Roadside Hardware Replacement Analysis: User Guide (2021)

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.

3 network safety. Appreciating the underlying risk reductions and costs of upgrading are fundamental to developing effective upgrading policies. The systematic data-driven process for roadside hardware replacement analysis described in this user guide is not meant to be prescriptive but informative. Engineers are likely to use this process as part of the decision-making process in combination with engineering judgement, agency experience, in-service performance history and other highway agency objectives. The method presented herein is just one of several considerations that should be included in making roadside hardware replacement decisions. A highway agency may also consider ease of maintenance issues, stock-piling materials, project scheduling, contractor familiarity with various designs and other issues. For example, an agency may elect to replace existing hardware that is otherwise in reasonably good repair because it is approaching the end of its service life and another project at the site in the near term is unlikely. Similarly, sometimes numerous spot repairs (e.g., blockout replacements, post replacement, post realignment, replacement of individual guardrail panels) might be needed to return a system to crash-ready condition and the agency may elect to simply replace with new hardware. The method presented herein, therefore is one facet of the decision-making process but could be used in conjunction with other considerations. 2 APPLICABILITY This Roadside Hardware Replacement Analysis User Guide can be used by highway agency engineers and their consultants when evaluating and planning individual or systemwide projects where existing roadside hardware is in good condition and appropriately located according the AASHTO Roadside Design Guide or highway agency roadside design policy but does not meet the evaluation criteria of MASH 2016 (8). A typical example of such projects include pavement restoration projects on roadways with existing guardrails and guardrail terminals where the roadside hardware is incidental to the primary objective of the construction. This method can also be used to develop policies for determining when damaged hardware should be upgraded to MASH 2016 or replaced in-kind. 3 PROCEDURE The roadside hardware replacement analysis process is comprised of 22 steps organized into four parts: • Relative risk assessment of existing hardware (Steps 1 through 8), • Relative risk estimate for replacement hardware (Steps 9 through 11), • Project details (Steps 12 through 17) and • Economic analysis (Steps 18 through 22). A brief description of each step is contained in Table 1 and a more detailed description of each step is contained in the following subsections. In executing this procedure, the analyst should assume the default decision would be to upgrade and replace the hardware. Where assumptions are necessary, the analyst should assume conditions most favorable to replacing the hardware. This ensures that if replacement is not recommended by the procedure, that the decision to leave existing hardware in place is not likely a good use of highway agency funds.

4 Table 1. Roadside Hardware Replacement Analysis Steps. Step Action Find the fatal and serious injury crash risk and police-reported crash rate for the existing hardware. 1 Enter the number of years of police-reported crash data obtained (YCD). 2 Enter the number of crashes of all severities involving the hardware under consideration (NCj) observed in the period reported in Step 1. 3 Enter the number of fatal and serious injury crashes involving the hardware under consideration (NFSCj) observed in the period reported in Step 1. 4 Enter the desired confidence interval (CL) for a two-tailed hypothesis test. The 85th percentile is recommended for design related decisions. 5 Calculate the mean fatal and serious injury crash risk (REHj) given a crash occurs with the existing hardware under consideration as Step3/Step2 . The confidence intervals (CI) shown to the left and right in the worksheet are calculated as: ⎣⎢⎢ ⎢⎢⎡CI = Step5 ∙ Z 2 ∙ Step21 + Z Step2 ± Z Step5 ∙ 1 − Step5N + Z4 ∙ Step21 + Z 𝑆𝑡𝑒𝑝2 ⎦⎥⎥ ⎥⎥⎤ 6 Enter the quantity of existing hardware in the crash reporting area (QEHj). Choose the appropriate units to the right in the worksheet where units can be in feet (e.g., guardrail, bridge rail, median barrier) or each (e.g., terminal, crash cushion, etc.). 7 Enter the amount of exposure in the crash reporting area over the period of data collection (EEHj). For continuous hardware (e.g., guardrails) report the exposure in million vehicle miles passing the hardware (MVMP) and for discreet hardware (e.g., terminals and crash cushions) report exposure in million vehicles passing the hardware (MVP). 