TESTING AGENCY VIEWS
John Rohde
University of Nebraska-Lincoln
Introduction
Compliance testing by the criteria outlined in NCHRP 350 over the past three to four years has yielded experiences that are important to consider in evaluating the efficacy of the document. Development and crash testing efforts at the University of Nebraska-Lincoln’s MidWest Roadside Safety facility have shown the need for both modification and clarification of current criteria. This report is primarily focused on the issues of construction practices and test evaluation, but will consider these in terms of certification assessment of hardware during the approval process.
This discussion has been divided into three topics: 1.) soil specification (with consideration of wood specification), 2.) construction tolerance (compliance and installation), and 3.) test evaluation and laboratory certification. Natural bias of the author (geotechnical engineering) will weight the bulk of the discussion to the first topic, but it should be noted that this does not necessarily relate to the priority of the various topics.
Soil Specification
During the development of NCHRP 350 it was identified that maintaining consistent test conditions both seasonally and between regions of the country was imperative to the certification process. To this end, the authors selected standard soils for compliance testing and further specified installation geometrics (underground). The general recommendation for a standard soil were materials meeting AASHTO M 147-65 (1990) specifications, grades A or B. This specification developed as a basis for base course material for highway construction is a well-graded crushed limestone. Further the authors state that "if a feature is likely to be installed in a soil that could be expected to degrade its performance, testing in one or more of the special soil described in Section 2.2.1.2. of NCHRP 350 may be appropriate." This referenced section discusses options for testing in saturated, frozen and "weak" soils.
In Section 2.2.1 of NCHRP 350 the methodology of placing the test article is described, with the intention that native soils would not effect the response of the embedded device. Criteria for placement of "standard soil" around the test article are either assurance that native soils do not influence the test or a fixed criteria of hole depth of 1.2 times embedment and lateral extent of 1.3 times embedment. Current research at UNL to define the soil response to dynamic impact has shown, that with the current soil specification, this lateral extent is a minimum if post rotation without the influence of surrounding soils is desired. The cost implications of this criteria are great, in the case of a standard guardrail post these criteria define an excavation 4 1/3 feet in depth and 4 2/3 feet in diameter. Compacted fill for this post would require about 5 tons of rock. The diameter of this excavation precludes the use of augers, requiring a backhoe to excavate line posts in a system. While it is acknowledged that this procedure would only be necessary on posts within a system expected to experience significant rotation during impact, our research has shown that surrounding soils, with posts set in 3 foot diameter excavations can significantly effect post response with as little as three inches of post movement at the ground line. A line drawing of the typical failure surface is shown in Figure 1. The shear strength of a particular soil controls the shape of the surface, with high shear strength materials such as 350 strong soil developing the largest envelope.
To quantify the effects of strong and weak soils on the load/deformation response of the impacted post, we performed a set of bogie vehicle impacts on wood and steel posts embedded in a series of soils. Shown in Figure 2, is the bogie mounted accelerometer response versus time for wood post impacts. The embedment conditions are outlined in the Figure’s legend. The sand utilized for these tests meets AASHTO M6-87 criteria, the NCHRP 350 designated "weak" soil. The cohesive soil is a low-plastic clayey silt (low moisture is about 4 percent below optimum and high moisture reflects nearly saturated conditions), and the crushed limestone is a well graded and meets AASHTO M 147-65 (1990). There are very significant differences in the response between the various soils. As shown in Figure 2, if the soil’s peak strength exceeds the wood posts bending strength, the post fractures.
Energy from the impinging vehicle in the case of a longitudinal barrier is dissipated through the post/soil and rail. If the energy absorption of the post/soil system is low, either due to low soil strength or post fracture, stress in the rail is increased. Because the strength of wood posts is also a variable within the test matrix, varying by a factor of 3 between various grades of wood, the effect of soil on the performance of the system is dependent on wood quality. For instance, if a longitudinal barrier placed in a strong NCHRP 350 soil incorporates some high-strength posts, the combination can generate localized pocketing and snagging problem. The influence of this performance in systems such as transitions is dependent on geometry. On a recent system with quarter post spacing, loads on the posts in the nested thrie-beam system were on the order of 9 kips at post bolt height. This is a load insufficient to cause wood fracture, but as shown by the bogie testing the load/deformation response of the posts in soils other than the crushed stone utilized in this test could have had a significant effect on post response.
In many instances, such as the previously mentioned transition test, the designer has the option of selecting an appropriate soil. Generally, posts placed adjacent to a bridge rail will be founded in good quality subgrade materials, probably best characterized by the low moisture cohesive soil in Figure 2. Comparing the response of posts founded in cohesive and 350 crushed limestone shows that for the load ranges resulting from the specified impact condition, there would be little difference in the performance of the posts.
Based on the average dynamic fracture strength of various classes of wood tested at UNL and the soil response shown in Figure 2, Table 1 was prepared to illustrate the resulting peak load experienced by the system. For the section of the table that is shaded, impact of sufficient magnitude will cause post failure. In the remainder of the combinations, impact would cause the soil to fail allowing the post to rotate. Considering Figure 2, these two scenarios are radically different in terms of energy absorption of the post/soil interaction.
The consideration of steel posts produces a different set of effects from varying soil strength. Bogie accelerometer response from impacts with steel posts set into the same set of soils previously described for wood posts is shown in Figure 3. When the yield stress is exceeded in steel posts, a plastic hinge is formed rather than the fracture discussed for wood posts. This hinge typically forms between 6 and 15 inches beneath the ground surface, increasing in depth as soil stiffness decreases. Also shown in Figure 3 is the result of W6X16 post that was tested to produce a soil failure in the NCHRP 350 strong soil, rather than plastic failure in the post. This heavy post has the same width and depth as a W6X9, but heavier flanges. Contrasting response of wood and steel posts two trends are obvious, first in soil material with a high angle of internal friction (particle interlock) steel posts have a notably higher soil resistance. This is true even though the steel posts are almost 2 inches thinner than wood posts. This phenomenon is postulated to be due to load carried on the backside of the flange opposite of the direction of travel. In general though, the performance of NCHRP 350 weak soil, which was significantly lower that of a nearly saturated cohesive soil for wood posts, is almost the same for steel. Additionally, the peak strength for a steel post that does not fail due to plastic yield is significantly higher in the NCHRP 350 strong soil.
So where does all of this lead in examining criteria in NCHRP 350? The current soils and embedment criteria were selected with sound engineering and economic rational. Subsequent testing has shown that the current criteria can be both very conservative and very non-conservative depending on the appliance being tested. Selection of weak soils in many instances is so conservative as to preclude its use.
