Transportation Research Record: Journal of the Transportation Research Board

The development and full-scale crash testing of a new, reusable, and essentially maintenance-free crash cushion designed to shield wide hazards are described. The energy-dissipating medium in this device is composed of high-molecular-weight, high-density polyethylene cylinders. This “smart” energy-dissipating thermoplastic is the same material used in the REACT family of narrow-hazard crash cushions and truck-mounted attenuators. It is self-restorative and reusable and possesses excellent energy-dissipation properties. The design process incorporated the results of impact testing with individual polyethylene cylinders, finite-element modeling, and a comprehensive full-scale crash testing program conducted in accordance with the guidelines of NCHRP Report 350. It is demonstrated that finite-element modeling is a cost-effective tool whose use can minimize the number of costly full-scale crash tests required to develop an effective impact attenuation device. This new, reusable, wide-hazard crash cushion satisfies all of the requirements of NCHRP Report 350 and has been accepted by FHWA for use in the National Highway System.

A new flared W-beam guardrail terminal that incorporates energy-absorbing technology has been developed and successfully crash tested to meet the criteria presented in NCHRP Report 350. The terminal, designated FLEAT-350, incorporates an impact head designed to dissipate impact energy by producing a series of plastic hinges in the W-beam as the impact head is pushed down the guardrail. The energy-absorption mechanism allows the flared terminal to absorb large amounts of kinetic energy before the vehicle is allowed to gate through the system. Most components of the new terminal are similar to those incorporated in the SKT-350 terminal. The terminal system is 11.4 m (37.5 ft) long and incorporates a straight flare with a final end offset of between 0.76 m (2.5 ft) and 1.2 m (4 ft). The terminal uses two breakaway posts in foundation tubes and five rough-cut controlled-release terminal posts of 150 by 200 mm (6 by 8 in.). The energy-absorbing design minimizes problems associated with vehicles that penetrate the terminal and strike slopes, ditches, or other hazards behind the guardrail while still traveling at a high rate of speed. Furthermore the new design has fewer components and is less sensitive to post location and installation details than some other flared systems.

A project to develop a flared guardrail end terminal system that would meet NCHRP Report 350 Test Level 3 (100 km/h) requirements was initiated at the request of a group of small guardrail manufacturers that did not have access to such a system. This was a major concern to this group that needed to have access to such a system before the October 1998 Intermodel Surface Transportation Efficiency Act of 1991 implementation date for NCHRP Report 350 roadside hardware. A set of design, performance, and cost objectives was established, and a decision was made to qualify the system for both weak and strong soils. The development project resulted in a new type of design and specifications for the wood posts used in the end terminal section. The redirective and gating end terminal (REGENT) system was designed and tested to meet the conditions described in NCHRP Report 350 (Test Level 3) and was approved for use on the National Highway System by FHWA. The REGENT system was tested when it was attached to both wood and steel downstream guardrail posts.

Highway agencies are continually called upon to make decisions about roadside safety projects by considering their relative benefits and costs. ROADSIDE software is a tool developed in the late 1980s for costeffectiveness analysis of safety features and has been used by different highway agencies. As a part of the continuing effort to improve such costeffectiveness procedures, the Roadside Safety Analysis Program (RSAP) has recently been developed at the Texas Transportation Institute. RSAP uses a simulation technique to replicate single-vehicle accidents on a highway facility and can test the effectiveness of the various countermeasures that are used to reduce the severities of these accidents. The application of RSAP to actual case study sites and the outcome of a study recently conducted at Wayne State University are described. Various median treatments on freeways were tested, and their respective benefit-cost ratios, as estimated by RSAP output, were examined. A set of sensitivity analyses of the RSAP output to various input parameters was also conducted. It is found that RSAP is user-friendly software and is fully operational on a personal computer with a Pentium processor in the Windows 95 environment. RSAP provides a summary of the accidents predicted for each scenario tested, and those are segregated by each feature coded along with its respective severity and cost. The economic analysis procedure is based on strong analytic principles, and the results are sensitive to accident cost, cost of construction, and interest rate. The software output is generally sensitive to the input parameters, and the trends appear to be logical. Additional research to calibrate the accident prediction model with historic accident data at the project site will be highly desirable.

