In-Service Performance Evaluation of Guardrail End Treatments (2018)

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.

NATIONALLY COORDINATED EVALUATION RESEARCH 77 Annex 3-1 Summary of Procedural Guides to In-Service Evaluation of Roadside Safety Devices The sections below describe four past or current activities to develop and demonstrate data collection and analysis methods for in-service evaluation of roadside safety devices: • The pilot in-service evaluation of guardrail end treatments con- ducted by FHWA (FHWA n.d.), • The AASHTO-FHWA Task Force guardrail end treatment investi- gations (Joint AASHTO-FHWA Task Force on Guardrail Terminal Crash Analysis 2015), • The procedure for in-service evaluation described in NCHRP Re- port 490 (Ray et al. 2003), and • The procedure developed by the Texas Transportation Institute for the Texas Department of Transportation (van Schalkwyk et al. 2004). Table 3-2 shows a summary of the data elements used in each procedure compared with the generic list given in Chapter 2, Box 2-2. Table 3-3 compares these past procedures with the nationally coordinated evaluation research program outlined in this chapter with respect to objectives, orga- nization, and data sources.

78 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS TABLE 3-2 Data Elements Used in Three Procedures for In-Service Evaluation of Roadside Devices Data Elements Relevant to Evaluation of Roadside Safety Device Evaluation Procedure Washington State DOT Demonstration of NCHRP Report 490 Procedure (Igharo et al. 2004) FHWA Pilot End Treatment Evaluation (FHWA n.d.) TTI Procedure for Texas DOT (van Schalkwyk et al. 2004) Crash scenario: • Collision type (e.g., run-off-road) • Sequence of events • Number, types of vehicles involved • Contributing factors (e.g., speeding) • Driver characteristics (age, sobriety, gender) • Injury severity • Number of injured occupants • Restraint usage • Severity of vehicle damage From site investigation: • Orientation of vehicle • Location of strike on end treatment From police crash report: • Number, severity of occupant injuries • Number, type of vehicles • Crash scenario Serious and fatal injury crashes: • NHTSA SCI team standard investigation: – Detailed scene inspection (including diagrams and photos) – Vehicle inspection – Witness interviews • Added detail on end treatment crash dynamics Minor crashes: • Data from police crash report • Site inspection by state personnel Data from police crash report Roadside device information: • Characteristics of devices involved (design type, installation details, condition, age, maintenance history) • Postcrash condition • Repair cost From special postcrash investigation, detailed descriptions and measurements of • Precrash end treatment, posts, connections, foundation • Postcrash damage and positions of end treatment elements and posts • Repair cost • Standard series of photos of crash- involved end treatments taken by state personnel • Detailed descriptions and measurements of performance in crash and postcrash damage and positions of end treatment elements and posts Site inspection by maintenance personnel at time of repair of crash damage, records (for end treatments): • Device types • Yes–no assessment of whether performed as intended • In Phase 2 inspection (for devices suspected of poor performance), eight additional data elements on device installation, condition • Standard site photographs Maintenance department record of repairs and cost Road and environmental conditions: • Road surface condition • Roadway geometry • Traffic characteristics • Light condition • Time of day • Weather condition From site inspection: shoulder material From existing road inventory: • Roadway description • Traffic volume Special survey to inventory end treatments by type • Road features from crash site inspections • “Snapshot” end treatment inventories for pilot substate areas • End treatment maintenance records as available Data from: • Police crash report • Maintenance personnel site inspection • Photos NOTE: DOT = Department of Transportation; NCHRP = National Cooperative Highway Research Program; NHTSA = National Highway Traffic Safety Administration; SCI = Special Crash Investigation; TTI = Texas Transportation Institute.