In conclusion, soil properties influence various appurtenances differently. As in the determination of impact conditions, soil must be approached from some rational basis for different classes of appurtenances. While soil classifications could be used as a basis for defining this performance, it may be more rational to specify both a classification and performance, so site soils at the various testing facilities could be quantified and utilized if they are appropriate. This issue must be addressed in those classes of appurtenances that are effected by soil interaction, without this effort we cannot generalize the performance during certification testing to installation conditions.
Construction Tolerances
In any construction practice, it is necessary to have both a dimension and a tolerance to both allow the constructor a limit and to assure the owner of a quality product. To date, the roadside safety industry has not provided this information to the constructor and consumer. In the consideration of NCHRP 350, it is critical that tolerancing issues be addressed. Some items to be considered include:
This issue needs to be addressed by hardware developers in the development process, making tolerances part of the package submitted for FHWA approval. Who better than the developers could define critical tolerances for a given system? If tolerances were a part of the approval process, then these same criteria could be applied to installations for compliance testing. This issue carries some very significant tort liability issues, but it is imperative that it be addressed to protect developers, fabricators, installers and owners.
Test Evaluation And Laboratory Certification
A significant amount of expertise is required in running, instrumenting, and analyzing the results of compliance tests. While FHWA has been making significant efforts into standardization of some of these processes, there remains a need to insure each of the testing houses is utilizing sound practices. There are two potential avenues to assuring this performance: (1) a certification procedure for each testing house, and (2) in depth review of practices as a measure of performance. Both of these approaches could entail significant expense, so this issue must be approached rationally. On the basis of cost, a system of self and mutual regulation seems to be the most prudent approach. In this system, reporting and documentation of standard practices would be required by FHWA, and a program of periodic review would be developed. This review process would have to depend on funding from FHWA or other sources. The ultimate result of this program, must be confidence on the part of FHWA that critical test parameters are accurately measured and reported.
Conclusions
In the test performance and compliance criteria described in NCHRP 350 provide a good foundation for the assurance of safe and economical roadside safety hardware. In consideration of updating these criteria, consideration of the effects of soil on the performance of various appurtenance is critical. Efforts to define salient parameters, with approaches analogous to those used to develop impact criteria, would provide rational test conditions and assurance of adequate performance in the field. The revision must address the issue of tolerancing both test installations and field installations. This is currently a shortcoming in our practice that causes innumerable tort liability problems, as well as potential risks to the motoring public and difficulties in quality assurance programs for owners. Finally, we need to develop a method of assuring that procedures utilized in compliance testing are both reasonably precise and accurate across the various testing facilities in the nation.
List Of Figures:
Figure 1: Diagram of failure plane in soil during post impact.
Figure 2: Accelerometer plots for bogie impact with wood posts in various soils.
Figure 3: Accelerometer plots for bogie impact with steel posts in various soils
Table 1: Controlling ultimate strength of wood post/soil response considering average dynamic post strength defined by grading and various soils.
PEAK LOAD EXPERIENCED BY THE POST AT RAIL HEIGHT
|
Soil Type |
#1 DOUG FIR |
#2 SYP |
#1 SYP |
#2D SYP |
#1D SYP |
DS-65 SYP |
|
Wet Clayey Silt |
6.1 |
6.1 |
6.1 |
6.1 |
6.1 |
6.1 |
|
Sand |
6.7 |
6.7 |
6.7 |
6.7 |
6.7 |
6.7 |
|
Dry Clayey Silt |
12.6 |
12.6 |
12.1 |
13 |
13 |
13 |
|
Optimum Clayey Silt |
12.6 |
12.6 |
12.1 |
13.3 |
14 |
14 |
|
350 Spec Crushed Limestone |
12.6 |
12.6 |
12.1 |
13.3 |
14.6 |
22 |
History
Guidelines for testing roadside appurtenances started back in 1962 with a one-page document—Highway Research Circular 482 entitled Proposed Full-Scale Testing Procedures for Guardrails. This included four specifications on test article installation, one size vehicle (standard design, 4000Å200 lb., center of gravity at 21 in), six test conditions, and three evaluation criteria. In 1974, NCHRP Report 153 Recommended Procedures for Vehicle Crash Testing of Highway Appurtenances was published. It provided the first complete test matrix. Parameters to be measured were specified along with methods and limiting values, and limited guidance on reporting formats was included. Transportation Research Circular 191 Recommended Procedures for Vehicle Crash Testing of Highway Appurtenances was published in 1978 provided limited interim changes to NCHRP 153 and then there was extensive revision in 1981 with the publication of NCHRP Report 230 Recommended Procedures for the Safety Performance Evaluation of Highway Appurtenances. The adoption of NCHRP Report 350 Recommended Procedures for the Safety Performance Evaluation of Highway Features in 1993 provided even more extensive changes including multiple services levels, inclusion of many other roadside features, change in the test vehicles, more and different test conditions, etc. In general, NCHRP Report 350 is considered a very adequate document with specific requirements and detailed guidance which allows testing agencies to perform valid unbiased tests.
Tolerances & Accuracy For Test Installations
The basic philosophy that should be followed is: any parameter that significantly affects test results should be measured and controlled within appropriate limits. This philosophy should be followed in construction of prototype test articles and in performance of tests. Although field construction is not a part of this discussion, this same philosophy should be followed in field construction.
Since more crash test requirements mean higher cost and more time, unnecessary requirements in testing or in field installations should not be imposed. Tolerances and accuracy for prototype test installations should be compatible with those for field installations. Both should be sufficiently well controlled to produce consistent, reproducible performance of the test article. It could be further argue that any safety device that is too sensitive to construction variations should not be used in the field. More emphasis should be made in reporting any special care in fabrication and installation that will probably not happen in production manufacturing to assure that it does happen. All special treatment in construction or installation of the test article or in fabrication of parts should be reported as specified in NCHRP 350.
Documentation
Material and geometric properties of the test article and values of test parameters should be sufficiently well documented to allow another test engineer to duplicate or reproduce a test. Whether or not all the documentation should be contained in the test report is point for discussion. An alternative to reporting all detailed documentation is to maintain all details in the testing agency files and report more general information, such as is now reported. Many take the position that testing reports do not now include enough detailed documentation of properties of the test article.
If more documentation is to be included, the testing agency, the sponsor, and the supplier/manufacturer must all participate in obtaining and/or supplying the information. For example: If a W-beam guardrail section is to be used, the following questions could be addressed:
Section 2.3 of NCHRP Report 350 generally requires that details such as those just listed be documented through reports, certification and supplemental testing. It, at least, implies that all details should be established and reported. That is not now always being done. All parties involved (testing agencies, sponsors, manufacturers/suppliers) need to take more responsibility in following the requirements now in NCHRP 350. Some information is best supplied by manufacturers or suppliers. Testing agencies should devote more effort to documenting test installations; however, the competitive bidding requirements of transportation agencies often discourage the extra effort. This point needs to be emphasized. Complete and detailed documentation of all material and geometric properties of components of test articles would drastically increase the costs of full-scale crash testing. Testing costs are already so high, that these aspects receive limited attention. Significant increases in cost would further limit the ability to develop and qualify safety hardware.