8 Calculate the total police-reported crash rate for crashes of all severities with the existing hardware (CREHj) as Step8 = . Estimate the fatal and serious injury crash risk of an impact involving the replaced hardware. 9 Enter an estimate of the risk of a fatal or serious injury given a crash (RRHj) with the replaced hardware being considered occurs. 10 Determine if the difference between the fatal and serious injury crash risk for the existing and replacement hardware is statistically significant by comparing the value in Step8 to the range calculated in Step5. If the value in Step8 is within the range found in Step5, the difference is not statistically significant at the significance level entered in Step4. 11 Calculate the relative risk reduction (RRRj) of the replacement hardware with respect to the existing hardware as . Determine the costs of upgrading the hardware. 12 Enter the average annual daily traffic (AADT) on the project roadway in vehicles/day in the design year. 13 Enter the amount of replacement hardware in the project (QEHj) in the units selected (i.e., feet or each). 14 Enter the unit installed cost (UCIHj)of the replacement hardware using dollars and the units chosen in Step 6. 15 Enter the unit cost of removing the existing hardware (UCRHj) using dollars and the units chosen in Step 6. 16 Calculate the total installed cost of the replacement hardware (CRHj) in dollars as Step13 ∙ (Step14 + Step15) . 17 Estimate the annual number of fatal and serious injury crashes avoided (FSCAj) by installing the replacement hardware as ∙ ∙ ∙ ∙ ∙∙ where Units = 5280 for continuous hardware (e.g., guardrails) and 1 for all others. Repeat Steps 1 through 15 if there are other hardware items associated with the replacement hardware.

5 Table 1. Roadside Hardware Replacement Analysis Steps. (continued) Use the information collected in the previous steps to perform an economic analysis of the feasibility of replacing the existing roadside hardware. 18 Calculate the total cost of the replacement (TCRH) by summing all the Step16 costs. 19 Enter the value of statistical life (VSL) or use the default value of $12,300,000 based on the FHWA recommended procedure. 20 Calculate the annual societal benefit of crashes avoided (ABCA) by upgrading the hardware as Step 19 ∙ ∑ Step 17 . 21 Enter the assumed service life of the roadside hardware replacement (YL). Generally, this should be between 20 and 30 years. 22 Calculate the internal rate of return (IRR) for the service life indicated in Step21. The internal rate of return is the interest or discount rate, which makes the net present value of a replacement project zero. In other words, the interest rate makes the investment in the construction year equal to the present value of all future societal benefits of the crash reduction during the service life. The IRR can be found by solving the following equation: ∑ ( ) 3.1 ROADSIDE HARDWARE REPLACEMENT ANALYSIS WORKBOOK The analysis procedure described in the following sections was organized into a convenient Excel macro-enabled workbook, which is shown Figure 1. Each step in the workbook is indicated by an integer number at the left and the activity in the step is described (e.g., 1. Enter the number of years of crash data). The 22 steps needed to complete the procedure are listed in Table 1 with a brief explanation for each step. The self-calculating Excel workbook performs all the necessary calculations that are indicated in unshaded boxes. The yellow-shaded cells in the workbook indicate user input is required, the grey-shaded cells indicate a default value which the user can change if desired, and the unshaded cells are self-calculating or descriptive. The worksheet indicated by “Form” can be copied and pasted such that multiple replacement analyses can be stored in a single workbook. The “Example” tab corresponds to the example presented at the end of this user guide. The workbook includes three primary columns, one for each type of hardware being considered. For example, a typical project might include consideration of replacing guardrails and any associated terminals and transitions. Guardrail data could be placed in one column, terminal data in another, and transition data in the remaining column. It is useful to assign a specific name to each analysis task in the designated space at the top of the workbook in Figure 1. The name may include a route number, mile posts, or a narrative description of the work evaluated. The name is only used to distinguish one project from another and does not affect the results. Likewise, the yellow-shaded cell at the top of the three main columns should be used to identify the type of hardware that is being considered for replacement. The value entered is not used in the analysis but is helpful for organizing the results.