NATIONALLY COORDINATED EVALUATION RESEARCH 79 TABLE 3-2 Data Elements Used in Three Procedures for In-Service Evaluation of Roadside Devices Data Elements Relevant to Evaluation of Roadside Safety Device Evaluation Procedure Washington State DOT Demonstration of NCHRP Report 490 Procedure (Igharo et al. 2004) FHWA Pilot End Treatment Evaluation (FHWA n.d.) TTI Procedure for Texas DOT (van Schalkwyk et al. 2004) Crash scenario: • Collision type (e.g., run-off-road) • Sequence of events • Number, types of vehicles involved • Contributing factors (e.g., speeding) • Driver characteristics (age, sobriety, gender) • Injury severity • Number of injured occupants • Restraint usage • Severity of vehicle damage From site investigation: • Orientation of vehicle • Location of strike on end treatment From police crash report: • Number, severity of occupant injuries • Number, type of vehicles • Crash scenario Serious and fatal injury crashes: • NHTSA SCI team standard investigation: – Detailed scene inspection (including diagrams and photos) – Vehicle inspection – Witness interviews • Added detail on end treatment crash dynamics Minor crashes: • Data from police crash report • Site inspection by state personnel Data from police crash report Roadside device information: • Characteristics of devices involved (design type, installation details, condition, age, maintenance history) • Postcrash condition • Repair cost From special postcrash investigation, detailed descriptions and measurements of • Precrash end treatment, posts, connections, foundation • Postcrash damage and positions of end treatment elements and posts • Repair cost • Standard series of photos of crash- involved end treatments taken by state personnel • Detailed descriptions and measurements of performance in crash and postcrash damage and positions of end treatment elements and posts Site inspection by maintenance personnel at time of repair of crash damage, records (for end treatments): • Device types • Yes–no assessment of whether performed as intended • In Phase 2 inspection (for devices suspected of poor performance), eight additional data elements on device installation, condition • Standard site photographs Maintenance department record of repairs and cost Road and environmental conditions: • Road surface condition • Roadway geometry • Traffic characteristics • Light condition • Time of day • Weather condition From site inspection: shoulder material From existing road inventory: • Roadway description • Traffic volume Special survey to inventory end treatments by type • Road features from crash site inspections • “Snapshot” end treatment inventories for pilot substate areas • End treatment maintenance records as available Data from: • Police crash report • Maintenance personnel site inspection • Photos NOTE: DOT = Department of Transportation; NCHRP = National Cooperative Highway Research Program; NHTSA = National Highway Traffic Safety Administration; SCI = Special Crash Investigation; TTI = Texas Transportation Institute.

80 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS TABLE 3-3 Comparison of Proposed Nationally Coordinated Evaluation Research Program with Selected Past In-Service Evaluation Procedures Nationally Coordinated Evaluation Research Program Evaluation Procedure Washington DOT Demonstration of NCHRP Report 490 Procedure (Igharo et al. 2004) FHWA Pilot End Treatment Evaluation (FHWA n.d.) TTI Procedure for Texas DOT (van Schalkwyk et al. 2004) Objectives • Validation of crash tests • Demonstration and assessment of benefits and costs of routine state highway agency in-service evaluation • Exploratory analysis toward development of a roadside device collision casualty risk model • Compare severity distributions of collisions with alternative guardrail end treatment types • Obtain data on correctness of installation and on repair costs for struck end treatments Develop and demonstrate data collection methods for state highway agencies to use in conducting in-service evaluations of end treatments Develop and test method for routine state highway agency in-service evaluation to detect roadside safety devices with relatively high failure rates Organization • Nationally coordinated research program with specified goals and term • Probably jointly governed by FHWA and AASHTO with participation by a small number of volunteer state highway agencies State-sponsored project conducted by university researchers with cooperation of state highway agency Project conducted by FHWA with cooperation of NHTSA NASS field teams, four state highway agencies, and a toll road authority State-sponsored project conducted by university researchers with cooperation of state highway agency Data Sources (see Table 3-2 regarding data elements) • For test validation, prospective database of crashes involving roadside devices; scientifically valid sample • For demonstration of state highway agency methods, roadside device condition and performance data collected as adjunct to maintenance management system Special investigations of all identified crashes involving end treatments in three maintenance districts for 1 year Crashes meeting study criteria identified from state sources; investigated by NHTSA NASS field teams Data collected by maintenance workers at time of repairs at sites of roadside device collisions; collated with police crash report NOTE: AASHTO = American Association of State Highway and Transportation Officials; DOT = Department of Transportation; FHWA = Federal Highway Administration; NASS = National Automotive Sampling System; NCHRP = National Cooperative Highway Research Program; NHTSA = National Highway Traffic Safety Administration. FHWA PILOT IN-SERVICE PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS FHWA undertook its pilot in-service performance evaluation of guardrail end treatments in 2015, with the principal objective of developing and dem- onstrating in-service evaluation data collection methods that could be used by the states (FHWA n.d., 7). Improving crash testing also is an objective of this pilot study; therefore, the procedures call for prompt and detailed postcrash data collection at crash sites by trained investigators.