Specifications in 350 for photo targets, markings, and documentary still photos seem to be adequate, yet allow some variation as evidenced in films and photos from various testing agencies. Specified targets would enhance ability to analyze and compare tests from the different testing agencies.
Soil Conditions
The type of soil and its condition, such as moisture content, have a very strong influence on behavior of some safety devices in crashes. Furthermore, type and condition of soil in the field can vary enormously -- from saturated "muck" to solid rock. For example, if a device functions by a post moving in the soil, the strength of soil is extremely important. The question becomes, what strength of soil should be used in testing? If a post needs to break for the device to function properly, strength of soil (in test and in field) must be above some value to assume post breaking. NCHRP 350 states that "Product developers and user agencies should assess the potential sensitivity of a feature to foundation conditions, and if the feature is likely to be installed in a soil that could be expected to degrade its performance, testing in one or more of the special soils may be appropriate." Other important questions include: Is moisture content important during testing? Should this be reported? How should measurements be made? Test data show that the strength of NCHRP standard soil falls drastically if the moisture content increases above 8%.
Test Vehicles
NCHRP 350's age limit requirement on test vehicles seems to be appropriate in terms of test vehicles being representative of those on the road. Of course, these older vehicles cannot do the impossible job of predicting the future, but the age is more representative of what is on the road. When choosing actual test vehicles, the make and model that is most representative of that particular class of vehicle should be used. NCHRP 350 gives specifications on the test vehicles. However, in the case of the 2000P there are differing opinions on what is actually represented on the road. Should the test specification describe in detail and specify make and model of vehicles that are acceptable as test vehicles? If descriptions and specifications are too strict, availability is reduced and costs are increased. If descriptions and specifications are too relaxed, test results could be influenced by choice of vehicle. Some think that the current specifications allow that to happen. NCHRP 350 specifies:
The certification label usually found on the driver’s door states original equipment, including tires and tire loading information for a vehicle. Testing agencies restore the test vehicle to original specifications as stated on the Certification label if any standard equipment has been modified. Still, there has been quite a bit of opposing opinion on whether to use pickups with 6 lug versus 8 lugs as well as Ford versus GM in regard to suspension characteristics. As efforts to update NCHRP Report 350 proceed, it is an appropriate time to check to see which vehicles ARE MOST representative on the road and make specific recommendations or requirements to be reflected in the update.
Occupant Compartment Deformation And Intrusion
There has been some discussion and disagreement on what deformation or amount of intrusion could cause "serious injury." Individuals even within testing agencies have differing ideas of what is "serious." Someone may say loss of limb is considered serious, while another would argue that even so they would still be alive.
This NCHRP 350 criterion remains subjective as it states: "This factor must be assessed in large part by the judgment of the test agency and the user agency, or both. ...In the absence of a widely accepted measure of risks associated with deformations or intrusions, it is essential that adequate documentation in the form of photographs and measurements of occupant compartment damage be made and reported. ... Until an acceptable methodology is developed," NCHRP 350 recommends the use of the OCDI in hopes that it "will permit some degree of quantification of occupant compartment damage. As experience is gained with its use, definitive acceptance criteria may be established in the future."
Now should be considered "the future" and an opportune time to consider what could be done to improve this criterion. This could include the consideration of specific guidance for pass/fail decisions with specific definitions (what is "serious") linked to certain or specific measurements. If nothing else, consider a project (sponsored by whom?) to investigate or compare the OCDI or some other measurement to "serious injury."
Test Article Selection
A higher level of detailed documentation would be beneficial in controlling the "Sunday Sample Syndrome" in test article selection. For example: In the selection of posts, would the installer of guardrails go through all wood posts available to be sure he got good posts throughout the entire installation? In the selection of vehicles would the choice between a Ford or GM product be made to influence the outcome of the test? And if this choice would make a difference, should that installation be accepted if it is that sensitive? What about field devices not being manufactured or installed as they were for testing? (i.e., the bolts provided by the sponsor for sign testing being different than those provided for field installations.) What about changes in design after testing? Should the procedure require that production samples be tested instead of prototype components?
Lab Certification
Under NCHRP 350, data system accuracies must conform to the requirements of SAE J211, Instrumentation for Impact Test. As stated in the scope of J211: "This SAE Recommended Practice outlines a series of performance recommendations which concern the whole data channel. . . It is not intended that each recommendation be taken in a literal sense, as necessitating a single test to demonstrate that the recommendation is met. Rather, it is intended that any agency proposing to conduct tests to this practice should be able to demonstrate that if such a single test could be and would be carried out, then their equipment would meet the recommendation. This demonstration should be undertaken on the basis of reasonable deductions from evidence in their possession, such as the results of partial tests."
Since 1987 NHTSA has been providing a signal waveform generator (SWG), which supplies precision test waveforms, to each of its crash and sled test contractors. This is used to evaluate the facility’s crash test data acquisition system (DAS) performance for compliance to SAE J211 requirements for FMVSS 208 type testing. The introduction of the NHTSA SWG Hardware Upgrade Operator’s Manual states that "In particular, the FHWA is putting forth a prerequisite that its contractor test facilities must pass NHTSA’s SWG performance evaluation as part of its certification requirement for testing."
On the surface, this requirement may sound logical and reasonable. But, if one looks at data system requirements for FMVSS 208 and NCHRP 350 they will see several large differences. First of all, FMVSS 208 uses dummy data to determine Head Injury Criterion (HIC). These measurements necessitate a wide bandwidth system to pass the Class 1000 dummy data. It is this Class 1000 channel that the NHTSA Signal Wave Generator was designed to test. NCHRP 350, on the other hand, typically requires only Class 180 data (integration for velocity or displacement, J211). Data channels designed to meet the requirement of J211, Class 180, such as telemetry, will not accurately pass the Class 1000 test signals of the NHTSA SWG.
If the signal wave generator is to be used for evaluating NCHRP 350 type data systems, it should be designed specifically for that purpose. A redesign of the SWG could be accomplished by simply revising the software. A series of waveforms could be developed that would test the amplitude accuracy and frequency response of a Class 180 system. The sample head impact wave should be replaced by a vehicle acceleration waveform for evaluation of NCHRP 350 type data systems.
For over twenty-five years TTI and other testing agencies have been successfully performing crash tests for the safety performance evaluation of highway hardware using accelerometers and other types of transducers. The accuracy of these measurements relied on internal quality control measures which required calibration at planned intervals by comparison to or measurement against national standards, natural physical constants, consensus standards, or by ratio type measurements using self calibration techniques. Supporting documentation relative to traceability would then be filed and available for examination upon request. These methods provided a simple, cost effective way of maintaining accuracy, not only on yearly intervals but from test to test.