6 Figure 1. Roadside Hardware Replacement Analysis Workbook. Intructions: Enter information on the project or policy being considered in the yellow shaded cells. Values in grey shaded cells are default values but can be changed if desired. Project Name: Relative Risk Assessment of Existing Hardware Symbol 1. Enter the number of years of crash data: YCD 2. Enter the total number of crashes with the existing hardware: NCj 3. NFSCj 4. Enter the desired cofidence interval (e.g., 85th percentile): CL 85% 85% 85% 5. REHJ < < < < < < 6. Enter the quantity of existing hardware in crash reporting area: QEHJ 7. EEHJ 8. Calculate the total police reported crash rate. CREHJ Relative Risk Estimate for Replacement Hardware 9. Estimate of risk of fatal or serious injury given a crash with the replacement hardware: RRHJ 10. Are the existing and replacement crash risk outside the 85th confidence interval? 11. Calculate relative risk reduction of replacement hardware compared to existing hardware: RRRJ Project Details 12. Enter the design year AADT of the project: AADT veh/day - veh/day - veh/day 13. Enter the amount of hardware installed on project: QEHJ 0 0 0 14. Enter the unit installed cost of the replacement hardware: UCIHJ $/0 $/0 $/0 15. Enter the unit cost to remove existing hardware: UCRHJ $/0 $/0 $/0 16. Calculate the total cost of installing the replacement hardware: CRHJ - $ - $ - $ 17. Estimate the number of annual fatal and serious injury crashes avoided: FCSAJ crashes/yr crashes/yr crashes/yr Economic Analysis 18. Calculate the total cost of the hardware replacement: TCRH - $ 19. Enter the value of statistical life (VLS) or use the default value of $9,600,000: VSL 12,300,000 $ 20. Calculate the annual societal benefit of the replacement: ABCA - $/yr 21. Enter the assumed service life or retain the assumed 25 year life. YL 25 yrs Type CTRL+SHIFT+R 22. Calculate the internal rate of return assuming a 25 year design life. IIR % to Clear and Reset Form The expected number of fatal and serious injury crashes avoided calculated in the previous step are summed and assigned an economic value based on the FHWA's value of statistical life (VSL) recommendation. The internal rate of return where the sum of all the benefits in future years equals the construction investment in the design years is calculated. Generally, if the internal rate of return is less than 2% it is not the best use of ROADSIDE HARDWARE REPLACEMENT ASSESSMENT WORKSHEET Using state or district wide police reported crash data determine the number of fatal and serious injury crashes that occurred involving the existing hardware. If an in-service evaluation of the hardware is avaiable use the crash severity performance from that source. Use as many years of crash data as are available and enter the appropriate values in the yellow shaded cells. Three columns are provided to account for associated roadside hardware. For example, replacing a guardrail may also require replacing the associated terminals, anchors and transitions associated with that guardrail. Place information for each type of hardware in one of the three Enter the number of fatal and serious injury crashes with the existing hardware: Calculate the mean risk of fatal or serious injury with the existing hardware with 85th percentile confidence: Enter the existing hardware exposure during the data collection period (i.e., million vehicle miles passing [MVMP] for guardrails and million vehicles passing for terminals [MVP]). Estimate the smallest likely risk of a fatal or serious injury with the replacement hardware. Generally, crash data or an in-service evaluation will not be available to determine this value so it must be estimated using engineering judgement. It should be less than the observed value for the existing hardware and must be greater than zero. Enter the estimates for each type of replacement hardware below: Enter information about the proposed project. Use the project information along with the values determined in the previous sections to estimate the number of fatal and serious injury crashes avoided by replacing the existing hardware with upgraded hardware.

