National Academies Press: OpenBook

Recommended Guidelines for Curb and Curb-Barrier Installations (2005)

Chapter: Chapter 5 - Analyses and Results

« Previous: Chapter 4 - Research Approach
Page 46
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 46
Page 47
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 47
Page 48
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 48
Page 49
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 49
Page 50
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 50
Page 51
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 51
Page 52
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 52
Page 53
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 53
Page 54
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 54
Page 55
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 55
Page 56
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 56
Page 57
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 57
Page 58
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 58
Page 59
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 59
Page 60
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 60
Page 61
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 61
Page 62
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 62
Page 63
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 63
Page 64
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 64
Page 65
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 65
Page 66
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 66
Page 67
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 67
Page 68
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 68
Page 69
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 69
Page 70
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 70
Page 71
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 71
Page 72
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 72
Page 73
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 73
Page 74
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 74
Page 75
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 75
Page 76
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 76
Page 77
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 77
Page 78
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 78
Page 79
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 79
Page 80
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 80
Page 81
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 81
Page 82
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 82
Page 83
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 83
Page 84
Suggested Citation:"Chapter 5 - Analyses and Results." National Academies of Sciences, Engineering, and Medicine. 2005. Recommended Guidelines for Curb and Curb-Barrier Installations. Washington, DC: The National Academies Press. doi: 10.17226/13849.
×
Page 84

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.

46 CHAPTER 5 ANALYSES AND RESULTS INTRODUCTION This chapter summarizes the analyses described in Chap- ter 4 and their results, as well as relevant analyses from prior studies. In most cases, significantly more detail was provided in one or more appendices or other publications. These results were used to develop the guidelines that were the pri- mary product of this research project, described in Chapter 6. PRIOR STUDIES The analyses of vehicle impact with curbs and curb–barrier combinations conducted in this study were limited to one vehi- cle type, a 2000-kg pickup truck. Thus, guidelines based solely on the results of those analyses would only be applicable to that one type of vehicle. In order to develop a more general set of guidelines, additional information was needed about the response of a broader range of vehicle types. The liter- ature provided an adequate amount of information on the response of various types of cars traversing curbs and also a limited amount of information from the results of full-scale crash tests regarding both cars and pickup trucks impacting curb–barrier combinations. There are many factors that influence vehicle behavior when traversing curbs, such as abrupt steering caused by the interaction of the front wheels with the curb; loss of contact between the tires and ground; excessive vehicle accelerations; and excessive roll, pitch, and yaw rates of the vehicle during impact. Although each of these factors may lead to loss of control of the vehicle, all the data that have been collected from full-scale tests and computer simulations suggest that total loss of control was unlikely except in extreme cases. A more important issue, however, may be the effects that these factors precipitate when curbs are placed in combination with roadside hardware (e.g., guardrail, crash cushions, or break- away poles). Even a slight increase in bumper height caused by traversing a curb may be sufficient to cause the vehicle to impact a roadside safety device at a point higher or lower than normal, which may lead to override or underride of bar- riers or adversely