Closing Remarks
A great deal of consideration should be given to exactly what is required from testing agencies and what needs to be done to achieve that goal. The above mentioned methods are achievable at reasonable costs to manage the programs and meet all the requirements. If a certification program is adopted, costs will increased and these costs will be added on to the price of each crash test. Certification should not be needed for a good reputable testing agency. However, reality may be that the "customers" position (e.g., European Union) on the matter may dictate that all test agencies become certified. It is recommended that lab certification not be imposed on testing agencies without a very careful study of the consequences.
Testing procedures for roadside safety features is only one part of the integrated activities that result in a safety device being designed, developed, tested, evaluated, approved as acceptable highway hardware and implemented for use on the national highway system. These activities are intended to improve the safety of the national highway system and reduce injuries and fatalities. The information to be presented here is one perspective from a hardware industry representative that is involved in several of the elemental parts of these integrated activities. Thus, it is necessary to discuss several of the related elements to be able to describe the strengths and weaknesses of the current programs and suggest improvements that could be made in a rewrite of the National Cooperative Highway Research Program Report Number 350 (NCHRP 350) (1).
INTRODUCTION
The hardware industry is made up of commercial companies that are in business to provide a reasonable return on the investment made by their owners and stockholders. If they are not able to do this, the investors will go elsewhere and the business will cease to exist. Thus, the companies must focus on providing quality products that can be sold at reasonable prices. This means that the products must be able to be developed and marketed at a reasonable expense and the sales volume must be sufficient to justify the costs to design, develop, test, set up manufacture and market the product.
The types of issues encountered by the hardware industry that reduce its ability to be successful include the following:
Addressing the concerns of the hardware industry requires looking at the whole design, development, testing, evaluation, approval, implementation and use process. The remainder of this presentation will focus on the key issues that affect these concerns and then break the issues down into specific categories.
KEY ISSUES
The vehicle fleet underwent major changes in the time period after the publication of the National Cooperative Highway Research Program Report No. 230 (NCHRP 230) (2). Vans mini-vans, pick-ups, four-wheel drive and utility vehicles became a significant (greater than 25%) part of the passenger vehicle fleet. These vehicles reacted very different to roadside safety features than the 1800C and 4500S test vehicles specified in NCHRP 230.
One of the major changes in the testing criteria that occurred when the NCHRP 350 was published was the substitution of the 2000P test vehicle (2000kg, ¾ ton pick-up truck) for the 4500S (4500 pound [2040kg] sedan) test vehicle used in NCHRP 230. The 2000P vehicle was to act as a surrogate for the high center of mass vehicles that had become a significant part of the vehicle fleet. However, this vehicle has characteristics that cause it to behave differently than previous test vehicles when widely accepted hardware is evaluated. This issue should be considered in a rewrite of NCHRP 350. The types of issues that also need to be addressed in a rewrite NCHRP-350 includes the fact that very narrow testing and evaluation considerations may lead to very broad approvals and vague evaluation factors can lead to inconsistent conclusions. Thus, testing and demonstrating acceptable performance of a breakaway feature in strong soil may lead to approval for all soils even though the device may not perform acceptably in soft soils. Also, criteria such as "moderate roll, pitch and yaw" (NCHRP-350) are fairly vague and may lead to inconsistent judgment calls by the developers, the approval agencies and even the end users.
The objectives of the developer and hardware producer must focus on minimizing costs. However, these objectives should not conflict with the need to properly develop, test, evaluate and publish the characteristics of the system. These characteristics must include limitations on performance, where and how the product should/ could be applied and how it would be expected to function under those conditions. Otherwise, serious product liability problems could result.
The approval agencies and specifiers need to focus on ensuring that the characteristics of the systems being considered match the characteristics needed for specific site conditions where application is being considered. This may require publishing information such as that provided in "Design Construction and Maintenance of Highway Safety Features and Appurtenances - Users Guide" (3) to provide better guidance to specifiers and end users. This guidance could include categories of devices based on specific performance parameters. It may also suggest mechanisms to enforce standards that are established.
The key issues break down into the categories of design and development, testing and evaluating, evaluation and approval, and application and use. In the subsequent sections these categories of potential improvement that could be made in a rewrite to NCHRP 350 will be discussed.
DESIGN AND DEVELOPMENT ISSUES
The design and development process should not start with compliance testing. Again, in an attempt to reduce development costs, developers may not start with good design, analysis and developmental testing programs. This can result in a product being approved for use with little design analysis work and only three or four full scale impact tests. How this device may perform in the field, the limitations of the product and applications information may not be understood nor transferred to the end user.
The designer/developer should establish the performance objectives of the device up front in the development process. The objectives should include anticipated performance, specific targeted applications, acceptable and unacceptable site conditions, anticipated limitations of the device, and estimates of sensitivity of system performance to site parameters. This will better allow the consideration of developmental and compliance test matrices before this expensive phase of the project begins. It is further suggested that these objectives and the proposed compliance test matrix be shared with and approved by the approval agency before compliance testing begins.
The designer/developer should compile the material specifications, drawings, and any other instructions that will be used by the testing agency prior to compliance testing. This will allow the testing agency to ensure compliance of the tested system to that documented by the designer/developer.
TESTING AND EVALUATION ISSUES
Testing and evaluation issues will be addressed in the categories of test vehicle selection, test article compliance to design documents, testing laboratory, evaluation factors and test matrix issues. A great deal of the improvements that could be made in a rewrite to NCHRP 350 are in these sections.
Test Vehicle Selection
The issues in this area include the appropriateness of the test vehicle for what is being evaluated, the choice of a specific model of vehicle on vehicle sensitive tests and verification of actual vehicle characteristics compared to specified characteristics.
The test vehicle used to evaluate containment of a longitudinal barrier should be able to load the barrier in a manner consistent with containment evaluation. If the vehicle would be expected to vault or be tripped and roll, another vehicle should be considered for the test. Thus, full consideration should be given to the appropriateness of the test vehicle for evaluating the characteristics of the system.
The type of test vehicle may be appropriate for a particular test but a specific model may be chosen that would result in either passing or failing the evaluation criteria for a given test. Because the bumper heights, center of mass heights and suspension are quite different between ¾ ton pickup trucks (Figure 1) (4), the choice of a specific model can affect the results depending on whether you are doing a redirective test, a frontal terminal or crash cushion test, etc. Therefore, it is proposed that a specific vehicle recommendation be made or that the vehicle parameter tolerances be tightened and enforced to avoid this problem.
FIGURE 1. VARIATIONS IN CRITICAL PROPERTIES OF SPECIFIC MODELS WITHIN A VEHICLE TYPE.
The testing agency should also validate that vehicle parameters are within the tolerances in the criteria.