Concerns about the Hutchinson and Kennedy (H&K) encroachment data used as the basis for AASHTO’s Roadside Design Guide (RDG) procedures for computing guardrail runout length have led to proposed revised guidelines based on more recent data collected by Cooper in Canada. The revised guidelines result in guardrail lengths that are approximately 35 percent shorter. Data from the original H&K and Cooper studies were analyzed to determine why the two studies seem to have inconsistent findings. The concerns about the H&K data expressed by the authors who proposed the revised guidelines were found to be invalid. However, team-by-team comparisons of the Cooper data revealed inconsistencies in encroachment rates, lengths, and departure angles for highways with similar speed limits. Documented data collection problems and unanswerable questions about the methodology used to measure encroachment lengths and departure angles make it impossible to validate the accuracy of the Cooper data. Adjustments were made to the Cooper data to compensate for possible data irregularities to examine the consistency between them and the H&K data. No evidence indicating that the encroachment lengths reported by H&K are inconsistently high compared with the encroachment lengths reported by Cooper was found. The findings suggest that reducing guardrail runout lengths from current RDG guidelines for highways with high speed limits [112 km/h (70 mph)] may not be prudent.

In-service performance evaluation is the process of assessing the performance of roadside safety hardware under real-world service conditions. The purpose of in-service evaluation is to determine and document the manner in which a safety feature performs during a broad range of collision, environmental, operational, and maintenance situations for typical real-world site and traffic conditions. An in-service performance evaluation of the concrete median barrier (CMB) in Connecticut was undertaken to determine how often CMBs are struck and how often such collisions are reported to the police. The method used to perform the inservice performance evaluation of the CMB is described, and the results are presented. Data were collected for a section of CMB located on I-84 in Hartford, Connecticut. Accurate information about impacts with CMBs is difficult to obtain because so many collisions are unreported. Even for severe collisions, damage to the barrier is rare, and little maintenance is required. In many cases, police reports are the only means of monitoring when, where, and how severe the CMB collision was. Videologging was used to accomplish the in-service performance evaluations of CMBs to document the extent of unreported collisions.

Each year more than 14,000 persons are killed and 1 million persons are injured as a result of roadside crashes. Recent estimates of the annual societal costs associated with these roadside crashes amounted to $80 billion. Clearly, there is a need for further understanding of the roadside safety problem. The impact performances of roadside safety features are typically evaluated through full-scale crash testing with two vehicles selected from the extremes of the passenger vehicle fleet in terms of weight and size. The implicit assumption is that if a roadside safety feature successfully passes the test requirements for vehicles at the extremes for the fleet, the feature will perform satisfactorily for all other vehicles in between. Many vehicle parameters could influence performance during impacts, and this assumption may or may not be valid. The safety performances of roadside features for various passenger car platforms and light-truck subclasses were evaluated. The study approach consists of evaluations of the frequency and severity of roadside crashes for these generic platforms and subclasses by using recent crash data from the Fatal Accident Reporting System, the General Estimates System, and the Highway Safety Information System.

The effects of wood post strength and soil strength on the dynamic performance of guardrail systems have long been a concern of the roadside safety community. Evidence from full-scale crash tests has suggested that these parameters may significantly affect guardrail system performance. Essentially identical tests have resulted in widely varying outcomes that might be the result of various post strengths and soil conditions. A finite-element model of a common guardrail terminal—the modified eccentric loader breakaway cable terminal—was developed to examine the effects of post strength and soil strength on the overall performance of the terminal system. A matrix of 12 simulations of a particular full-scale crash test scenario was conducted with the explicit nonlinear dynamic finite-element software LS-DYNA3D to establish the combinations of post and soil strengths that produce favorable results. The parametric simulations show that certain combinations of soil and post strengths increase the hazardous possibilities of wheel snagging, pocketing, or rail penetration, whereas other combinations produce more favorable results. Conditions that will maximize the safety and reliability of the guardrail terminal system are identified.

