Selection and Placement Guidelines for Test Level 2 Through Test Level 5 Median Barriers (2022)

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.

D-1   Probability of Passing Through, Over, or Under a Barrier (THRBAR) A P P E N D I X D CONTENTS Chapter 1 Introduction Chapter 2 Background Chapter 3 Simulated Encroachment Data Chapter 4 THRBAR for Guideline Development References

D-2 Selection and Placement Guidelines for Test Level 2 Through Test Level 5 Median Barriers CHAPTER 1 INTRODUCTION The ability to reasonably predict the number of penetrate-the-barrier, roll-over-the- barrier, and vault-the-barrier crashes (THRBAR) is critical to understanding barrier performance and selecting a barrier. Often, guidelines presented in past editions of the Roadside Design Guide (RDG) have presumed that barriers were 100% effective in preventing a vehicle from crossing to the other side of the barrier. While barriers are highly effective at minimizing penetration, roll over, and vault-over crashes, a small percentage of vehicles still cross the barrier line and interact with hazards shielded by the barrier. Often these penetration-roll over-vault crashes are severe. An accurate prediction of barrier performance must include a reasonable method for estimating the number of penetration, roll over, and vault collisions for a particular barrier. The proportion is assumed to be a function of the vehicle mix where heavier vehicles are more likely to breach the barrier; however, the crash data are dominated by passenger vehicles. Properly incorporating both heavy vehicles and passenger vehicles is important in selecting the appropriate test level barrier. As discussed here, a penetration implies a complete structural failure of the barrier that allows the vehicle to pass through. A roll-over-the-barrier collision is one where the vehicle rolls over the barrier to the field side, whereas a redirection roll over is one in which the vehicle rolls over after redirection while remaining on the same side of the barrier (i.e., the traffic side). A vault of the barrier is when a vehicle vaults over the barrier to the other side after impact. A study of the probability of a barrier breach (i.e., penetration, roll-over-the-barrier, or vault-the- barrier) given a crash with the barrier was undertaken using RSAPv3. The model development and analysis of the simulated data are documented in this appendix. This appendix documents a model developed from simulated RSAPv3 data to represent THRBAR for these guidelines. Vehicle type, barrier type, and barrier placement were evaluated as explanatory variables for THRBAR.

Probability of Passing Through, Over, or Under a Barrier (THRBAR) D-3   CHAPTER 2 BACKGROUND The RSAPv3 Engineer’s Manual developed in NCHRP Project 22-27 reviewed the advantages and disadvantages of employing a mechanistic approach or an empirical approach to estimating barrier penetrations. (Ray 2012) Briefly, most mechanistic penetration models assume that penetration occurs once capacity has been reached; however, this is only the beginning of the failure process. The barrier may contain and redirect the vehicle even though there are structural failures, as was determined by Ray et al. in their NCHRP Project 22-12(03) review of 50 full-scale crash tests on concrete bridge rails and median barriers. (Ray 2014) In other words, reaching capacity does not necessarily mean the vehicle will penetrate the barrier. One way to overcome this overly conservative assumption is by adopting an empirical approach using field-collected crash data. A complete understanding of the physics is not required since the data represents real observed events that incorporate a realistic range of impact conditions and material failures. Sufficient quantities of some vehicle and/or barrier types, however, are not available within the crash data (e.g., new MASH barriers), so the empirical approach is limited by the availability of crash data. There are, however, reliable data on vehicle mix. NCHRP Research Report 892: Guidelines for Shielding Bridge Piers (Ray 2018) recently studied the distribution of vehicle types and vehicle properties for the development of guidelines that are proposed for incorporation in the RDG. The variations and distributions of each vehicle group on different types of roadways were considered. This distribution of vehicle mix varies slightly based on the percentage of trucks (PT) in the traffic stream. The traffic mix findings from NCHRP Research Report 892 are shown in Table D-1. Table D-1 NCHRP Research Report 892 Traffic Mix Summary for Roadside Design (Ray 2018) Rural Urban F H W A V eh ic le C la ss Vehicle Category In te rs ta te s an d A rt er ia ls C ol le ct or s an d L oc al s In te rs ta te s an d A rt er ia ls C ol le ct or s an d L oc al s 1 Motorcycles Unknown Unknown Unknown Unknown 2 Passenger Cars 0.75(1-PT) 0.75(1-PT) 0.80(1-PT) 0.80(1-PT) 3 Pickups, Vans, and SUVs 0.25(1-PT) 0.25(1-PT) 0.20(1-PT) 0.20(1-PT) 4–7 Single-Unit Truck and Bus 0.25(PT) 0.70(PT) 0.35(PT) 0.80(PT) 8–10 Single-Trailer Truck 0.70(PT) 0.30(PT) 0.60(PT) 0.20(PT) 11–13 Multi-Trailer Truck 0.05(PT) 0.00 0.05(PT) 0.00