Test Article Compliance To Design Documents
The recommendation was made earlier that the designer/developer provide the testing agency with specifications, design documents, etc., for the test article. These documents should be used to ensure the test article is in compliance with the design values. Otherwise there is no "chain of custody" that will exist between what was designed, tested, approved and used in the field.
Testing Laboratory
The key issues under this category includes laboratory accreditation, mounting of data measurement and collection devices, and changes to the test article during the compliance testing process.
The hardware agency would prefer to have a situation where compliance testing completed at one laboratory would be recognized by all compliance testing laboratories and approval agencies. This is currently not the case because there are no mechanisms in place to allow either recognition or acceptance. The development of a testing laboratory accreditation program for all approved compliance testing laboratories would help solve this problem. This program would also make it easier to get acceptance of tests done by accredited laboratories done in different countries (i.e., U.S., Europe, Asia).
It is proposed that the accreditation program include quality programs, auditing procedures (internal and external), an enforcement body, etc. We should strive to harmonize the programs with an international body to help get mutual recognition of approved laboratories.
The data measurement devices are sometimes mounted to thin sheet metal areas on the floor of the test vehicle. If this sheet metal is impacted by the engine/transmission or if the sheet metal buckles during a test, the measurements will be altered. Standardization and improvements in the mounting of data measurement devices should be considered in the rewrite.
The testing laboratory should not allow the test article to be changed during a compliance testing program without proper documentation. Approval of the changes by the ultimate approval agency is also recommended.
Evaluation Factors
Several of the evaluation factors in NCHRP 350 are vague and need to be made more specific in a rewrite. The vagueness results in inconsistent conclusions relative to the evaluation of the crashworthiness of safety features. Thus, a feature that is considered acceptable to one group may not get approval, while another feature that is considered unacceptable may be approved.
One of the vague evaluation factors is post impact vehicle trajectory and intrusion into adjacent traffic lanes. The evaluation criteria is that any intrusion into adjacent traffic lanes should be "minimized" so as not to present a hazard to adjacent traffic. Based on approvals, it would appear that intrusion of up to 20 meters is acceptable. The approach proposed in the CEN prEN 1317 document where the vehicle must stay within the confines of an imaginary box should be considered as an alternative.
Detached elements should not pose a "hazard" to other motorists, pedestrian or personnel in a work zone. What constitutes a hazard is not defined in NCHRP 350. Wheels and suspension elements from the test vehicles and sizable pieces of the test article may be in adjacent traffic lanes or in areas where pedestrians or work zone personnel would be expected to be and the device may be approved. On the other hand, there may be no way to prevent elements of the vehicle from becoming detached in certain tests (2000P redirect tests). This issue also needs to be addressed in the rewrite.
The deformation of the occupant compartment is measured and reported in NCHRP 350. No specific criteria nor guidance are given as to what is or is not acceptable. With the types of occupant compartment deformation seen in some tests Figure 2 and Figure 3 (5, 6), more specific criteria is needed.
FIGURE 2. OCCUPANT COMPARTMENT DEFORMATION- FRONTAL IMPACT
FIGURE 3. OCCUPANT COMPARTMENT DEFORMATION- ANGLED IMPACT.
The roll, pitch and yaw of the test vehicle need to be "acceptable" as stated in NCHRP 350. What is "acceptable"? Again, very little guidance is given and thus, we have a vague criteria that leads to inconsistent conclusions.
Environmental factors such as high/low temperature, UV degradation, ozone effects, corrosion, moisture, etc. are not considered as evaluation criteria in NCHRP 350. However, once approved, the safety feature will be placed in these environmental conditions and be expected to perform acceptably. Again, more guidance is needed in this area in a rewrite.
Test Matrix
The compliance test matrix for a safety feature should be developed around the design objectives of the safety feature, the proposed applications and what approval is specifically being requested. If a safety feature is developed only for unidirectional traffic conditions, it would not make a lot of sense to conduct wrong way redirect tests. On the other hand, if it is anticipated that the feature will be placed in two way traffic, that same test should be mandatory. A similar argument can be made for issues such as the size of the clear zone required for a safety feature, transition design options, redirective/non-redirective, gating/ non-gating, etc.
The issue of soil conditions used in the compliance testing and how performance could be affected by soil conditions where the safety feature is expected to be applied also need to be addressed. A detailed assessment of the soil conditions present during testing should be required. Further, an engineering assessment should be made to determine if the performance of the safety feature would be adversely affected under different soil conditions. At this point, if performance is expected to be adversely affected, the feature should be tested in the worse soil condition or the approval should be restricted to those conditions where the feature would be expected to perform acceptably.
Many of the impacts into safety features are from a non-tracking vehicle and may include side impacts. These issues are not addressed in NCHRP 350 and should be considered in a rewrite. In several product categories in NCHRP 350, optional tests were proposed. These tests were intended to gather information in order to support improved testing criteria for the future. However, the optional tests are not being run or reported and thus, the rewrite will have little support in these areas. The tests should be required to be conducted but it should be optional whether the approval agency requires a specific evaluation criteria to be passed.
EVALUATION AND APPROVAL ISSUES
The evaluation and approval process has historically lumped devices with widely varying characteristics into broad categories of safety features (i.e., crash cushions, end terminals and longitudinal barriers).
An example of how this causes problems is when an end terminal with a large clear zone requirement is approved for use. The approval does not indicate the clear zone requirement and typically the published information on the system does not give any guidance. Thus, the end terminal is applied to sites that have small or no clear zone and injuries or fatalities result. Other approved products with very small clear zone requirements could have been used and provided performance more applicable to that site.
The rewrite should cause the characterization of systems to be narrowed in order to bring focus on the site specific characteristics of the safety features. This could cause a reduction in the number of compliance tests for some products that have narrow applications and an increase in the number of tests required if very broad approvals are being requested.
The "Users Guide" contains valuable information on the site characteristics that should be considered and the system characteristics that should be published. It would be appropriate to consider this information in categorizing systems being evaluated in the rewrite.
The approvals that result from an evaluation of a safety feature should be very specific on where and how the feature should and should not be used. Performance and site limitations should be described in sufficient detail to give the end user guidance on how to properly use the feature.
APPLICATION AND USE ISSUES
The highway designer or safety engineer needs to have access to accurate, thorough and consistent information on the physical and performance characteristics of approved safety features. Having access to this information will greatly help reduce the misapplication of safety features, improve the ability to have the features properly installed and maintained and result in better utilization of limited funds.
The published characteristics of approved safety features should include selection guidelines, applicability to site specific characteristics and limitations from both a performance and installation perspective.
CONCLUSION
The NCHRP 350 document represented a significant improvement from the previous guidelines in NCHRP 230. The result of adopting NCHRP 350 testing and evaluation criteria was the development of new and the refinement of existing highway safety features that made an improvement to the safety provided to the motoring public. A great deal of money and valuable resources were used to make the changes but this hardware industry representative believes the expense was worthwhile.