Vehicles that collide with highway safety features do not always produce the same effects. The outcome for each vehicle that collides with a feature will be different, depending greatly not only on its mass, velocity, and impact angle but also on the position of its center of mass and on the tire-pavement friction. The ways in which barrier and striking vehicle behaviors are affected by changes in the position of the center of mass and in the tire-pavement side friction were analyzed to identify the most dangerous loading systems. Toward this aim numerous simulations of a collision of a truck against a steel road safety barrier were carried out through nonlinear, dynamic finite-element analysis. The results obtained demonstrate that the position of the center of mass is of greater consequence in collisions in which the impact energy is close to the maximum barrier containment capacity and under conditions that favor friction. The longitudinal position of the center of mass has a large impact on the risk that the vehicle will pass over the barrier or roll over. If it is in the back, the most common loading system, the risk of passing over the barrier is rather limited compared with the risk of rolling over. As the position of the center of mass is pushed forward, the risk of rolling over is reduced but the risk of passing over the barrier increases. If the height of the center of mass increases, the rollover risk is higher, but a higher center of mass does not significantly influence displacement and the risk of passing over the barrier.

A bullnose median barrier was designed and successfully crash tested head-on at 100 km/h with a 2000-kg pickup truck. After a failed pickup truck test, LS-DYNA was used to simulate the failed system, to determine the cause of the failure, and to evaluate a solution to the problem. Subsequent testing substantiated the LS-DYNA predictions. To keep up with the design project deadlines, some features of the simulation model were simplified. For other features, however, great attention to detail was required to make a useful model. Specifically, a considerable amount of effort went into defining the material failure criteria and appropriate mesh density for the guardrail, rolling tires for the truck model, and application of the relatively new cable element in LS-DYNA. The simulation effort is documented to demonstrate how LS-DYNA can be used to solve difficult design problems in roadside safety applications.

Freestanding concrete highway barriers were originally implemented for work-zone applications. These barriers are becoming more widespread in their use and have been implemented in some innovative counter-flow applications. As the amount of freestanding barriers increases, so does the demand for detailed operational performance data. However, investigations of their crash protection and operation have been limited compared with those for the more common post-and-beam guardrail configurations. Computer simulation is one method for evaluating the performance of freestanding concrete barriers during vehicle collisions. A two-dimensional model of vehicle and barrier dynamics was developed to investigate the concrete barrier systems used by the British Columbia Ministry of Transportation and Highways. The simulation program was developed by using a multibody formulation for the barrier segments. A description of the model formulation is presented with a comparison of model output of recorded crash test data.

“Resource sharing” refers to projects that use public right-of-way for telecommunications facilities and that result in some form of compensation to the public agency. The facilities most often installed are transmitter antennas supported on aboveground structures (wireless facilities) or buried fiber-optic cable. Research indicates that fewer than 10 state departments of transportation (DOTs) are leasing or licensing wireless sites to telecommunication companies, but many more states desire data concerning the revenue that they might receive from such installations. The current range of fees that state DOTs receive from the telecommunications companies for the wireless sites, and the fees that the DOTs receive from other companies (colocators) that place their facilities on a structure owned by a first locator, are described. The typical site locations, the group (DOT versus company) that identifies lease sites, typical support structures for the antennas, the manner in which the revenue may be used in the various states, and legal difficulties associated with wireless resource sharing are also described.

Facilitation of the development of the information superhighway, which is supported by fiber-optic technology, has begun to emerge on the agendas of some transportation agencies as they search for means to satisfy communications needs, create greater efficiencies and economic benefits, and possibly stem the trend toward increasing population concentrations in urban centers. In Minnesota, the state department of transportation has embarked on a process that has evolved into a landmark public-private partnership called Connecting Minnesota to build, operate, and maintain a high-speed, statewide fiber-optic communications backbone. When complete, nearly 75 percent of the state’s population will be within 16 km (10 mi) of this fiber-optic backbone. Many areas of rural Minnesota receive communications services via traditional copper wire or by limited lower-capacity fiber-optic cable. The current infrastructure does not meet the requirements for application of new and emerging technologies for the transfer of information between businesses, public agencies, educators, and friends and families. Access to advanced telecommunications technologies and services is necessary if outstate (remote) communities are to remain vital and prosperous. It has been suggested that in the next century the information superhighway will be as important to business and commerce as development of the railroads and the Interstate highway system was in the 19th and 20th centuries. Despite challenges, Connecting Minnesota and its partners are proceeding on a course for completion in 2001. Lessons learned from Minnesota’s experience should help smooth the road for other states attempting to facilitate and accelerate the construction of the information superhighway.

