Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
102 Chapter 5 Conclusions and Future Research 5.1 Conclusions Throughout the research period, compatibility between vehicles and roadside hardware systems was investigated. Findings of the project indicated that the performance of roadside systems during typical crash configurations varied greatly according to the overall characteristics of the impacting vehicle class. However, the relationship between detailed vehicle characteristics and adverse crash outcomes could not be easily linked using currently available data. 5.1.1 Data Analysis A review of NASS/CDS and FARS data was conducted to evaluate the compatibility of existing roadside systems and an evolving vehicle fleet. A historical review of fatalities during roadside hardware impacts was conducted using crash data from 1990-2000. Although NCHRP 350 criteria has lead to significant changes in roadside systems designs since it's publication, many roadside systems that remained in service during these years did not vary significantly in design. Using the assumptions and adjustments outlined in the following paragraphs, the variation in injury and fatality outcomes during impacts with roadside devices could be attributed to evolving vehicle characteristics. Adjustments to crash exposure were based on vehicle population for each class. As more vehicles were registered, the occurrence of a fatal crash involving that vehicle class was expected to rise proportionally. R.L. Polk registration data was used to transform fatality counts into fatality rates per registered vehicle. Since the likelihood of an impact by an errant vehicle increases proportionally as the population of each device type increases, a second adjustment for the number of installed roadside devices would have been beneficial. However, most states do not maintain an accurate inventory of installed devices. Therefore the device installation counts could not be obtained. In addition the increased roadwork activity and resulting increase in the numbers of temporary barrier systems through the 1990âs may have been responsible for an increase in barrier related deaths. In assessing fatality trends for vehicles impacting each class of device, the inherent crashworthiness and level of occupant protection provided by the vehicle directly relates to crash injury outcomes. The improvement in the safe design of vehicles was not adjusted for during this analysis. The
103 effect of these improvements would lead to reduced fatality counts for a given impact condition when compared with crash outcomes for impacts involving earlier model vehicles. Based on the analysis of guardrail, concrete median barrier and small to medium pole impacts involving each of the investigated vehicle classes (i.e. small, midsize and large cars, small, midsize and large SUVâs, and small and large pickups), two clear trends were observed. The first observation regarded fatality rates for longitudinal barrier (guardrails and concrete median barriers) impacts by small and midsize SUVâs. These fatalities increased at a rate higher than the vehicle population increase. Further investigation of small and midsize SUV crash behavior indicated that the increase in fatalities correlated directly with the occurrence of vehicle rollovers. A comparison of fatality rates for these vehicles with and without rollover involvement (Tables 2.2 and 2.3) suggested that the inherent instability introduced to these vehicles during longitudinal barrier impacts had a more significant effect than the same condition involving pickup trucks. This finding suggests that pickups may not adequately represent the worst case impact for the vehicle fleet and selection of future crash test vehicles should account for this. The second observation indicated improved outcomes involving small and midsize cars during impacts with longitudinal barrier systems. Small and midsize car impacts with longitudinal barrier systems occur frequently, however fatality rates have declined since 1990. This behavior may be attributed to improved vehicle handling characteristics reducing the frequency of such impacts, improved occupant protection by the vehicle, and improved roadside device performance. 5.1.2 Individual Case Review Anecdotal evidence of poor interaction between roadside structures and impacting vehicles was gathered from review of existing NASS/CDS and CIREN case data. (See Section 2.2) This review lead to valuable insight into characteristics of common roadside crash configurations, however inadequate documentation of device characteristics and a lack of specific information regarding vehicle kinematics during impact hindered the analysis from drawing further conclusions. Key findings from this review include the following. 1. Side impact crash outcomes involving barrier end-terminals depend largely on direct engagement of the vehicle structure (i.e. door sill or pillars). Taller SUV and pickup structures engage existing terminal devices adequately while lower small and midsize car structure often do not. Door structures often cannot prevent excessive door deformation and occupant compartment intrusion during these impact conditions.