There is a need to make further improvements to the whole process as described in this paper. These improvements include a rewrite of NCHRP 350 and changes in the areas discussed. However, we should not loose focus on the main objective stated in the beginning -- Saving lives and reducing injuries through good design, development, testing, evaluation, approval, implementation and use of highway safety features. It takes a lot of teamwork.
REFERENCES
"Recommended Procedures for the Safety Performance Evaluation of Highway Features," H. Ross, D. Sicking, R. Zimmer, and J. Michie, NCHRP Report 350, 1993.
"Recommended Procedures for the Safety Performance Evaluation of Highway Features," H. Ross, D. Sicking, R. Zimmer, and J. Michie, NCHRP Report 230, March, 1981.
"Design Construction and Maintenance of Highway Safety Features and Appurtenances - Users Guide", B. Bowman, J. Bryden, D. Gripne, Auburn University - Highway Research Center, April, 1997. "Safety Appurtenance Design and Vehicle Characteristics", B. Stephens, TRC 453, February, 1996. "Occupant Compartment Deformation, ¾ ton pick up - NCHRP350, Test 3- 44, Angled Impact for Non-Redirective Crash Cushions". Crash Test Report 20.8.1996, Autobahnmeisterei Graz, conducted by Institute for Mechanics, Technical University of Graz, Austria, October, 1996.
We all have become familiar with "NCHRP 350" during the past few years. Its history briefly includes the following highlights:
With the growing number of products available, we have been surprised at the reluctance of some states to implement improved safety features before the 1998 deadline, especially when some of the new products are offered at the same price the products certified to the less stringent requirements of NCHRP 230.. However, many progressive states have implemented "350" specifications as required, and the list is growing.
Also, we have come to realize the great difference between successful laboratory testing and successful in-service performance of hardware on the roadsides. Unfortunately, few states have active and regular disciplined in-service performance evaluation programs, so the data coming back on actual performance of products is sketchy at best. We believe a return to the requirements limiting use of a new product during an experimental phase as a good idea, and should be re-adopted.
CONCLUSION
Since "350" has demonstrated that it is substantially more demanding in its safety requirements than its predecessor NCHRP 230, at this time, we do not see any need to make its requirements even more demanding than they are at present. However, since it has been adequately demonstrated during the past several years that "350" compliant products can be designed, successfully tested, certified and successfully used on our roadways, we certainly do not favor any kind of reduction in safety requirements (or testing criteria) of NCHRP Report "350". Let's be sure not to lower the safety bar already established for our nation's highways. In the interest of public safety, we would suggest that more stringent side impact protection in safety devices be developed and required for any modified "350" specification as it is considered for revision. In summary, we recommend no reductions in safety requirements and the addition of the requirement of side impact protection.
HARDWARE INDUSTRY PERSPECTIVE
Dean C. Alberson
Safety Quest, Inc.
General Thoughts
What has NCHRP 350 done for you? That all depends on your point of view. If you are a manufacturer, you've seen large quantities of green go the wrong way on the balance sheet, at least in the short run. In the bigger picture, if you have successfully complied with NCHRP 350, you look around and the number of competing products is somewhat limited and you are enjoying a favorable market condition. If you are a test facility employee, you've seen an increase of the green stuff flowing in as a result of the manufacturers testing their products to meet NCHRP 350. As a bonus, you've seen many exciting tests because what used to be a test faced with confidence is now a nerve racking, gut wrenching experience for the manufacturers.
If you're a Federal Highway employee that has the dubious task of evaluating the NCHRP 350 tests for the manufacturers, done by the test facilities, you're under the constant fire of competing manufacturers crying foul and throwing sand in the sand box. If you're a citizen that drives a Ford Festiva, you're clueless. If you're a citizen that drives a pickup or sport utility vehicle, you're enjoying a roadside that is becoming more forgiving than ever in history. Your chances of surviving serious impacts with roadside appurtenances is improving daily.
Pull Back on the Standards?
Are the test criteria too severe? I mean, how many people drive 3/4 ton pickups? That all depends on your point of view. If you're a small business enterprise that has relied on non-proprietary terminals and crash cushions to complete your product lines, the tests are probably too severe. Your product line is shrinking and will continue to shrink. You are being forced to purchase proprietary items to package with your highway projects. Your margins are down because you are paying for the increased requirements for highway safety products. If you are a large company that has the financial clout to test to the NCHRP 350 standards and the perseverance to see it through to the end, you've decided the criteria is not too severe and the risk is worth the reward. If you are a test facility employee or a Federal Highway employee, you recognize the enhancement to safety that has been brought about by NCHRP 350. You may have even been involved in the establishment of the standards. If you're a road user, the evaluation criteria are definitely not too severe. As a taxpayer you might question the associated costs for a time but will eventually agree the standards are "OK," especially if someone dear to you has attempted to test those standards personally.
Obviously the number of tests has increased and will continue to increase with the passage of time. Are we beyond the point of good economy? We understand the concept of benefit/cost ratios and until a better mousetrap comes along, that will be how we measure if we are spending the tax dollars wisely. Can the evaluation criteria be lessened? It is probably a little late for that. The number of products successfully surviving the NCHRP 350 test gauntlet is increasing almost daily. To step backwards would be just that, a step backwards. Clearly products can meet the criteria, so why would we want to drop our standards.
Improvements?
Where are we deficient in our evaluation of current products? As many are aware, we have a limited specification for soil types. What we don't have is compaction and moisture content specified or at the very least recorded. This information clearly has an impact on the performance of many systems, particularly guardrails. We need to be at least reporting this information.
Wood is an interesting product. Does the performance change with age? I'll bet individuals could even select specific posts for crash tests that can significantly affect the test performance. If one selects a post with high ring density, no knots, and full dimensions for use in a strength test, does this accurately reflect what will be installed in the field? Or perhaps a post with low ring density, lots of knots and small dimensions for use when frangibility is desirable? That narrow window of performance has been widened a bit. What about plastics and steel? Reporting in these areas can be improved and should be outlined in the performance specification. We currently measure occupant compartment deformation. How much is acceptable? Perhaps a quantification on a percent basis would be suitable. The standard is somewhat subjective at this point.
Off-tracking vehicles are quite common in run-off-road accidents. The safety community has indicated their awareness of this problem with side impact tests. This area warrants more investigation. The level of performance that can be achieved must be identified and then a test to evaluate this will probably be needed. A full side impact will be the most severe case and will therefore be conservative in nature.