The combination of airborne laser and digital camera technologies can be used as a cost-effective and efficient means of digital mapping for civil and environmental applications. A framework for the use of these remote-sensing technologies in conjunction with Global Positioning System (GPS) receivers for digital mapping of infrastructure projects is presented. An example of the benefit of airborne laser terrain mapping combined with differential positioning with a GPS is low-altitude, high-speed laser scanning of up to 81 km² (20,000 acres) per day, which achieves an accuracy of 10 cm with reduced operating constraints, such as cloud and vegetation cover, traffic and usage, and time of day. The three-dimensional digital coordinate data are directly loaded into terrain mapping and computer-aided design software. This leads to efficient and error-free data processing and map generation. Conventional topographic surveys are slow, and aerial photogrammetry is limited by operating constraints and time-consuming data processing and interpretation. Examples of recent applications of the system for diverse projects like pipeline networks, power transmission lines, railroad tracks, and freeway interchanges are presented.

As part of NCHRP Project 15-15, Collection and Presentation of Roadway Inventory Data, the authors conducted a nationwide survey of the state departments of transportation (DOTs), visited several DOTs and vendors, and reviewed the literature regarding roadway inventory data collection. On the basis of that work, a process to help state DOTs match inventory needs to data collection technologies is presented. The process that is described may help state DOTs narrow their focuses to the most relevant inventory data collection technologies. The authors present a four-step process that can be used to select inventory data collection technologies. First, the authors help agencies view their inventory data collection in context. The authors show how successful inventory data collection in the past has focused on the provision of timely information that is used to solve problems. Second, inventory units must select the proper mode of transport for the data collection technologies. Units should consider vans, the popular choice, versus satellites at one extreme and backpacks at the other extreme. Next, inventory units need to select the georeferencing technology. The relative merits of global positioning systems, inertial navigation systems, distance-measuring instruments, and range finders are discussed. Finally, inventory units must select sensors to provide descriptive information. For each common inventory element, the authors estimate the accuracy of collection with six common technologies and other special techniques. The authors also describe the initial costs and labor needed for the different collection technologies.

Surveyor is a mobile highway data collection system designed to collect measurement data about objects, features, structures, and landmarks located along highways and roadways for highway planning, management, and maintenance. It creates classified inventories annotated with object dimensions, object position relative to the road, and global position reference. The mobile data collection part of the system consists of a high-speed multifunction vehicle with minimum complement of a right-of-way video system with a precisely calibrated high-resolution video camera, a distance-measuring instrument for spatial positioning, a gyroscopic geometries system, and an ultrasonic grade system for precise measurement of vehicle attitude. Data acquisition is facilitated by multiple on-board computers, and the right-of-way video uses a time code for synchronization to the geometric and position databases. The interactive (postprocessing) part of the system uses a workstation to retrieve and buffer the video for the user to identify targets by using point and click with a mouse, to classify them, and to request position or size measurements. The software can make measurements on multiple images by using triangulation or on single images by using the idealized plane of the highway as a reference.

The means by which three technologies, geospatial imaging, soft-copy photogrammetry, and alignment optimization, can assist in generating and evaluating plans for new road alignments are assessed. The technologies are illustrated by two case studies. One, based on satellite data, was selected as a case study of the technologies because planning had been in progress for several years and all the data except the satellite images were available. The study was conducted at a broad level to demonstrate the ability of the technologies to produce high-quality results quickly and to prepare material for presentation. Constraints were kept simple because the aim was to demonstrate the technology, not to undertake a detailed study of the options; and the road authority’s costs for earthwork and other construction were used to provide comparability with the ongoing study. The costs of the alignments produced by the optimization were about 25 percent less than those from the ongoing study, and there was a consensus that the combination of technologies could have cut 12 to 18 months off the planning process. The second study, based on aerial photography, was a straight application of the technologies with no comparison with other studies and had to provide all the detail required for public presentations. It was a full-scale planning study in which aerial photography and alignment optimization technologies were used to identify the best transport corridors, to generate a set of low-cost alignment options within the corridors, and to assess the various alignment options.

