Adhesive paste on side of metal spacer.
General I selected two approximately 300' segments of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the AKT delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were all present and in working condition along the GA 341 installation site. Delineators showed minimal evidence of exposure and road-grime, and appeared to be functional. The Carsonite delineators were mounted using an S-shaped bead of glue on the side of the metal spacer. Delineators were temporarily secured in place with two wooden clothespins while the adhesive set.
Installation Failures: On the segment I selected, there were no failures (see Figures 1-2). I observed one failure case within the installation area. In that case, the delineator had lost adhesion and fallen off the spacer. All other delineators within the inspection area were secure and upright. Of note, however, the clothes pins used to secure them to the metal spacer temporarily had never been recovered and were still on the delineators.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated very little dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately clean", and "extremely dirty", these delineators are subjectively rated as "extremely clean". The reflective surface could still be easily seen, and reflectivity was not degraded significantly. In Figure 3, the top half of the delineator was cleaned with a damp cloth. A

61

comparison between the top and bottom halves shows the lack of accumulated roadgrime. The delineators were also compared to a clean delineator with the same retroreflective sheeting (see Fig. 4).

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on the delineators at this site.

62

Field Inspection Report Davidson Plastics Delineators I-85 South, vicinity mile 56 Installed: 19 February 2002 Inspected: 26 September 2003 Method: Adhesive paste on top of metal spacer.
General I selected an approximately 500' segment of guardrail for a detailed inspection for signs of wear, installation failures, and dirtiness of the Davidson Plastic type delineator. I also conducted a general visual inspection of other segments within an area of approximately two miles of the chosen segment.
Overall Assessment: Delineators were all present and in working condition along the I-85 installation site. Delineators showed evidence of exposure and road-grime, but all appeared to be functional. The Davidson Plastics delineators were mounted using a single thick bead of adhesive glue along the top of the metal spacer.
Installation Failures: On the particular segment I selected, there were no failures (see Figures 1-2). I observed no failure cases within the installation area. In each case, the delineator was secure and upright.

Fig. 1

Fig. 2

Reflectivity: The delineators had accumulated dirt during the time since installation. On a scale of "extremely clean", "moderately clean", "slightly dirty", "moderately dirty", and "extremely dirty", these delineators are subjectively rated as "moderately dirty". A grayish film had accumulated on the reflective surface. The reflective surface could still be seen, but reflectivity was degraded to some degree. In Figure 3, the top half of the delineator was cleaned with a damp cloth, the bottom half shows the level of accumulated grime. The delineators were also compared to a clean delineator of the same type (see Fig. 4).

63

Fig. 3

Fig. 4

Wear: There were minimal signs of exposure damage on most of the delineators at this site. None of the delineators inspected showed signs of damage (see Fig. 5-6).

Fig. 5

Fig. 6

64