Approval Process
NCHRP 350 is a performance specification. In a society that is going to largely proprietary and significantly different items, a performance specification is the only viable method for evaluating the respective devices. Federal Highway's acceptance of the respective roadside safety devices in largely based on NCHRP 350, but not solely guided by it. At FHWA's discretion, manufacturers may and often are required to run non-standard tests before the "FHWA Approval Letter" will be issued. Is this good practice? Perhaps. Perhaps not. Does this practice assure equality across the spectrum? Is there a process by which this opinion can be appealed? Furthermore, if NCHRP 350 has been accepted by the governing agency as the standard and a device has successfully met the criteria set forth in the standard, can the agency arbitrarily place further restrictions on the device? If devices are perceived to slip though cracks in the standard, revisions to the standard would be the best policy. Certifying a device with questionable performance is the last thing a manufacturer should want to do in our litigious society. Therefore, accept or reject devices based on a standard or modify the standard to accurately reflect the views of the majority of the experts in the field.
Conclusion
To date, NCHRP 350 is the most comprehensive document we have to guide us in our evaluation of roadside safety features. Have we arrived? Certainly not! Are we headed in the right direction? The responses are varied but generally, yes. Let us continue to move forward and increase the safety on all roadways.
BREAKOUT GROUP SUMMARY
Group A. Relevancy of Evaluation Procedures
It has been suggested that any effort to update NCHRP Report 350 that may lead to new performance requirements should be shown to be "relevant." The breakout group debated this issue for a full day, considering many factors in attempting to first define "relevancy' and then determine how to establish it. Cost, crash severity, benefit/cost ratios, and in-service experience were discussed as means to define relevancy. The perspective of the DOTs, hardware developers, and testing agencies were considered, but after a day's effort, no consensus on relevancy was achieved. It was noted that threats to credibility include variability in test results, differences in testing items, soil conditions, equipment, the limited number of tests that have to be passed, and the lack of in-service data to validate that the tests are indicative of field performance. Methods proposed to determine relevancy included the use of crash data, biomedical measures, or jury systems, but again no consensus was achieved.
Group B. Test Matrices and Conditions
Group B was charged with reviewing the test matrices and conditions in NCHRP report 350 to identify updating needs. The research needs identified included by priority level:
Priority A
Priority B
Priority C
Group C. Full-Scale Crash Testing
Group C was charged with reviewing the updating needs for full-scale crash testing. They noted a strong interest in reducing crash test costs, since not all hardware gets extensive application. The needs identified included by priority level:
Priority A
Priority B
Priority C
Group D. Evaluation Criteria
Group D was charged with reviewing the evaluation criteria specified in NCHRP Report 350. The needs identified included by priority level:
Priority A
Group E. Processes for Certification
Group E was charged with reviewing the process for getting new hardware certified through crash tests. The needs identified included:
Appendix A Participant List "Future Directions in Roadside Safety," Woods Hole, MA, July 1997
Safety Quest, Inc.
505 University Drive East, Suite 701
College Station, TX 77840
409-268-2235
Richard Albin
Washington State DOT
P.O. Box 47329
Olympia, WA 98501-7329
360-705-7269
Andrew Artar
Gregory Highway Products, Inc.
4100 13th Street, SW
Canton, OH 44708
330-477-4800
Nicholas Artimovich
Federal Highway Administration
Nassif Building, HNG 14
Washington, DC 20590
202-366-1331
Joseph Batley
Buffalo Specialty Products, Inc.
894 Marcon Blvd., Suite 100
Allentown, PA 18103
610-266-4133
Nabih E. Bedewi
FHWA/NHTSA NCAC
20101 Academic Way
Ashburn, VA 22011
703-729-8361
Roger Bligh
Texas Transportation Institute
College Station, TX 77843
409-845-4377
Rodney Boyd
Trinity Industries
302 Steel Drive
Elizabethtown, KY 42701
502-769-3354
Maurice Bronstad
Dynatech Engineering
8023 Vantage Suite 900
San Antonio, TX 78230
210-525-8862
William Bryson
Bryson Products, Inc.
2240 Schoenersville Road Suite 201
Bethlehem, PA 18017
610-954-8210
Lance Bullard
Safety Quest, Inc.
505 University Drive East, Suite 701
College Station, TX 77840
409-268-2235
Monique Burns
Connecticut Department of Transportation
2800 Berlin Turnpike
Newington, CT 06131
860-594-3292
John F. Carney, III
Worcester Polytechnic Institute
100 Institute Road
Worcester, MA 01609-2280
508-831-5222
Gerald Cawley
Highway Safety Corp.
P.O. Box 358
Glastonbury, CT 06033-0358
800-544-4929
Al Clifford
Universal Engineered Products
10 Ferry Wharf Building
Newburyport, MA
508-463-9771
Lincoln C. Cobb
Energy Absorption Systems, Inc.
3617 Cincinnati Avenue
Rocklin, CA 95765
916-645-8181
Mack Christensen
Utah Department of Transportation
4501 South 2700 West Box 143200
Salt Lake City, UT 84114-3200
801-965-4264
Subhasis Chatterjee
Vanderbilt University
Box 1831 Station B
Nashville, TN 37235
Novice Cole
Lu, Inc.
P.O. Box 607
Kingston Springs, TN 37082
615-952-5501
Gary Consolazio
Rutgers University, CE Dept.
P.O. Box 909
Piscataway, NJ 08855
732-445-4414
Cary Corkin
The Entwistle Company
Bigelow Sstreet
Hudson, MA 01741
508-481-4000
Daniel Dearasaugh
Transportation Research Board
2101 Constitution Avenue NW
Washington, DC 20418
202-334-2955
John W. Duckett
1100 William
Carson City, NV 89701
702-885-2500
John Durkos
Energy Absorption Systems
One East Wacker Drive, Suite 3000
Chicago, IL 60601
312-467-6750
Owen Denman
Energy Absorption Systems, Inc.
3617 Cincinnati Avenue
Rocklin, CA 95765
916-645-8181
Arthur Dinitz
Transpo Industries, Inc.
20 Jones Street
New Rochelle, NY 10801
914-636-1000
Mike G. Dreznes
Energy Absorption Systems, Inc.
One East Wacker Drive
Chicago, IL 60601
312-467-6750
Michael Essex
Energy Absorption Systems, Inc.
One East Wacker Drive, Suite 3000
Chicago, IL 60601
312-467-6750
Vittorio Giavotto
Technical University of Milan
32 Piazza leonardo da Vinci
Milano, 20133 ITALY
392-2399-4030
Don Graham
Trinity Industries, Inc.
2525 Stemmons
Dallas, TX 75207
214-689-0545
703-285-2508
James Hatton
Federal Highway Administration
Nassif Building HNG-14
Washington, DC 20590
202-366-1329
Herwig Hausdorf
Fed. Ministry for Economic Affairs
/Stubenring 1
Vienna, AUSTRIA A-1010
43-1-71100-5512
Rex Hedges
MIRA
Watling Street
Nunerton,Warwickshire CV10 OTU
ENGLAND
Kathy Hoffman
Roadway Safety Foundation
1776 Massachusetts Avenue, NW, Suite 500
Washington, DC 20036
202-857-1200
Deborah Irvin
Transportation Research Board
2101 Constitution Avenue, NW
Washington DC 20418
202-334-3237
Peter Johnsen
Road Directorate
Niels Juels Gade 13
Copenhagen, Denmark1059
Don Johnson
Syro, Inc.