Storm-water runoff from highway pavement is capable of mobilizing and transporting a wide gradation of particulates and solids. A significant fraction of this particulate matter is generated from traffic activities, including the abrasion between tires and pavement. Particulate accumulation and characteristics from an urban Interstate pavement site are investigated. The southbound asphalt pavement section studied consisted of four traveled lanes and two paved shoulders with a measured average daily traffic of 110,000. The pavement section has a transverse slope of 0.020 m/m and a longitudinal slope of 0.004 m/m. Recovery of deposited particulate matter indicates that accumulation of particulate matter along the east paved shoulder (the downslope shoulder) is significantly greater than that along the the west paved shoulder (the upslope shoulder) or the pavement. Particles larger than 25 μm were separated mechanically to generate particle size distributions (PSDs) and also to separate particles for further analysis. The east shoulder PSDs were consistently coarser across the entire size gradation than the west shoulder and pavement PSDs. Particle specific gravity ranged from 2.70 to 3.01 g/cm³ across all gradations. Particle specific surface area increased with decreasing particle size for all gradations. In contrast, total particle surface area decreased with decreasing particle size. Integration of metal element concentration (in micrograms per gram) of zinc, copper, cadmium, and lead with the mass fraction of each particle size demonstrates that most of the metal mass is associated with the coarse to mid-range-size particles. Study results are useful for design of roadway drainage appurtenances and water quality control devices that clarify storm water mainly through the separation of particles.

Local pier scour has been the topic of numerous research studies on noncohesive alluvial materials. However, because of the complexities involved in cohesive-material scour, very little effort has been devoted to the study of scour around bridge piers situated in cohesive materials. The parameters that affect the scour mechanism at bridge piers located in saturated and unsaturated clayey soils are identified, and prediction equations for quantification of the effects of some of the cohesive properties on the resulting local scour are developed. The scour depth predictors developed from the analysis of the experimental data express pier scour in terms of the approach flow conditions, initial water content, compaction, and soil shear strength of the cohesive bed materials. These predictors are related to scour in noncohesive alluvial material to derive factors that express clay scour as a percentage of sand scour. At bridge locations where exact soil properties are unknown, these estimators may be used to provide bounds to scour estimations.

Factors related to the probability of joint flooding on paired streams were investigated. Stream pairs were considered to have flooded jointly at the design-year flood threshold (corresponding to the 2-, 10-, 25-, or 50-year instantaneous peak stream flow) if peak stream flows at both streams in the pair were observed or predicted to have equaled or exceeded the threshold on a given calendar day. Daily mean stream-flow data were used as a surrogate for instantaneous peak stream-flow data to determine which flood thresholds were equaled or exceeded on any given day. Instantaneous peak stream-flow data, when available, were used preferentially to assess when the flood threshold was exceeded. Observed probabilities of joint flooding were computed as the ratios of the number of days when stream flows at both streams concurrently equaled or exceeded their flood thresholds (events) to the number of days when stream flows at either stream equaled or exceeded its flood threshold (trials). Logistic regression equations for estimating the probability of joint flooding at the 2-year flood threshold were developed on the basis of event-trial ratio and basin characteristic data. Distance between drainage area centroids, the ratio of the smaller drainage area to the larger drainage area, mean drainage area, and the centroid angle adjusted 30 degrees were the characteristics most closely associated with the probability of joint flooding on paired streams in Ohio. In general, the probability of joint flooding decreased with an increase in centroid distance and increased with increases in drainage area ratio, mean drainage area, and centroid angle adjusted 30 degrees.

The results of laboratory studies investigating the energy dissipation caused by the placement of roughness elements to impede the downward thrust of water near pier faces are summarized. The purpose of the investigation was to evaluate the effectiveness of the method as a viable countermeasure for local pier scour. The results indicate that the roughness element size and spacing had various effects on the energy dissipated from the downward thrust of water and, therefore, the potential to reduce local pier scour. On the basis of the results that are presented, it should be feasible to develop countermeasures that can be retrofitted to existing piers, which will measurably reduce local pier scour. However, additional laboratory and prototype scale testing will be required.