NATIONALLY COORDINATED EVALUATION RESEARCH 81 TABLE 3-3 Comparison of Proposed Nationally Coordinated Evaluation Research Program with Selected Past In-Service Evaluation Procedures Nationally Coordinated Evaluation Research Program Evaluation Procedure Washington DOT Demonstration of NCHRP Report 490 Procedure (Igharo et al. 2004) FHWA Pilot End Treatment Evaluation (FHWA n.d.) TTI Procedure for Texas DOT (van Schalkwyk et al. 2004) Objectives • Validation of crash tests • Demonstration and assessment of benefits and costs of routine state highway agency in-service evaluation • Exploratory analysis toward development of a roadside device collision casualty risk model • Compare severity distributions of collisions with alternative guardrail end treatment types • Obtain data on correctness of installation and on repair costs for struck end treatments Develop and demonstrate data collection methods for state highway agencies to use in conducting in-service evaluations of end treatments Develop and test method for routine state highway agency in-service evaluation to detect roadside safety devices with relatively high failure rates Organization • Nationally coordinated research program with specified goals and term • Probably jointly governed by FHWA and AASHTO with participation by a small number of volunteer state highway agencies State-sponsored project conducted by university researchers with cooperation of state highway agency Project conducted by FHWA with cooperation of NHTSA NASS field teams, four state highway agencies, and a toll road authority State-sponsored project conducted by university researchers with cooperation of state highway agency Data Sources (see Table 3-2 regarding data elements) • For test validation, prospective database of crashes involving roadside devices; scientifically valid sample • For demonstration of state highway agency methods, roadside device condition and performance data collected as adjunct to maintenance management system Special investigations of all identified crashes involving end treatments in three maintenance districts for 1 year Crashes meeting study criteria identified from state sources; investigated by NHTSA NASS field teams Data collected by maintenance workers at time of repairs at sites of roadside device collisions; collated with police crash report Identification of Crash Cases The important features of the procedure for identifying and investigating crashes are as follows: • The project is to obtain crash data in a 2-year period for four states (California, Massachusetts, Missouri, and Pennsylvania) with the cooperation of the four state highway agencies and the Pennsylva- nia Turnpike Authority.

82 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS • Property-damage-only and minor injury crashes are to be identi- fied and documented by the state agencies. Fatal and serious injury crashes are identified by the states and investigated by NHTSA Special Crash Investigation (SCI), an established network of teams stationed throughout the country. • FHWA has established a special arrangement with each agency for notification of crashes. For fatal or serious injury crashes, a desig- nated contact person in the agency receives notice of a crash from maintenance staff or state police and in turn notifies FHWA and NHTSA. • FHWA screens the crash notices to identify crashes that are within the scope of data collection and, for fatal and serious injury crashes within the scope, the SCI team is dispatched to conduct the inves- tigation. Investigations are to be conducted within several hours of the crash. • For property-damage-only and minor injury crashes, a similar noti- fication and screening procedure is followed, except investigations are by state personnel rather than by the SCI team. • Crashes are identified and investigated in a selected substate region (such as a maintenance district) within each state. In addition, statewide fatal crashes are included in at least one state. • Consistent data collection depends on the functioning of the stan- dard chain of communication from state maintenance staff or police through designated contact persons in the state agency to FHWA. FHWA conducted training of state personnel involved and pilot-of-pilot trials to test procedures. Considering the effort required to arrange for continuous, consistent cooperation among all the parties—the state agency central office and main- tenance staffs, state police, FHWA, and NHTSA crash investigators—it seems likely that broadening the scope of the safety evaluation objectives of such a project beyond guardrail end treatments, while adding incremen- tally to the cost, could increase the cost-effectiveness and total value of the activity. Data Elements Collected Highlights of the data collection procedures are as follows: • Standard data collection forms for the on-scene investigations of fatal and serious injury crashes and for property-damage-only and