7 3.2 RISK ASSESSMENTS FOR EXISTING HARDWARE Knowing how well the existing hardware is performing is a necessary first step in the assessment of the roadside hardware replacement analysis process. Performance of roadside hardware in this analysis procedure is defined as the risk of observing a fatal or serious injury police- reported crash involving the subject roadside hardware given a crash occurs. Minimizing fatal and serious injury crashes is consistent with the design objectives of both the Roadside Design Guide and MASH 2016 (8, 9). 3.2.1 Step 1: Number of Years of Available Crash Data The analyst must first identify and gain access to the highway agency police-reported crash database covering the highway agency’s area of responsibility. Many highway agencies routinely have access to five years of police-reported crash data that can be used for safety analysis studies like those discussed in this guide. Ideally, more than one year of data should be used but there is generally no need to use more than 10. Five years of police data is usually sufficient and is recommended. Enter the number of years of available police-reported crash data (YCD) in the yellow-shaded cells for Step 1. 3.2.2 Step 2: Total Number of Crashes with Existing Hardware The number of crashes with the existing roadside hardware (NCj) used on the project shall be determined from the highway agency historical police report crash data identified in Step 1. The roadside hardware types listed on a typical police report are often generic. For example, the police report may indicate a guardrail was struck but may not indicate if the guardrail was a cable barrier, w-beam guardrail, or concrete barrier. Some highway agencies have inventory information that can be matched by location to the crash data such that the performance of each specific roadside hardware type can be determined. Other highway agencies may only be able to identify generic types of hardware. Still others may not even have the type of object struck listed with any specificity at all on the police report. As such, there are several methods for estimating the total number of crashes with each type of existing hardware considered. • Best approach – identify all the crashes in the police-reported data that involve a harmful event involving the roadside hardware of interest (e.g., guardrail, bridge rail, terminal, etc.) during the time period identified in Step 1. If a hardware inventory is available, match highway agency police-reported crashes with generic types of hardware to inventory locations of specific hardware and record the total number crashes for each type of specific roadside hardware. Also record the number of fatal and serious injury crashes for the subject roadside hardware for use in Step 3. • Acceptable approach – using the generic categories from the police-reported crash data, determine the total number of crashes involving the category of interest during the time period identified in Step 1. Also record the number of fatal and serious injury crashes for the subject roadside hardware for use in Step 2. • Adequate approach – if local data is not available, the only feasible alternative is to use data from another highway agency with more detailed data for the same types of roadside hardware of interest in the analyzed project. In such a case, use one of the two approaches listed above. Enter the total number of crashes for each type of roadside hardware in the yellow-shaded cells for Step 2.

8 3.2.3 Step 3: Number of Fatal and Serious Injury Crashes with Existing Hardware The number of fatal and serious injury crashes for each type of existing roadside hardware to be considered is a subset of the data collected in the Step 2. Fatal and serious injuries are assessed using the injury scale used on most police reports. The common definition of fatal and serious injury crashes is contained in the Model Minimum Uniform Crash Criteria (MMUCC) where fatal and serious injuries are defined as follows: Fatal Injury (K): A fatal injury is any injury that results in death within 30 days after the motor vehicle crash in which the injury occurred. If the person did not die at the scene but died within 30 days of the motor vehicle crash in which the injury occurred, the injury classification should be changed from the attribute previously assigned to the attribute “fatal injury.” Suspected Serious Injury (A): A suspected serious injury is any injury other than fatal, which results in one or more of the following: • Severe laceration resulting in exposure of underlying tissues/muscle/organs or resulting in significant loss of blood • Broken or distorted extremity (arm or leg) • Crush injuries • Suspected skull, chest, or abdominal injury other than bruises or minor lacerations • Significant burns (second and third degree burns over 10% or more of the body) • Unconsciousness when taken from the crash scene • Paralysis.” (10) The total number fatal and serious injury crashes involving each the types of roadside hardware (NFSCj) under consideration during the time period found in Step 1 (YDC) is determined using the police-reported crash data for all the roadways maintained and operated by the highway agency. Enter the total number of fatal and serious injury crashes (NFSCj) for each type of roadside hardware in the yellow-shaded cells for Step 3. 3.2.4 Step 4: Confidence Level The confidence level (CL) is a statistical parameter that describes the likelihood that the measurement observed adequately represents the true population value. For example, the percent of fatal and serious injury crashes involving guardrails will vary from year to year for a particular highway agency. In order to compare the performance of one roadside hardware to another, it is necessary determine if any observed differences are just random variation that occur from year to year or a real difference in performance. As more and more data are collected, the resulting performance measure can be relied on with increasing confidence. The 85th percentile CL is used in many types of engineering design calculations and is recommended as the default value for these analysis procedures. The 85th percentile represents relatively high confidence while acknowledging the many uncertainties regarding input values. The value chosen should be greater than 80 and less than 100 for conventional analysis. Enter the desired CL or accept the default 85th percentile value in the yellow-shaded cells for Step 4. The value is used to compute a two-tailed CI in Step 5 and a two-tailed hypothesis test in Step 10.