104 2. Oblique longitudinal barrier impacts involving SUVâs do not directly lead to rollover, however increased vehicle instability, driver overcorrection, and inherent vehicle kinematics lead to subsequent vehicle rollover. 3. The performance of airbag systems during roadside hardware crashes is not well understood, however it is possible that soft longitudinal pulses may delay airbag deployments. This condition may lead to reduced levels of occupant protection. 4. Frequently, existing NASS/CDS data collect techniques do not adequately document barrier crash conditions, barrier performance and structural interactions. Currently, NASS/CDS investigators code a category of impacted device. While this information is helpful, an investigators ability to recognize cases of incompatibility is limited due to a lack of specific barrier/vehicle attributes. Case photos are available, however NASS/CDS investigations focus largely on the vehicle, the occupant compartment and restraint systems. Less consideration is given to the impacted device. Through the case review process, additional data points have been defined that help to adequately characterize the performance of the vehicle/roadside device. Of particular interest are the specific installation attributes for crash involved devices. Currently, state highway engineers use guidelines for installation practices, however each installation is tailored to the terrain, road use and devices available. These conditions lead to complex installation practices and difficult crash investigations. To aid in data collection following a crash, three additional crash investigation forms, similar to existing CDS forms, have been created as a product of NCHRP 22-15. These proposed forms improve data concerning the following items: Device Design Characteristics Post, block out, rail types, barrier profiles, installation heights Location of impact relative to device features Distance downstream, distance from splice, distance from roadway, curb presence Verification of Proper device installation Barrier heights, protection of dangerous features Estimated Impact Mode Impact angle, tracking vs. non-tracking, rotational conditions Overall estimate of device performance Improved crash photos, device deformation
105 5.1.3 Crash Testing Currently, the test methods (NCHRP 350) evaluating the performance of roadside hardware devices use only a small sample of vehicle platforms. These tests are often performed without occupants or Anthropomorphic Test Dummys (ATDs). Due to the limited number of vehicles used, assessing compatibility of the entire fleet with a given device is not possible. A broader cross section of test vehicles is required to verify the appropriate identification of the most extreme case. Currently only the 820kg car and full sized pickup truck are tested. To determine the vehicles included in this wider sample a database, which identified and measured the key attributes of over 300 different vehicles, was created. Additionally, less detailed data for over 5000 vehicles also was aggregated. This data, linked with vehicle registration data, was reviewed to establish trends in the vehicle markets and determine how the attributes of the United Stateâs vehicle fleet are changing. It was found that the small, midsize and large SUV populations were the most rapidly growing. These vehicles were somewhat similar to the full size pickup truck. However, these vehicles differ from the pickup truck in CG height, Static Stability Factor (SSF), weight distribution and other characteristics important in vehicle to roadside hardware impacts. These vehicles now account for a sizable portion of the US vehicle fleet, necessitating their compatibility with guardrails. It is important that testing be performed using these vehicles to determine that their level of compatibility with roadside hardware systems is adequate. The crash test study shown in Chapter 3 is a compelling example of a vehicle to roadside system incompatibility. These test outcomes suggest that compatibility cannot be estimated using gross vehicle characteristics like vehicle mass, wheelbase and track width alone. Detailed structural attributes of vehicles must be considered with respect to the particular impacted device to understand crash behavior. The behavioral differences observed between these vehicles suggests that additional test requirements should be considered for all vehicle structures rather than a single representative vehicle or a vehicle chosen at the extreme of the entire vehicle fleet. The mechanism for implementing these additional tests must be determined however. 5.1.4 Industry and Government Awareness The NCHRP 22-15 workshop displayed a gulf between Automobile manufactures and the roadside hardware creators in understanding the issue of roadside hardware compatibility. In order to improve the overall safety of vehicle entering the roadside, initiation of a cooperative approach between the roadside safety community and the automotive safety community is necessary. When such
106 cooperation was discussed during the project workshop, both roadside and vehicle safety representatives expressed interest. A proposed concept involves the creation of a working group within an existing professional society, The Society of Automotive Engineers (SAE). In addition, automotive industry involvement at the Transportation Research Board Annual meetings would stimulate future activity with the goal of safety improvement in mind. 5.2 Future Research 5.2.1 Roadside Collision Data Collection Future evaluations of roadside hardware/vehicle compatibility will require improved data sources. Real world crash data is an important resource for the evaluation of roadside hardware compatibility. However, existing sources have a series of shortcomings. To rectify this, an evaluation of the proposed roadside crash investigation forms must be conducted. The ability for NASS/CDS investigators to collect