2525 Stemmons
Dallas, TX 75207
214-589-8138
Michael Kempen
Roadway International
46-04 245th Street
Douglaston, NY 11362
718-229-0046
James Kennedy
Battelle Memorial Institute
505 King Avenue
Columbua, OH 43201
614-424-4831
Glenn Korfhage
Minnesota DOT
MS 696 Transportation Bldg
St. Paul, MN 55155
612-296-4859
Kaddo Kothman
Interstate Steel
P.O. Box 3241
Big Spring, TX 79721
915-263-3725
Brett Lewis
APTEK
1257 Lake Plaza Drive
Colorado Springs, CO 80906
719-576-8100
David R. Lewis
David R. Lewis Group, Inc.
263 Bradford Drive
Canfield, OH 44406
330-533-6490
John Logan
JL Associates
P.O. Box 15189
Seattle, WA 98115-0189
206-440-0400
King Mak
TTI, Safety Division
Texas A&M University
College Station, TX 77843-3135
409-845-6385
Charles McDevitt
FHWA, HSR-20
6300 Georgetown Pike
McLean, VA 22101-2296
703-285-2418
Malcolm McDonald
Transport & Road Research lab
Crowthorne, Berkshire, ENGLAND
44-1344-770506
Richard McGinnis
Department of Civil Engineering
Burknell University
Lewisburg, PA 17837
717-524-1127
Dhafaer Marzougui
NCAC/GWU
20101 Academic Way
Ashburn, VA 22011
703-729-8361
Fredrick Mauer
Marion Steel Co.
4 Caswell Drive
Greenland, NH 03840
603-430-9350
Leonard Meczkowski
FHWA
6300 Georgetown Pike
McLean, VA 20101
703-285-2420
Wanda Menges
Texas Transportation Institute
TAMU Riverside Campus
College Station, TX 77843-3135
409-845-6157
John Melvin
G M R&D Center, MC480-103-001
30500 Mound Road
Warren, MI 48090-9055
810-986-1698
Robert Mileti
Roadway Safety Systems, Inc.
P.O. Box 1165
Andover, MA 01810
508-470-3656
Mark Mondo
Mondo Polymer Technologies
P.O. Box 250
Reno, OH 45773
614-376-9396
Daniel Mushett
Buffalo Specialty Products, Inc.
P.O. Box 178
Sunbright, TN 37872
423-628-2530
Ken Opiela
Transportation Research Board
2101 Constitution Avenue, NW
Washington, DC 20418
202-334-3237
Terry Otterness
Arizona DOT
205 S. 17th Avenue
Phoenix, AZ 85007
602-255-7341
Arthur Perkins
New York State DOT
1220 Washington Avenue Bldg. 5-408
Al any, NY 12232
518-457-4501
Richard Peter
California Department of Transportation
5900 Folsom Blvd. MS-5
Sacramento, CA 95819
916-227-7257
Richard Powers
Federal Highway Administration
Nassif Building HNG-14
Washington, DC 20590
202-366-1320
Elizabeth Ray
Conference Coordinator
2607 Friendship Street
Iowa City, IA 52245
319-341-9653
Malcolm Ray
Center for Computer-Aided Desigin
University of Iowa
Iowa City, IA 52242
319-384-0523
John Reid
University of Nebraska-Lincoln
104N WSEC (0656)
Lincoln, NE 68588
402-472-3084
John Rohde
University of Nebraska-Lincoln
W348 Nebraska Hall
Lincoln, NE 68588
402-472-8807
Hayes Ross
Texas Transportation Institute
Texas A&M University
College Station, TX 77843
409-845-4368
Barry Rosson
UNL-Midwest Roadside Safety Facility
W348 Nebraska Hall
Lincoln, NE 68588
402-472-8773
Payam Rowhani
Boeing Commercial Airplane Group
P.O. Box 355
Renton, WA 98057
425-965-0330
Randa Radwan Samaha
NHTSA, NRD-11
400 Seventh Street, SW
Washington, DC 20590
202-366-4707
Charles Sanders
Illinois DOT
2300 South Dirksen Parkway
Springfield, IL 62764
217-785-0720
Dale Schauer
Lawrence Livermore National Laboratories
7000 East Avenue L-140
Livermore, CA 94550
510-423-0303
Ronald Seitz
Kansas Department of Transportation
Decking State Office Bldg, 915 Harrison
Topeka, KS 66612
913-296-3890
Dean Sicking
University of Nebraska-Lincoln
W348 Nebraska Hall, P.O. Box 880531
Lincoln, NE 68588
402-472-9332
Robert Slagter
Anro Supply Company
1322 142nd Avenue
Wayland, MI 49348
616-877-4723
Ken Stack
GM -Vehicle Safety Center
30200 Mound Road
Warren, MI 48090
810-947-1766
Barry Stephens
Energy Absorption Systems, Inc.
3617 Cincinnati Avenue
Rocklin, CA 95765
916-645-8181
Alrik Svenson
FHWA/MITECH
6300 Georgetown Pike
McLean, VA 22101
703-285-2616
David Tarrant
Mondo Polymer Technologies
P. O. Box 250
Reno, OH 45773
614-376-9676
Harry Taylor, HHS-10
Federal Highway Administration
400 Seventh Street, SW
Washington, DC 20590
202-366-2175
Frank Tokarz
Lawrence Livermore National Lab
7700 East Avenue, L-644
Livermore, CA 94551
510-423-3459
Rod Troutbeck
Queensland University of Technology
P.O. Box 2434
Brisbane, QLD 4001 AUSTRALIA
Thomas Turbell
VTI
S-581 95 LINKOPING
SWEDEN
46-13-204369
Michael Vail
Buffalo Specialty Products
190 Tunnel Road
Vernon, CT 06066
860-895-6424
Jay Walter
Roadway Safety Service, Inc.
80 Remington Blvd.
Ronkonkoma, NY 11779
516-588-6200
Jerry Wekezer
FAMU-FSU, College of Engineering
2525 Pottsdamer Street
Tallahassee, FL 32310-6046
850-487-6143
William Wendling
Consultant Engineer
HCR 71 Box 119
Camdenton, MO 65020
573-873-3353
Richard Young
Mississippi Dept. of Transportation
P.O. Box 1850
Jackson, MS 39215-1850
601-359-7007
Richard Zimmer
Texas Transportation Institute
Texas A&M Univ. MS 3135
College Station, TX 77843-3135
409-845-6385