affect the breakaway mechanism of other roadside devices. Two of the studies identified in the literature review addressed the issue of override and underride indirectly using both full-scale testing and computer simulation: Olsen et al. (22) and Holloway et al. (24). In those studies the response of various types of cars traversing a number of dif- ferent curb types was obtained and the information was used to assess vehicle stability and to estimate the potential for barrier override and underride. Roll and pitch displacement- time histories and relative bumper trajectory–time history of vehicles traversing curbs were collected in their studies. Var- ious impact conditions and curb types were investigated in those studies, and all impact conditions were considered equally likely since data were not available to discern the most probable impact conditions of crashes. Only the maxi- mum values of angular displacement and bumper height from the various studies were considered when synthesizing the data for use in this study. The maximum encroachment angle of both the Olsen et al. study and the Holloway et al. study was 20 degrees, whereas the maximum encroachment angle used in the current study was 25 degrees. Also, since the vehicle used in the Olsen et al. study was a 1965 Ford four- door sedan, those results may not be representative of the current vehicle fleet. The results and conclusions from the study by Olsen et al., however, were similar to those obtained in both the Holloway et al. study and the current study. CRASH AND INVENTORY DATA ANALYSES This section describes six analyses of crash and inventory data that were conducted in this study. Extent of the U.S. Curb-Related Safety Problem The goal of this effort was to define the extent of the national safety problem related to curbs using FARS and NASS-GES data. A more detailed description of this task can be found in Appendix C. Table 10 presents FARS data for the 1994–99 period con- cerning how often “Curb” was noted as the FHE in fatal crashes on roads with speed limits of 40 mph (65 km/h) or greater for all crashes and SV fatal crashes. Table 11 presents NASS-GES national estimates for crashes of all severity on these higher-speed roads. These estimates are based on the GES weighting system, which was applied to the 50,000 cases collected each year. In both tables, the FHE differs from the

47 MHE in that the FHE was not necessarily the fatality or injury-producing mechanism. Curb impacts are actually very seldom the MHE in fatal crashes. As shown in Table 10, curbs on higher-speed roads were noted as the FHE in slightly over 1% of all fatal crashes on these roads each year; and, while varying somewhat, the fre- quency and percentage were relatively stable across the 6-year span. Curbs were noted as the FHE in approximately 2% of all SV fatal crashes on these roads. Table 11 shows that in terms of total crashes (fatal and nonfatal), curbs were noted as the FHE in fewer than 1% of all GES crashes each year, a rate even lower than that for fatal crashes. As with fatal crashes, while varying slightly, the frequency and percentage were relatively stable across the 5-year span for crashes reported in the GES database. Curb impacts were the FHE in approximately 2.5% of the SV crashes. These analyses were for roads with speed limits of 40 mph (65 km/h) or greater. If one included all speed limits, and thus urban streets where curbs are standard, the percentages for both fatal crashes and total crashes would increase, but not to a large extent. For example, in the 1999 FARS data, there are a total of 599 fatalities in which curb was the FHE on all roadways (regardless of speed limit). This represents approx- imately 1.6% of the total fatal crashes in 1999. Similarly, the 1999 weighted GES data for all speed limits indicate curb crashes represent approximately 1.0% of the crashes nation- wide. Clearly, curbs are the initial objects struck in only a small portion of fatal or total crashes on the roadways of interest and on all roadways. The remainder of the analyses in this task examined other factors related to these fatal and nonfatal crashes. All were restricted to the higher-speed roads (i.e., speed limits of 40 mph [65 km/h] or greater) and to the 1999 FARS and NASS-GES data. While only the highlights of the findings are included here, more detail is presented in Appendix C. • Curb crashes were more urban than other crashes: 72.3% of the curb-related fatal crashes were on urban roads, with 26.7% of the total on urban Interstates or other freeways/expressways. For the GES national estimates, almost half (49.5%) of the total higher-speed curb- related crashes were in urban areas with populations greater than 100,000, and 71.9% were in areas with pop- ulations greater than 25,000. The location of these curb- related fatal and total crashes differed significantly from the location for all SV crashes on these higher-speed roads: 71.2% of fatal SV crashes occurred on rural roads, and 61% of total SV crashes occurred on road- ways within areas of