NATIONALLY COORDINATED EVALUATION RESEARCH 83 minor injury crash investigations were developed by FHWA work- ing with the states and NHTSA. • Data collection includes a standard set of photographs to be taken at each crash site. A guidance document for investigators specifies the angles and distances of the photographs (FHWA n.d.). Photo- graphs are to be taken of the guardrail end treatment and adjacent guardrail and of the roadway and roadside near the crash site. • Data collection forms provide for recording information about the type of end treatment, indications of preexisting damage, instal- lation details, and conditions after the crash; the characteristics and condition of the adjacent guardrail; the crash scenario; and roadway and roadside conditions and geometry at the site (FHWA n.d.). • In addition to crash data, each participating agency is to provide an inventory of the guardrail end treatments on its system. Some agencies may provide maintenance records for the end treatments. Analysis The photographs are analyzed by FHWA or NHTSA experts to extract and code information about crash circumstances and end treatment per- formance. A database is to be compiled containing a record for each involved vehicle and the data from the forms and from the analysis of the photographs. AASHTO-FHWA TASK FORCE INVESTIGATIONS OF GUARDRAIL END TREATMENTS FHWA and AASHTO formed a task force in 2014 composed of representa- tives from nine state DOTs, AASHTO, and FHWA. The task force report states its purpose as follows (Joint AASHTO-FHWA Task Force on Guard- rail Terminal Crash Analysis 2015, 5): The Joint Task Force on Guardrail Terminal Crash Analysis (Task Force) was assembled to answer the following questions: • Are there performance limitations with the ET-Plus 4-inch extruding W-beam guardrail terminal and other extruding W-beam guardrail ter- minals (e.g., SKT [Sequential Kinking Terminal] and FLEAT [Flared Energy Absorbing Terminal])? • If yes, what are the next steps to further analyze or address these per- formance limitations? The Task Force assembled a set of historical records of crashes involv- ing end treatments from a variety of sources. A Federal Register notice was

84 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS published soliciting cases (Federal Register 2014). Crashes were identified from data submitted by states and from private individuals, from media reports, and from federal safety research databases. Of 1,200 initial cases identified, 161 were selected for detailed review, mainly on the basis of two criteria: completeness of information and in- volvement of a relevant device. Each case was reviewed by three indepen- dent expert teams. The reviews focused on identifying cases that revealed performance limitations of the end treatment. The Task Force’s analysis of performance limitations and its conclu- sions regarding end treatment performance and further evaluation needs are summarized in the section on validating crash test procedures earlier in Chapter 3. NCHRP REPORT 490 PROCEDURES FOR IN-SERVICE EVALUATION NCHRP Report 490: In-Service Performance of Traffic Barriers presented the findings of a project to “develop a practical procedures manual for conducting in-service performance evaluations of roadside barriers” (Ray et al. 2003). The In-Service Performance Evaluation Procedures Manual produced by the project (Ray and Weir 2002) was published as Appendix D of NCHRP Report 490. The manual presents a step-by-step procedure for in-service evaluations by state highway agencies. The procedure is intended to be applicable “for continuous monitoring [with] data collection . . . incorporated into routine maintenance and repair procedures” as a “long-term part of a safety man- agement system” as well as for special evaluations of specific devices (e.g., a device type that the agency is considering for replacement) (Ray and Weir 2002, 2). The manual organizes the procedure in three phases: planning and preparation, data collection, and analysis (see Figure 3-1). In the planning and preparation phase, the agency defines the objective of the evaluation and plans data collection and analysis. Planning steps include • Determining the sample size needed to meet the evaluation ob- jectives and selecting the geographic area and duration of data collection; • Assembling data to characterize the agency’s inventory of the de- vices to be evaluated; and • Establishing working relationships with police and with road maintenance personnel who will participate in identifying crashes, collecting data, establishing crash notification procedures, and training personnel.