9 3.2.5 Step 5: Mean Risk and Confidence Interval The risk of a fatal or serious injury crash involving the subject existing hardware (REHj) is determined by dividing the number of fatal and serious injury crashes found in Step 3 by the total number of crashes found in Step 2 as follows: REH = NFSCNC → Step5 = Step3Step2 REHj is a direct measure of the performance of the existing hardware since it describes the risk of a fatal or serious injury based on the observed police-reported crash data obtained earlier in Step 1. The risk of a fatal or serious injury crash involving the subject existing hardware (REHj) is automatically calculated and entered in Step 5. Table 2 contains some mean risk values for a variety of guardrails, median barrier, bridge rails and guardrail terminals that were collected from the literature. The data were re-analyzed so that the 85th percentile CI could be listed. These data are based on police-reported first and only harmful events with the roadside hardware listed and most only consider passenger vehicles. Values of REHj measured and calculated for replacement projects are expected to result in similar values and are provided as convenient reference for the user. Table 2. Typical Mean Risk Values for Select Roadside Hardware. Roadside Hardware 85th percentile lower bound Risk of fatal or serious injury 85th percentile lower bound Source Guardrails and Median Barriers Report 350 Cable 0.0028 0.0050 0.0090 (11) Report 350 W-Beam 0.0068 0.0084 0.0103 (11) Report 350 Concrete 0.0137 0.0159 0.0185 (11) Bridge Railings Closed-Face Concrete 0.0137 0.0159 0.0185 (11) Terminals Report 350 Tangent Terminals 0.0204 0.0305 0.0455 (12) Report 350 Flared Terminals 0.0218 0.0482 0.1031 (13) Pre-Report 350 Terminals 0.0183 0.0455 0.1084 (13) As discussed in Step 4, differences in the performance measure REHj for two similar devices (e.g., 27-inch tall w-beam guardrail versus 31-inch tall w-beam guardrail) may be due to random variation that occurs from year to year or may indicate real performance differences. Determining which is most likely requires the calculation of a CI. If the estimate of one REH lies within the CI on another, the difference may be random variation and not a real difference in performance. For example, if REHA for guardrail A is 0.0094 and REHB for guardrail B is 0.0120 with an 85th CI of ±0.0050, the difference cannot confidently be attributed to a difference in performance but may just be the result of random variation since the value of REH lies within the confidence range of the other

10 device. On the other hand, if the 85th CI was ±0.0020 for guardrail B, the difference is likely a real difference in performance. In statistical terms, fatal and serious injury crashes are generally rare events (e.g., less than 0.10). The Wilson score interval is a confidence interval that is used to ensure that the CI does not go below zero. The Wilson score interval is a function of the number of cases (N), the estimate of the percent of total crashes that are fatal and serious injury crashes for the subject hardware (REHj) and the two-tailed Z score which is based on the CL as shown in Table 3. The Wilson score interval is given by (14): CI = RFEH Z 2NC1 + Z NC ± Z REH (1 − REH )N + Z4NC1 + Z NC CI = Step5 Z 2 ∙ Step21 + Z Step2 ± Z Step5 ∙ (1 − Step5)N + Z4 ∙ Step21 + Z 𝑆𝑡𝑒𝑝2 The CL (α) is used to calculate the CI for the percent of fatal and serious injury crashes (REHj) estimated for each hardware type (j). The ZCL score is a value determined from the choice of the CL (α) and is determined from the standard normal distribution assuming a two-tailed test. Values for ZCL for typical confidence levels (α) are shown in Table 3 but are also available in any standard statistical manual or textbook. The CI is automatically calculated by the roadside hardware replacement analysis workbook or it can be manually calculated using the above equation or looked up in Table 2. Table 3. Common Confidence Levels and Z Scores. Confidence Level (CL) Two-Tailed ZCL Score 0.99 2.58 0.95 1.96 0.90 1.64 0.85 1.44 0.80 1.28 3.2.6 Step 6: Quantity of Existing Hardware in the Crash Reporting Area The quantity of existing hardware in the same area used for collecting police-reported crashes (QEHj) is needed. This value may come from an inventory of roadside hardware maintained by the highway agency or it may be estimated. One way to estimate is to select several typical routes in the data collection area and inventory the hardware on those routes. The total amount of hardware can