D-4 Selection and Placement Guidelines for Test Level 2 Through Test Level 5 Median Barriers CHAPTER 3 SIMULATED ENCROACHMENT DATA RSAPv3 simulations were generated to capture both the mechanistic and empirical calculations from the barrier breach module already coded within the RSAPv3 software. A subset of 114 available field-reconstructed passenger vehicle trajectories included in the analysis is identified in Table D-2 using the identifier from the research project in which the trajectories were gathered. (Mak 2010) Unfortunately, there are no field-reconstructed trajectories for heavy vehicles. Passenger vehicle trajectories were therefore limited to reasonable encroachment angles and speeds using previously documented methodologies (Ray 2017) to represent Single Unit Trucks (SUTs) and Tractor Trailers (TTs). The limitations resulted in assuming a data set of 38 trajectories to represent the SUTs and 30 trajectories to represent the TTs. These trajectories are also shown in Table D-2 marked with an asterisk or double asterisk for SUT and TT trajectories respectively. Table D-2 Passenger Vehicle Trajectories Used in Simulations 300421000** 134003385 146000704** 146003943** 209002222 657000589** 102002123** 134003586** 146001442 146003961** 209002882 657000595** 102002626* 134004265 146001661 146004363 209002883 659200356 129000676 134004706** 146001682 157008122 471400344 660500268 129000716 139001022 146001744 166002351 471600569 661100681 129001416** 139001781 146001761** 166002573** 471600589 661300240 129002054 139002264 146001764** 166003213 626300646 661300244 134000865** 139002302 146002182* 170001347** 655800263* 661300266* 134000908 139002405** 146002224** 170002267** 655800450* 661300402 134001646 139002542 146002723 170003586 655800471 778400222 134001707 139002925 146002742 200003532 655800566 819003805 134001745 139003243 146002884 207003902 655800684 881004121 134001987 139003481** 146003082** 207004182* 655800688 881004202 134002205 139003882 146003401* 207004284 655800691 916800203 134002365 139003961 146003481** 207004343 656500481 881004202 134002486 139004242 146003563** 207004742 656500681 916800203 134002507 139004343** 146003703 209000841 656500686* 134002627 139004501 146003746** 209001561** 657000372 134003026** 139004541** 146003821** 209001681 657000471 134003066 139004762** 146003843 209001764** 657000472 **SUT and TT trajectories * SUT trajectories The longitudinal barriers shown in Table D-3 were simulated at offsets from 1 foot to 99 feet from the travel edge in one-foot increments. Offset to the barrier was measured from the travel edge to the center of the barrier.