NATIONALLY COORDINATED EVALUATION RESEARCH 85 FIGURE 3-1 In-service performance evaluation process proposed in NCHRP Report 490. SOURCE: Ray and Weir 2002, 3.

86 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS In the data collection phase, the manual includes instructions for four processes: prompt notification to the evaluators that a crash within the scope of the evaluation has occurred, assembly of data on each crash from preexisting sources (police reports and maintenance reports), collection of data at the site of the crash, and monitoring the frequency of unreported crashes. A detailed procedure for on-site data collection is specified. Data collec- tion would be by a two-person team, which could be agency maintenance personnel who had received training for the task. Data collection forms are provided for recording descriptions and measurements of the guardrail and end treatment. A standard set of photographs is specified. The manual includes instructions for taking the required measurements. The procedure for estimating the frequency of unreported collisions is to select a sample of control sites—locations with barriers of the types un- der evaluation. Maintenance personnel inspect the control sites on a regular schedule and record evidence that the barriers were struck. The data analysis outlined in the manual uses the rate of severe crashes per vehicle passing the device as the measure of performance. A method is presented for comparing the performance of two alternative device types by using accident modification factors to isolate the effect of a device type on the severity from the effects of differences in roadway characteristics at locations where the alternative types are installed. The Washington State DOT guardrail end treatment evaluation sum- marized in Chapter 2, Box 2-1, followed the NCHRP 490 procedure with some adjustments, and was intended as a trial and demonstration of the method. Figure 3-2 is an example of one of the data collection forms used in the Washington State evaluation. TTI PROCEDURE FOR IN-SERVICE PERFORMANCE EVALUATION Objectives In 2004, TTI completed a project to develop a procedure for in-service evaluation of roadside safety features for the Texas DOT (van Schalkwyk et al. 2004). The procedure applies to guardrails, guardrail end treatments, transitions, impact attenuation devices, and signs. The authors note that the agency did not have experience conducting evaluations and acknowledge that implementation of the evaluation would be unfeasible if the procedure placed too great a burden on agency personnel. Therefore, the procedure developed emphasizes practicality, especially with regard to limiting time demands on maintenance staff: “[The Texas DOT] required a process that would be easy to implement, and put as little strain as possible on monetary

NATIONALLY COORDINATED EVALUATION RESEARCH 87 and manpower resources while achieving the basic objectives [of evalua- tion]” (van Schalkwyk et al. 2004, 17). The intended purpose of the evaluations is “improving the decision- making process regarding approval, application, and maintenance of road- side safety features” (van Schalkwyk et al. 2004, 76). The report notes that the agency frequently adopts new device designs; that staffing, budget, and priorities of its maintenance program are changing; and that vehicle and traffic characteristics change over time. It argues that monitoring is neces- sary to catch any performance problems that emerge on account of these changes. Methods The report explains that limitations of the state’s existing information sys- tems necessitated the introduction of ad hoc data collection processes for in-service evaluation. The highway agency did not have an inventory of its roadside safety devices, the maintenance management information system could not be linked to the accident report database, and police accident reports contained little information about devices struck and recorded only the first harmful event of a crash. The procedure developed entails “working within existing maintenance reporting procedures [and] limiting data collection to specific key variables related to the site, installation of the features, and performance of the features” (van Schalkwyk et al. 2004, 17). It is structured in two stages. The Phase I evaluation is routine, continuous monitoring to identify device types that have relatively high failure rates. Failures are defined as crashes in which the device does not perform as intended because of a design defi- ciency, improper installation or maintenance, or an impact that exceeds the design capacity of the device. A Phase II evaluation would conduct more detailed postcrash site investigations of crashes involving devices identified in the Phase I evaluation as having a high failure rate, to determine the causes of failures. The procedure does not involve any special arrangements for noti- fication of crashes. Maintenance personnel who are at a site to repair a damaged device are to complete an evaluation form and take standard photographs. The maintenance office assembles a file for each collision that includes • The evaluation form; • Documentation of the maintenance performed (which can be linked to cost data); • The police accident report, when available; and • Site photographs.