11 then be extrapolated based on the total miles of roadway in the area. For guardrails, median barriers, and bridge railings the quantity is entered in ft whereas for terminals, transitions, and other discreet objects they entered in units of “each.” Enter the quantity of existing hardware in the same area used to collect police-reported crash data (QEHj) in the yellow cells for Step 6. Select the appropriate units in the yellow-shaded cell next to the quantity just entered (e.g., ft or ea.). 3.2.7 Step 7: Hardware Traffic Exposure Most highway agencies collect traffic volume data for a variety of operational and planning purposes. The objective of this step is to calculate the volume of traffic that passed the roadside hardware (EEHj) during the same period police-reported data was being collected for Steps 1 through 3. Obtain the highway agency traffic data and do one of the following: • Best Approach – If traffic volume data is available by route and milepost or location and there is an inventory of roadside hardware, the roadside hardware locations can be matched to traffic volume counts. For continuous hardware (e.g., guardrails and median barriers) multiply the length of the hardware in miles by the traffic volume at the site times the number of years of data collection from Step 1. This results in a count of the number vehicles passing during the data collection period in units of million vehicle-miles passing (MVMP). For discreet hardware (e.g., terminals, transitions, etc.) sum the volume at each discreet hardware location and multiply by the number of years of data collection from Step 1. The result is a count of the number of vehicles passing during the data collection period in units of million vehicles passing (MVP). • Acceptable Approach – Many highway agencies do not have roadside hardware inventory data so another less accurate but acceptable approach is to calculate the AADT in the data collection area and multiply that value by the quantity of existing hardware entered earlier in Step 6 and the number of years of data collected from Step 1. The result is the hardware traffic exposure in MVMP for continuous hardware (e.g., guardrails) and MVP for discreet hardware (e.g., terminals). One way to make the resulting calculation more accurate is to categorize roadways by their functional classification and tabulate the exposure by functional class. • Adequate Approach – If data was obtained from another jurisdiction to complete Step 2 then the same should be done for estimating the hardware exposure. Use either method in the two paragraphs above depending on the availability of data in the source data collection area. Enter the resulting existing roadside hardware traffic exposure (EEHj) in the yellow-shaded cells for Step 7. The units are automatically chosen based on the earlier response to Step 6. 3.2.8 Step 8: Police-Reported Crash Rate The estimated police-reported crash rate for the existing hardware (CREHj) is automatically calculated in the roadside hardware replacement analysis workbook shown in Figure 1 as follows: CREH = NCEEH → Step8 = Step3Step7

12 The police-reported crash rate (CREHj) is automatically calculated and entered into the box for Step 8. 3.3 RELATIVE RISK ESTIMATES FOR REPLACEMENT HARDWARE The most difficult part of the roadside hardware replacement analysis process is estimating the improvement resulting from installing the upgraded replacement hardware. 3.3.1 Step 9: Estimate Fatal and Serious Injury Crash Risk for Replacement Hardware Estimating the fatal and serious injury crash risk for the replacement hardware (RRHj) is difficult because there will generally be no field performance data available for newly developed devices. Instead, the analyst must make an estimate based on new design features, crash test results and engineering intuition. For example, many highway agencies are transitioning to 31-inch tall w- beam guardrail. The reason 31-inch guardrail was designed and tested was because of poor performance of 27-inch guardrail in Report 350 crash tests with pickup trucks. Since the purpose was to improve the performance with pickup trucks, the analyst could base the risk for the 31-inch guardrail on the anticipated improved performance with pickup trucks. This approach is illustrated at the end of this document in the example problem. Enter the value for the fatal and serious injury crash risk for the replacement hardware (RRHj) in the yellow-shaded cell for Step 9. 