population less than 25,000. These findings probably reflect the fact that curbs were more often located in urban areas. • The MHE in fatal curb crashes was often a rollover, but the MHE in total curb crashes was the curb impact itself. Year Crashes with curb as FHE (SL>=65 km/h) Percent of total fatal crashes (SL>=65 km/h) Fatal single vehicle crashes (SL>=65 km/h) Fatal SV crashes with curb as FHE (SL>=65 km/h) Percent of fatal SV crashes (SL>=65 km/h) 1994 1995 1996 1997 1998 1999 27191 28005 28464 28171 28453 28527 291 387 396 391 325 329 1.1% 1.4% 1.4% 1.4% 1.1% 1.2% 14520 15206 15293 14906 15227 15062 271 345 359 362 308 307 1.9% 2.3% 2.3% 2.4% 2.0% 2.0% Total fatal crashes (SL>=65 km/h) TABLE 10 Fatal crashes with curb as FHE and speed limit 40 mph (65 km/h) or greater Year Total crashes nationwide (SL>=65 km/h) Crashes with curb as FHE (SL>=65 km/h) Percent of crashes with curb as FHE (SL>=65 km/h) SV crashes nationwide (SL>=65 km/h) SV crashes with curb as FHE (SL>=65 km/h) Percent SV crashes with curb as FHE (SL>=65 km/h) 1995 2765377 23680 0.9% 855097 21784 2.5% 1996 2857985 23470 0.8% 899940 21761 2.4% 1997 2839031 20107 0.7% 876545 18981 2.2% 1998 2781930 23908 0.9% 844783 23002 2.7% 1999 2753457 21807 0.8% 835853 20843 2.5% TABLE 11 National estimates of crashes with curb as FHE and speed limit 40 mph (65 km/h) or greater, weighted data

48 When curbs were noted as the FHE on these higher- speed roads, 38.9% of the 368 vehicles involved in fatal crashes were coded as having “Overturn” as the MHE. This was very similar to the 39.6% of all fatal SV crashes in which overturn was the MHE. Unfortunately, the data did not reveal whether the rollover was related to tripping on the curb or to an embankment or other object behind the curb. Only 18 (4.9%) of the 368 vehicles in fatal curb- related crashes, of which 15 were motorcycles, were coded as having the curb as the MHE. In contrast, for total crashes on these higher-speed roads, only 12.7% of the vehicles were coded as having “Overturn” as the MHE, and 50% were coded as having “Curb” as the MHE. The higher percentage of “Overturn” in the fatal data was the result of the nature of an overturn: once it occurs, it is likely to be fatal. • The curb-related fatal crashes occurred predominately at nonjunction locations (80.5%). An additional 10.6% were at interchanges, with the majority of these being on ramps. The total crashes were also more likely to be at nonjunction locations, but not to the same extent (53%). Here, approximately 26% were at intersections, and an additional 11.6% of the total crashes were at interchanges, mostly on ramps. • Pavement conditions (e.g., dry or wet) for fatal and total curb crashes were very similar to those for all SV fatal and total crashes: 90.3% of the curb-related fatal crashes occurred on dry pavement, 7.9% on wet pavement, and 1.2% in snow/slush/ice; 74.7% of the curb-related total crashes occurred on dry pavement and 21.8% on wet pavement. • There were only subtle differences in the vehicle maneu- vers prior to the crash (e.g., “going straight,” “changing lanes,” or “turning”) for the curb and total SV crashes for both the fatal and GES samples. • Vehicle types in curb-related fatal and total crashes did differ somewhat from vehicles in the comparable SV groups on these higher-speed roadways; they were more likely to be passenger cars (and motorcycles for the fatal subset), and somewhat less likely to be SUVs or pickup trucks. While these vehicle-related findings might be related to differential exposure (e.g., more passenger cars on urban roads where more curbs were located), they do not seem to indicate greatly increased curb- related problems for SUVs or pickups. Again, this con- clusion is tenuous given the lack of exposure data in both the FARS and GES files. In summary, curb-related fatal crashes on roadways with speed limits of 40 mph (65 km/h) and above represented a very small percentage of total fatal crashes (approximately 1%). Curb-related total crashes represented an even smaller percent of all crashes (less than 0.5%). Curbs were very sel- dom the MHE in fatal crashes (approximately 5%), but much more likely to be the MHE in total curb-related crashes (53%). This implies that curb impacts caused enough prop- erty damage to result in a reportable crash but that fatalities were more likely to result from a rollover. Finally, both fatal and total curb-related crashes differed from other SV fatal crashes on these higher-speed roadways in that they were much more likely to occur on urban roads and more likely to involve passenger cars rather than SUVs or pickups. Curb-Related Rollover Risk and Nature Given a Crash Rollover occurrence and risk is of particular interest when curbs are being studied, since the severity of impacts with this low-profile object would be expected to be related to whether a vehicle overturned rather than to energy exchange in the impact itself (unlike impacts with guardrails, for exam- ple). The FARS analysis described in the preceding section highlighted the fact that the