Probability of Passing Through, Over, or Under a Barrier (THRBAR) D-5   Name MASH Test Level Height (inches) Width (inches) Crash Data PRV (%) Energy Capacity (ft-lbf) Load Capacity (lbf) BT0 Concrete TL2 24 12 --- 47,000 43,200 BT1 Cable TL3 30 6 4.00 40,000 110,000 BT2 W-beam TL3 31 24 2.00 110,000 40,000 BT3 Concrete TL3 29 12 --- 47,000 90,000 BT4 Concrete TL4 36 12 --- 47,000 128,000 BT5 Concrete TL5 42 12 --- 47,000 264,000 The objective of this study was to model the probability of barrier breach for different barrier types, vehicle types, and barrier offsets. A simple relationship was evaluated for predicting barrier breach explained by the levels of vehicle type and barrier type as well as the continuous measure of offset. It was found that there is not a significant difference between TL3 Cable, TL3 W-beam, and TL3 F-Shape at the 95th percentile. These results show that the development history of test levels influences the ability to distinguish between test levels. It is therefore recommended that each test level of longitudinal barrier consider a single representation of THRBAR. Further consideration was given to the ability to differentiate between highway types, where the vehicle mix (i.e., mix of sedans, pickups, SUTs, and TTs) does vary, as shown in Table D-1. There is no statistically significant difference of predicted THRBAR values for a particular barrier type on any highway type considered. The attempted modeling showed it was not possible to distinguish between variations of THRBAR at different values of PT. This is believed to be due to the lack of available empirical data for heavy vehicles. Table D-3 MASH Barriers Studied

D-6 Selection and Placement Guidelines for Test Level 2 Through Test Level 5 Median Barriers CHAPTER 4 THRBAR FOR GUIDELINE DEVELOPMENT It is understood that not all vehicles will be contained under all impact conditions. It is presumed, however, for guideline development, that each test level contains the types of vehicles it was designed for. For example, TL4 barriers are specifically designed to contain SUTs, and TL5 barriers are specifically designed to contain TTs. The distributions of each type of vehicle can be represented by empirical data as shown in Table D-1. Using this equation as a model where PT is a number (i.e., not decimal), the lack of empirical data necessitated the following assumptions to estimate the value of THRBAR for each test level: TL2 All trucks penetrate, roll over, or vault a TL2 barrier. Passenger vehicles are contained when posted speed limits are less than or equal to 45 mph. TL3 Barriers contain passenger vehicles but all trucks breach the barrier. TL4 Barriers contain passenger vehicles and SUTs but all TTs breach the barrier. TL5 Barriers contain everything. It should be recognized that there are no assurances that all crashes of any type will be contained or will not be contained; however, these assumptions were necessary to differentiate among different test levels of barriers. Table D-4 shows values for the coefficient A based on the traffic mix documented in Table D-1. Table D-4 Values for THRBAR Coefficient A for Guideline Development Test Level A 2 1.00 3 1.00 4 0.75 5 0.000

Probability of Passing Through, Over, or Under a Barrier (THRBAR) D-7   REFERENCES Mak, King K., Dean L. Sicking, Francisco Daniel Benicio de Albuquerque, and Brian A. Coon, NCHRP Report 665: Identification of Vehicular Impact Conditions Associated with Serious Ran-off-Road Crashes, Transportation Research Board, Washington, DC, 2010. Ray, Malcolm H. and Christine E. Carrigan, NCHRP Project 22-12(03), "Recommended Guidelines for the Selection of Test Levels 2 Through 5 Bridge Railings," Transportation Research Board, Washington, DC, 2014. Ray, Malcolm H., Christine E. Carrigan, and Chuck A. Plaxico, "Heavy Vehicle Encroachment Trajectories," Presented at First International Roadside Safety Conference, San Francisco, CA, 2017. Ray, Malcolm H., Christine E. Carrigan, and Chuck A. Plaxico, NCHRP Project 22-27, "Roadside Safety Analysis Program (RSAP) Update, Appendix B: Engineer's Manual RSAP," Transportation Research Board, Washington, DC, 2012. Ray, Malcolm H., Christine E. Carrigan, and Chuck A. Plaxico, NCHRP Research Report 892: Guidelines for Shielding Bridge Piers, Transportation Research Board, Washington, DC, 2018.