88 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS FIGURE 3-2 Data form for recording end treatment postcrash condition. SOURCE: Igharo et al. 2004, Appendix A. Used with permission. IN-SERVICE PERFORMANCE EVALUATION OF ROADSIDE SAFETY HARDWARE STRUCK GUARDRAIL TERMINAL FORM 2: DAMAGE REPORT DATA COLLECTION Device Type: Date of Data Collection: Location (route #, direction, milepost, placement): A. Collision Scenario: Are you able to determine collision scenario? Y/N: If no, please explain in comment box. B. Which one of the eight diagrams below Most Closely represents the collision scenario?

NATIONALLY COORDINATED EVALUATION RESEARCH 89 FIGURE 3-2 Continued. IN-SERVICE PERFORMANCE EVALUATION OF ROADSIDE SAFETY HARDWARE STRUCK GUARDRAIL TERMINAL FORM 2: DAMAGE REPORT DATA COLLECTION C. Damage to End Treatment: 1. Max Rail Deflection at rail height (ft): 2. Total Rail Length Damaged (ft): 3. Total Rail Length Extruded (ft): 4. Total no. of secondary impacts: 5. Distance from post #1 to POI (ft): 6. How much did Post#1 foundation move at Ground Line?: 7. Was Soil tube at Post#1 pulled out?: 8. How much did Post#2 foundation move at Ground Line?: 9. Was Soil tube at Post#2 pulled out?: 10. Did cable release mechanism detach from Rail?: D. Damage Details along Terminal Profile Post No. Ground Line Deflection (ft) Deflection at top of Rail (ft) Post Broken (y/n) Post Bent (y/n) Splice Failed (y/n) Rail Torn or Broken (y/n) E. Final Position of Nose/Extruder: 1. Nosed Moved (Y/N) 2. Distance moved from Post#1 (parallel to Edge Line) (ft) 3. Lateral distance from Edge Line (ft) Comments:

90 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS Presumably a threshold minimum crash severity or device damage severity would be specified for triggering creation of a file. The project developed forms for the site investigations—a shorter ver- sion for the Phase I evaluation and a more detailed form for Phase II evalua- tions. The forms have questions specific to each category of roadside device (guardrails, end treatments, impact attenuators, transitions, and signs). For devices other than signs, the Phase I form has only one question about performance in the crash (van Schalkwyk et al. 2004, 127): DID THE SYSTEM PERFORM AS INTENDED? [Note that this ques- tion refers to a system failure or any condition where the system did not perform as we intended it to when it was installed (typically if the design load was exceeded etc.).] ___ YES ___ NO The Phase II form asks the same question but also asks for an explanatory sketch and comments, and adds a few more questions about performance. Box 3-2 Definition of End Treatment Failures in the TTI Evaluation Method The following is the text of a presentation slide in the training materials prepared for state highway agency participants in the trial of the evaluation method. END TREATMENTS ¨ Normally gating & redirective ¨ Failures Ø Impacts on end § Improper feeding of W-beam through impact head § Improper “gating” § More problematic: Smaller vehicles § Improper activation • May cause excessive rotation on the vehicle - Present side of occupant compartment - Guardrail deformation of the occupant compartment Ø Redirective impacts § Rail rupture § Loss of anchorage § And/or excessive pocketing SOURCE: van Schalkwyk et al. 2004, 163.