additional barrier data must be evaluated through increased involvement of the National Center for Statistical Analysis (NCSA), the center that maintains the NASS systems. A pilot program to evaluate critical factors involved in the collection of this additional data must be performed. This program must address the following issues: Data forms and collection techniques must be optimized so that investigators can collect the information within a reasonable time frame. The effect of the inclusion of additional barrier data on the time on scene and subsequent case analysis must be evaluated. Roadside device crashes often occur on high-speed roadways. Performance of the detailed crash scene investigations, which were proposed in Section 4.3 of this report, requires some addition risk to investigators. An investigation into the willingness of investigators to assume this additional risk is necessary. Further, the authority of NASS/CDS investigators to block traffic when necessary for investigations must be understood. Once new roadside crash investigation forms are adopted, guidelines must be established to ensure their suitability for investigation of existing and emerging roadside devices. A method to include new device designs into existing collection sheets must be implemented so that data collection remains useful. The evaluation of proposed data forms and collection techniques should involve existing crash data collection studies. CIREN crash investigations offer great detail in evaluation of the causation, vehicle dynamics and occupant outcomes of vehicle crashes. These case investigations, although limited
107 in number, currently include information that provides insight into vehicle/roadside hardware compatibility. A current shortcoming of this system however is the exclusion of vehicle rollovers. As demonstrated throughout the research period, rollover is a major component of vehicle to roadside hardware crash compatibility. A proposed special crash study would use these highly specialized CIREN investigators to inspect these types of crashes and their outcomes. 5.2.2 Evaluation of Alternative Test Methods Since the US vehicle fleet is constantly changing, annual monitoring of vehicle populations should be conducted. As consumer tastes change, the vehicles that best represent the current vehicle fleet should be chosen for use in the hardware tests and the finite element models. Current development of vehicles that are compatible with one another are converging on higher degrees of geometric alignment for frontal structures. Synergy between the roadside community and vehicle manufacturers is necessary so that bumper and frame rail height requirements are common to serve vehicle to vehicle compatibility as well as vehicle to roadside hardware compatibility goals. In order to verify improved performance of the vehicle/roadside system in the future, multiple vehicle platforms within a class should be tested (similar to Lumina and Taurus testing) for the most common roadside hardware devices being installed. Testing with similar vehicles within a class ensures that the barrier performs adequately across the class of vehicles, not only for a specific vehicle. Current testing showed that that concrete barrier design uses barrier shape and vehicle lift to control vehicle damage and lateral accelerations. However, NASS/CDS and FARS data has shown that when vehicle lift is applied to some SUVs and pickups, these vehicles may roll or lose control due to their inherent instability. The HARM due to this increased instability and subsequent rollover is believed to exceed the potential HARM that results from the redirection of these vehicles by a more vertically oriented barrier. Further testing and research should be done to ensure the compatibility of these vehicles with these devices. In addition, impacts with roadside systems may not provide a clear crash signature to airbag control modules. The emergence of improved side impact energy management systems (side airbags or energy absorbing side structures) has increased the levels of tolerance of occupants to the lateral acceleration of vehicle. In order to further address this issue, testing using occupants (ATDs) would help to identify if a more aggressive redirection of vehicles would lead to injury causing occupant loads. Care also must be taken to ensure that the airbags continue to deploy properly when vehicles impact guardrails. Specifically, impacts at a 15-25 degree impact angle may lead to delayed or improper
108 deployments of frontal and side airbags. Testing recent vehicle models with dummies is important to understand if the firing of the airbag is timed properly in these types of impacts. 5.2.3 Compatibility Evaluation Using Detailed Finite Element Models A collection of finite element models, developed by the FHWA/NHTSA National Crash Analysis Center and the FHWA Centers of Excellence, has been created that can aid in the testing of roadside hardware devices. These models are publicly available to both the automotive manufacturers and the roadside hardware manufacturers. The models should be utilized to test a wide variety of vehicle types with several different roadside devices. In addition, the devices can be tested in almost any configuration at relatively low cost. Continuous renewal of these models is required to ensure that the effects of recent design trends can be simulated properly. In addition, research should be done to combine the vehicle finite element models, provided by manufacturers, with roadside hardware models, created by the Centers of Excellence and other FHWA laboratories. These models may be exercised by an unbiased research organization, by safety engineers at the vehicle manufacturers, by FHWA staff or other proposed groups. This exploratory program will help to establish protocols and introduce the concept of vehicle design for improved interaction of emerging vehicles with roadside hardware systems.