MHE in most fatal curb-related crashes is a rollover. This set of analyses was conducted to further examine the incidence and nature of rollover in curb- related crashes. In order to help ensure that the curb was directly related to the rollover under study, databases chosen for use had to have a sequence of events that would allow examination of only those rollovers preceded by a curb impact. NASS-CDS data and the HSIS data from both Michigan and Illinois included such a sequence variable and were thus used in the analyses. This subsection presents the results of the NASS cases and the Michigan and Illinois cases separately. A detailed description of this analysis can be found in Appendix D. NASS-CDS Analysis and Results NASS-CDS data for the 1997–99 period were used in the analysis. Using the investigator-supplied sequence of events, cases were chosen that involved at least one impact with a curb on roads with posted speed limits of 40 mph (65 km/h) or greater. The resulting sample was very small, particularly for cases involving rollover. In the 3 years, there were 101 SV crashes involving a curb, and 38 of these involved a roll- over somewhere in the sequence. Of primary interest were those impacts in which a curb was the first event in the sequence (92 of the 101 curb-involved cases) and in which a rollover immediately followed the curb impact and thus was assumed to be related to it. As noted earlier, the NASS-CDS data are from an unequal probability sample extracted from police reports from across the nation and overrepresent more severe crashes. The data can be presented in two forms, unweighted and weighted. The unweighted, or raw, data represent the actual number of cases in the sample. The weighted data represent the total number of such cases that would have occurred nationwide, given that the sample and the assigned weights are accurate, that is, given that the sample cases as a whole do in fact reflect the national incidence of all such crashes. The weight

for each case is assigned by NASS. The weighted estimates are considered reliable when a large number of cases is being analyzed, but there are serious questions concerning the reli- ability of these estimates when relatively small samples are being studied, as is the circumstance here. Table 12 shows NASS-CDS data for the frequency of overturn in SV crashes in which the curb impact was the first event in the crash sequence. In this sample, three cases have extremely high weights, and those cases largely determine figures in the “Weighted” column. For this reason, a trun- cated version of the weighted data is also presented; it gives estimates based on the sample excluding the three very high weight cases. As shown in Table 12, the unweighted data indicate that in 17% of these SV crashes in which a curb was the first event, an overturn occurred immediately after the curb impact. In an additional 21% of these cases, an overturn occurred at some point in the crash sequence but could not be attributed to the curb impact. When the full NASS weights are applied to the same data, an overturn occurred immediately after the curb impact in 9% of the cases. An overturn occurred subsequently in the crash sequence but could not be attributed to the curb impact in an additional 6% of the weighted crashes. When the truncated weights are used, the overturn occurred in 13% of cases immediately after curb impact and in an additional 8% of the cases as a later event in the crash sequence. The rollover cases were further examined to see if the investigator noted “Curb” as the “Tripping object” in the 16 cases in which the rollover immediately followed the curb impact, as one would expect. This was only true in ten cases, with three other cases having “Ground” as the tripping object and the remaining three being uncoded. Thus, there is some lack of certainty concerning the percent of SV curb impacts resulting in an overturn. It would appear that such overturns occur in at least 7% of the curb impacts (based on the weighted data where the investigator noted “Curb” as the tripping object), and may be attributed to the curb impact in as many as 17% of the cases (based on the unweighted data, and assuming all overturns immediately fol- lowing the curb impact were caused by the curb). The best 49 estimate might be approximately 10%, based on the weighted and truncated-weight distributions. These curb-related cases were also examined to see if information could be extracted concerning vehicle impact speed and whether the vehicle impacting the curb was track- ing or nontracking. Unfortunately, the data did not provide such information. Michigan and Illinois Analyses and Results The HSIS databases for Illinois and Michigan used in this analysis included SV crashes that occurred in 1996 and 1997 on sections of roadways with curbs and posted speed limits at or above 40 mph (65 km/h). All such crashes were included in which a vehicle impacted a curb or another fixed object as the FHE or the first “substantial” harmful event. The latter subset included crashes in which