NATIONALLY COORDINATED EVALUATION RESEARCH 91 Both forms are much less detailed than the forms developed in NCHRP 490. The TTI report cites the NCHRP 490 procedure and states that a sim- pler data collection process was necessary in Texas to be compatible with the time and budget constraints in the agency’s maintenance department (van Schalkwyk et al. 2004, 19). The project developed a training program for maintenance staff who would complete the evaluation forms at crash sites. For each category of de- vice to be evaluated, the training materials contain a list of possible collision outcomes that would merit a response of yes to the perform-as-intended question. As an example, Box 3-2 shows the kinds of failures of guardrail end treatments defined in the training materials. These lists of failures do not appear on the crash data forms. A pilot of the procedure was conducted with the participation of main- tenance personnel in four local maintenance offices. In addition to conduct- ing training at the outset of the pilot, the researchers met monthly with the maintenance staff who were collecting data. These regular meetings were found to be valuable for maintaining cooperation and quality control (van Schalkwyk et al. 2004, 43). One aim of the pilot was to determine whether the maintenance work- ers were able to consistently respond to the perform-as-intended question. The researchers concluded that, in general, the workers were able to do so but acknowledged that “this was, however, problematic in some cases due to the complexity of the crash or other factors” (van Schalkwyk et al. 2004, 44). The report recommends that, for serious crashes, a technical expert investigate the crash site in addition to maintenance personnel. It was also concluded that the time burden of completing the form was not a hindrance to maintenance worker cooperation in the pilot. Analysis The primary analysis in the procedure is tracking the failure rates of devices in each category and manufacturer’s type. (The failure rate was defined as the fraction of all collisions in which the device failed to perform as intended.) The report describes calculations of the sample size necessary to detect a failure rate of a given magnitude with a specified degree of reliabil- ity and the time required to observe the required sample size in a specified substate area. In an example calculation in the report, 11 months of data collection would be required to observe the 1,150 crashes needed to reliably estimate a failure rate of 5 percent for end treatments with data from three large urban maintenance offices (presumably assuming all end treatments were of the same type) (van Schalkwyk et al. 2004, 62). The report also outlines a cost-effectiveness analysis to compare alternative device types. The sample size calculation described in the report may be too conser- vative as a criterion for beginning a Phase II investigation of a device. For

92 PERFORMANCE EVALUATION OF GUARDRAIL END TREATMENTS example, if the failure rate of a device is 5 percent, the probability that two or more failures will be observed in 50 crashes is more than 70 percent, and such an observation ought to be sufficient to trigger follow-up. In the pilot of the TTI procedure, eight of 28 collisions with impact attenuators were coded as “did not perform as intended.” The researchers judged this observation as sufficient evidence to warrant further investigation (van Schalkwyk et al. 2004, 53). In the interviews with state highway agencies conducted for the com- mittee, Texas DOT staff reported that the agency has not conducted in- service safety performance evaluations of roadside devices (Heimbecker and Lohrey 2016, 44), so it appears that the procedure developed and demonstrated by TTI was not adopted by the state. REFERENCES Abbreviations AASHTO American Association of State Highway and Transportation Officials FHWA Federal Highway Administration GAO Government Accountability Office NCHRP National Cooperative Highway Research Program AASHTO. 2009. Manual for Assessing Safety Hardware 2009. AASHTO, Washington, D.C. AASHTO. 2011. Roadside Design Guide, 4th ed. AASHTO, Washington, D.C. AASHTO-FHWA Measurement Task Force. 2015. AASHTO-FHWA Task Force on ET-Plus 4” Dimensions. March 11. https://www.fhwa.dot.gov/guardrailsafety/dimensionsreport. pdf. AASHTO-FHWA Measurement Task Force. n.d. FHWA Review of ET-Plus. https://www.fhwa. dot.gov/guardrailsafety/mtf.cfm. Bischoff, D., and I. Battaglia. 2007. ET-2000 End Treatment for Guardrail: Final Report. Wis- consin Department of Transportation. December. http://wisconsindot.gov/documents2/ research/wifep-03-07guardrailendtreatment.pdf. Federal Register. 2014. ET-Plus Guardrail End Terminal. Vol. 79, No. 247, Dec. 24, p. 77595. https://www.federalregister.gov/documents/2014/12/24/2014-30081/et-plus-guardrail- end-terminal. FHWA. n.d. Data Collection: In-Service Performance Evaluation of Guardrail End Terminals. GAO. 2015. Auto Safety: Status of NHTSA’s Redesign of Its Crashworthiness Data System. GAO-15-334. http://gao.gov/assets/670/668900.pdf. Heimbecker, C., and E. C. Lohrey. 2016. Examples of State Highway Agency Practices Re- garding Design, Installation, Maintenance, and Evaluation of Guardrail End Treatments. Background Paper for Special Report 323: In-Service Performance Evaluation of Guard- rail End Treatments. Transportation Research Board, Washington, D.C. Igharo, P. O., E. Munger, and R. W. Glad. 2004. In-Service Performance of Guardrail Terminals in Washington State. Report No. WA-RD 580.1. Washington State Depart- ment of Transportation, Olympia. June. https://www.wsdot.wa.gov/research/reports/ fullreports/580.1.pdf.