3.3.2 Step 10: Statistical Significance If the estimated fatal and serious injury crash risk for the replacement hardware falls within the CI range for the existing hardware calculated in Step 5 the differences between the risks are not statistically significant. If the replacement hardware risk is outside the range of the existing risk, there is likely a performance differences between the existing and replacement hardware. This information is not used in the analysis but is automatically calculated for the convenience of the analyst. 3.3.3 Step 11: Relative Risk Reduction The reason for considering replacement of existing roadside hardware with new hardware is to improve roadside safety performance by reducing the risk of observing a fatal or serious injury crash. The risk for the replacement hardware identified in Step 9 should be less than the calculated risk for existing hardware determined in Step 5. The relative risk reduction (RRRj) measures the reduction in risk as a percentage of the existing risk as follows: RRR = REH − RRHREHS → Step11 = Step5 − Step9Step5 The relative risk reduction (RRRj) associated with replacing the existing hardware with upgraded hardware is calculated and automatically entered in the box for Step 11. 3.4 PROJECT DETAILS The project details are the most easily obtained information since they are generally available on the initial project scoping documents or surveys. Information regarding the project details are entered in Steps 12 through 15. 3.4.1 Step 12: Design Year Traffic Volume The AADT in the design year for the proposed project may be determined using whatever method is used by the highway agency to make traffic projections for design decisions. The results

13 do not depend on how the AADT is estimated, only the value in the design year is required. One common method for projecting future traffic volume is simple geometric growth as follows: AADT = AADT ∙ (1 + G) Enter the design year traffic volume (AADT) in the yellow-shaded boxes for Step 12. 3.4.2 Step 13: Quantity of Replacement Hardware The quantity of each type of existing roadside hardware (QEHj) is determined for the proposed project based on the project scope or survey documents. Quantities are entered in ft for continuous roadside hardware like guardrails, median barriers, and bridge railings whereas quantities of “each” are used for discreet roadside hardware like guardrail terminals, transitions, crash cushions, signs, and other appropriate hardware. Recall that this procedure is only used for projects where replacement hardware is only considered where hardware already exists. Enter the quantity of existing hardware (QEHj) for each type of hardware in the yellow-shaded cells for Step 13. 3.4.3 Step 14: Unit Installed Cost of Replacement Hardware The unit installed cost of each type of replacement hardware (UCIHj) can usually be determined directly from the highway agency’s construction bid data. The cost should include the cost of the materials, labor, equipment charges, and any other expenses associated with installing a complete and final installation of the roadside hardware. For example, if the site of each new MASH terminal requires new grading, this cost should be included in the unit cost of the replacement MASH terminals. Enter the installed unit cost of each type of replacement hardware (UCIHj) in the yellow- shaded cells for Step 14. 3.4.4 Step 15: Unit Cost of Hardware Removal The unit cost of removing existing hardware (UCRHj) can also usually be determined as a bid item in the highway agencies construction cost data. All labor and equipment costs associated with removing the existing hardware must be included. Usually scrape value is not considered but if highway agency policy is to include the scape value of the removed hardware it should be deducted from the unit cost of hardware removal. Unit costs are in dollars/ft for continuous objects like guardrails and median barrier and dollars/each for discreet hardware like guardrail terminals, crash cushions, transitions, and signs. Enter the unit cost of removing each type of existing hardware (UCRHj) in the yellow-shaded cells for Step 15. 3.4.5 Step 16: Total Cost of Each Type of Replacement Hardware The total cost of replacing each type of hardware (CRHj) is the sum of the installation cost and the removal cost multiplied by the quantity of hardware. The total cost of replacing each type of replacement roadside hardware is calculated as: CRH = QEH ∙ UCIH + UCRH → Step16 = Step13 ∙ (Step14 + Step15) The total cost of replacement hardware (CRHj) is automatically calculated and shown in the box for Step 16.