the curb impact was not the first event, but was only preceded by nonobject events such as “uncoded or errors,” “loss of control,” or “ran off road left/right.” For each case, it was also noted whether the vehi- cle that struck the curb or fixed object was involved in an overturn subsequent to the impact with the curb or fixed object. For comparison with the NASS-CDS data, the first analysis involved the overall rollover percentage for Michigan and for Illinois. The crashes were then categorized by land use (i.e., urban or rural) and roadway classification (i.e., Interstate or non-Interstate) for each state. Since assigned operating speed for the roadway where the crash occurred is a combination of land use and roadway class, distributions of rollover per- centages were generated for four operating speed categories within each state. The assigned operating speeds were based on results from the analysis described in Appendix B. Table 13 presents the rollover percentages for each state. In the Michigan data set, 5% of the SV curb crashes resulted in a subsequent overturn, while in Illinois, 8% subsequently overturned. While the percentage of overturns in curb crashes is the same as for other objects in Michigan, the percentage of overturns in curb crashes is higher than for other objects in Illinois (8% versus 2%). Both are in the same range as, but slightly lower than, the 10% best estimate from the CDS data. Number of vehicles Percent of vehicles Incidence of overturn No weighting Full weighting “Truncated” weighting No weighting Full weighting “Truncated” weighting Did not overturn 57 30178 19530 61.96% 85.57% 79.33% Overturn immediately following curb impact 16 3109 3109 17.39% 8.82% 12.63% Overturn, not immediately following curb impact 19 1978 1978 20.65% 5.61% 8.04% Total 92 35265 24617 100.00% 100.00% 100.00% TABLE 12 Frequency of overturn in NASS-CDS SV curb impacts in which the curb impact was the first event in the crash sequence

Similar tables of rollover percentages for curb and other- fixed-object crashes were also produced for urban and rural Interstate and non-Interstate roadway classes. The samples of Illinois curb-related crashes were too small to be meaningful except for the urban non-Interstate category. The sample for the Michigan curb-related crashes in the rural Interstate cate- gory was also too small. In the other three categories (i.e., rural non-Interstate, urban Interstate, and urban non-Interstate), the Michigan data indicated that the curb-related rollover per- centages were very similar to the rollover percentages for other objects. For the urban non-Interstate higher-speed roads, the Illinois data indicated a higher curb-related rollover per- centage than was found for other objects that are struck first (7% versus 1%). When the data from the two states were combined, the curb-related rollover percentage was slightly higher on urban Interstates than on urban non-Interstates (8% versus 5%). The final analysis involved curb-related crashes classified by roadway operating speed for their crash location (see Table 14). As can be seen from Tables 13 and 14, the Illinois curb-related rollover percentages were consistently higher than the corresponding Michigan percentages. This could have been related to curb design or placement standards or to differences in crash reporting between the two states. If non- injury crashes in Illinois were systematically reported less often than in Michigan, the rollover percentage for Illinois would be higher since rollover crashes, which are more likely to result in injuries, were most likely to be reported fully in both states. In both Illinois and Michigan, the proportion of 50 curb crashes resulting in rollover appeared to increase as the assigned operating speed increased. Summary The early analyses of curb-related crashes on higher-speed roads indicated a relatively high frequency of “rollover” in FARS fatal curb-related crashes (40%), and a significant, though lower rollover percentage in the GES (all crashes) data (13%). Since neither of these databases included a sequence of events allowing a better link between the roll- over and the curb or other-fixed-object impact, NASS-CDS, Michigan, and Illinois data were analyzed. The relatively small sample size of curb-related crashes on higher-speed roads and the issue of weighting led to difficulties in draw- ing firm conclusions from the 1997–99 NASS-CDS data. The Michigan and Illinois data provided somewhat larger samples. Even though conclusions were difficult with the CDS data, the “combined estimate” of 10% rollover was sim- ilar to, but slightly higher than, the rollover estimates from Michigan and Illinois. The Michigan data indicated that the curb-related rollover percentages were very similar to the rollover percentages for other objects for the three roadway categories where adequate samples were found (i.e., rural non-Interstate, urban Interstate, and urban non-Interstate roads). The Illinois data for urban non-Interstate roads, the only category with adequate sample size, indicated