NATIONALLY COORDINATED EVALUATION RESEARCH 93 Johnson, N. S., and H. C. Gabler. 2014. Injury Risk in Frontal Crashes with Guardrail and Guardrail End Terminals. Presented at 93rd Annual Meeting of the Transportation Re- search Board, Washington, D.C. Joint AASHTO-FHWA Task Force on Guardrail Terminal Crash Analysis. 2015. Safety Analysis of Extruding W-Beam Guardrail Terminal Crashes. https://www.fhwa.dot.gov/ guardrailsafety/safetyanalysis. Mak, K. K., D. L. Sicking, and H. E. Ross, Jr. 1986. Real-World Impact Conditions for Run- Off-the Road Accidents. Transportation Research Record, Vol. 1065, pp. 45–55. Mak, K. K., D. L. Sicking, F. D. Benicio de Albuquerque, and B. A. Coon. 2010. NCHRP Report 665: Identification of Vehicular Impact Conditions Associated with Serious Ran-off-Road Crashes. Transportation Research Board of the National Academies, Washington, D.C. Marzougui, D., C. D. Kan, and K. S. Opiela. 2014. Assessing Options for Improving Roadside Barrier Crashworthiness. Presented at 13th International LS-DYNA Users Conference, Detroit, 2014, p. 9. http://www.dynalook.com/13th-international-ls-dyna-conference/ simulation/assessing-options-for-improving-roadside-barrier-crashworthiness. Michie, J. D. 1981. NCHRP Report 230: Recommended Procedures for the Safety Per- formance Evaluation of Highway Appurtenances. TRB, National Research Council, Washington, D.C. NCHRP. 2013. NCHRP 22-30 [Pending] In-Service Performance Evaluation of W-Beam End Terminals. http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3669. NCHRP. 2016. NCHRP 17-11(02) [Active]: Development of Clear Recovery Area Guidelines. http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=2586. Ray, M. H., and J. Hopp. 2000. Performance of Breakaway Cable and Modified Eccentric Loader Terminals in Iowa and North Carolina: In-Service Evaluation. Transportation Research Record: Journal of the Transportation Research Board, Vol. 1720, pp. 44–51. Ray, M. H., and J. Weir. 2002. Appendix D: In-Service Performance Evaluation Procedures Manual. Contractor’s Final Draft, NCHRP Project 22-13(2). Transportation Research Board of the National Academies. http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_ rpt_490aD.pdf. Ray, M. H., J. Weir, and J. Hopp. 2003. NCHRP Report 490: In-Service Performance Evalu- ation of Traffic Barriers. Transportation Research Board of the National Academies, Washington, D.C. Ross, H. E., Jr., D. L. Sicking, R. A. Zimmer, and J. D. Michie. 1993. NCHRP Report 350: Recommended Procedures for the Safety Performance Evaluation of Highway Features. TRB, National Research Council, Washington, D.C. van Schalkwyk, I., R. P. Bligh; D. C. Alberson; D. L. Bullard, Jr., D. Lord; and S.-P. Miaou. 2004. Developing an In-Service Performance Evaluation (ISPE) for Roadside Safety Features in Texas. Texas Transportation Institute, College Station. December. http:// d2dtl5nnlpfr0r.cloudfront.net/tti.tamu.edu/documents/0-4366-1.pdf.