14 3.5 ECONOMIC ANALYSIS The last portion of the analysis is an economic analysis where the expected annual number of the fatal and serious crashes avoided on the project are transformed into a dollar value and compared to the total cost of replacing the existing hardware with new upgraded hardware. 3.5.1 Step 17: Annual Number of Fatal and Serious Injury Crashes Avoided The reduction of risk resulting from installing the new replacement hardware is expected to result in fewer fatal and serious injury crashes. This reduction in the fatal and serious injury crash risk is the benefit that will be accrued to society in the future if the hardware is replaced. The number of fatal and serious injury crash avoided (FSCAj) is estimated as follows: FSCA = 365 ∙ CREH ∙ AADT ∙ QEH ∙ REH ∙ RRRUnits ∙ 10 Step17 = 365 ∙ Step8 ∙ Step12 ∙ Step13 ∙ Step5 ∙ Step11Units ∙ 10 The value of “units” in the above equation are 5280 for continuous hardware measured in ft and 1 for discreet hardware. The calculated annual number of fatal and serious injury crashes avoided (FSCAj) is automatically calculated and shown in the box for Step 16. 3.5.2 Step 18: Total Cost of Replacement Hardware The total cost of replacement hardware (TCRH) is simply the sum of all the hardware replacement costs (CRHj) listed earlier in Step 16 for each type of hardware. Most projects involve several types of hardware. For example, if guardrails are replaced, the associated terminals and transitions attached to the guardrails would likely also be replaced in the same project (i.e., m = 3). The costs of all hardware replaced on the project are summed to find the total project replacement cost (THRC) as follows. TCRH = CRH → Step 19 = Step 16 The TCRH is automatically calculated and shown in the box for Step 18. 3.5.3 Step 19: Value of Statistical Life According to Kniesner, “the VSL is the local tradeoff rate between fatality risk and money. When the tradeoff values are derived from choices in market contexts the VSL serves as both a measure of the population’s willingness to pay for risk reduction and the marginal cost of enhancing safety. Given its fundamental economic role, policy analysts have adopted the VSL as the economically correct measure of the benefit individuals receive from enhancements to their health and safety” (15). The U.S. DOT, including the FHWA, has used the VSL as a measure of crash cost since 2008 (16). The VSL can roughly be considered the cost of a fatal crash. A U.S. DOT memorandum on the use of the VSL states that “the relative values of injuries of varying severity were set as a percentage of the economic value of a life” (15). In 2008, the FHWA recommended using a VSL value of $5.8 million. The VSL has been periodically revised in subsequent years and in

15 2016 the recommend value was $9.6 million (17). The VSL increases from year to year and the FHWA provides a method for estimating a new VSL (17). Using the FHWA method for extrapolating the VSL to the year 2020 results in a VSL of $12.3 million, which is used in this procedure as the default value. The VSL is a method used to transform the number of crashes avoided into dollars so that an economic analysis can be performed. The methodology was specifically developed to assess the economic impacts of safety design decisions, so it is appropriate for use in roadside hardware replacement analysis. In these guidelines, the VSL is used to represent the benefit in dollars for both fatal and serious injury crashes. This is a conservative approach since it will tend to overestimate the benefit of replacing roadside hardware. If the highway agency uses its own locally derived crash cost values, the crash cost for a fatal crash can be substituted for the VSL. The analyst may retain the default VSL of $12.3 million shown in the grey-shaded box for Step 19 or enter another value if a different VSL is desired. 3.5.4 Step 20: Annual Societal Benefit of Crashes Avoided The ABCA by replacing existing roadside hardware with upgraded replacement hardware is calculated by summing the estimated number of annual fatal and serious injury crashes avoided (FSCAj) from Step 17 and multiplying by the VSL in Step 19 as follows: ABCA = FSCA ∙ VSL → Step 20 = Step 19 ∙ Step 17 The ABCA is assumed to be a constant value that is realized each year throughout the project life. The annual societal benefits of crash avoided by (ABCA) by replacing the hardware are calculated and displayed in the box for Step 20. 3.5.5 Step 21: Design Life Roadside hardware can remain crash-ready and functional in the field for a very long time. A design life of 25 or 30 years is not unreasonable if the hardware is properly maintained and repaired after a crash. A design life (YL) of 25 years is assumed in the procedure (Table 1), but the analyst could choose any other reasonable value based on local highway agency policy. The analyst may retain the default 25-year design life shown in the grey-shaded box for Step 21 or enter another value if a different design life is desired. 3.5.6 Step 22: Internal Rate of Return The IRR is the discount or interest rate that makes all net present worth values of the alternative zero. The IRR is the interest rate where the annual benefit of avoiding fatal and serious injury crashes by replacing hardware is exactly equal to the cost of implementing the replacement hardware alternative. In other words, the present worth of the annual benefit of fatal and serious injury crashes avoided (ABCA) over the life of the project exactly equals the total hardware replacement costs (THRC). The final step of the economic analysis is to calculate the IRR for the expected service life as follows: (18)