a higher curb-related rollover percentage than was found for other objects (7% versus 1%). Finally, because of the small sample FHE = curb impact FHE = other fixed object impact State Did not overturn Overturned Percentage overturns Did not overturn Overturned Percentage overturns Michigan 1,487 83 5% 6,156 305 5% Illinois 361 30 8% 1,969 36 2% Total 1,848 113 6% 8,125 341 4% TABLE 13 Frequency of overturning vehicles in SV crashes in which either a curb or another fixed object was struck (Michigan and Illinois data, 1996–97) TABLE 14 Frequency of overturning vehicles in SV curb crashes categorized by roadway operating speed (Michigan and Illinois data, 1996–97) Michigan IllinoisAssigned operating speed Did not overturn Overturned Percentage overturns Did not overturn Overturned Percentage overturns NA 50 4 7% 46 1 2% 40-49 mph 193 6 3% 59 6 9% 50-59 mph 633 28 4% 172 13 7% 60-69 mph 597 40 6% 75 8 10% 70-79 mph 30 5 14% 9 2 18% Total 1503 83 5% 361 30 8%

sizes and poor quality of the raw data available for these cases, it was not possible to extract further information from the NASS-CDS data on vehicle tracking/nontracking prior to curb impact or vehicle impact speed, both of which would be presumably related to rollover risk. In summary, rollover after impacts with curbs appears to be a relatively low-frequency occurrence in all crashes. How- ever, it remains a problem worthy of design attention due to the severity of rollovers, as demonstrated by the higher rollover percentages in fatal curb-related crashes. Crash, Injury, and Rollover Rates for Guardrail Sections with and without Curbs Since rollover after striking a curb on higher-speed roads could be a significant cause of injury, different data sources were examined in an attempt to gather more information on rollover in the presence of curb and curb–guardrail combi- nations. The analyses described previously concerned the risk of rollover once a crash has occurred and therefore used crash data and a rollover subset within that data. The basic goal of the analysis described in this section was to examine curb–guardrail-related crash risk and rollover risk, which is similar to “crash rate” and “rollover rate,” per passing vehi- cle for segments of highway with guardrails and segments with curb–guardrail combinations. These guardrail and curb– guardrail sections were not compared to roadway sections without a curb or guardrail since reporting of crashes on the latter section is a function of the nature of the roadside beyond the shoulder, which is not in any roadway inventory file. A more detailed description of this effort can be found in Appendix E. To examine guardrail-related crash and rollover risk or rate per passing vehicle, a database was needed that allowed identification of specific segments of roadway with guard- rails and with curb–guardrail combinations that could be linked with run-off-road crash and rollover counts, AADT, and other characteristics of the roadway, such as road classi- fication, curvature, and speed limit. By definition, when one is attempting to compute “risk” or “rate,” the analysis record needed is a segment of highway, not a crash, since one must also include segments of highway which have had no roll- overs or crashes. The only database available for such an analysis, and probably one of very few such databases in the nation, was the Michigan HSIS database. While most states have roadway inventory files that include AADT and details of the cross-section of the roadway to the edge of the shoul- der, few include any information on guardrails. Michigan had developed and maintained a separate guardrail inventory file up through 1992. Each record identifies a section of guardrail. The inventory provided details of location (i.e., side of the highway), beginning and ending milepoints, and details of the guardrail such as type, end treatment, and off- set from the pavement edge. Since there was one record per 51 guardrail section, there could be multiple records referring to the same milepost on the highway, as a result of guardrails being on each side of the road or even in the median as well as on each side of the road. In contrast, the Michigan Road- way Inventory File was organized by homogeneous segments of roadway, a new segment beginning when any change occurred in a major variable (e.g., divided/undivided, shoul- der width/type, or lane width). When divided highways were present, there were separate inventory items for each direc- tion of travel, but the record included both directions. The presence of curbs on the roadway was found under the “shoul- der type” variable, and there were either two or four shoul- ders on each homogeneous segment, depending on whether the roadway was undivided or divided. Finally, the Michigan crash file had information on the crash milepost and the direction of travel of each of the vehicles, but did not specify the side of the divided roadway on which the crash occurred. The complicated nature of the guardrail file resulted in a complex data screening and merging effort involving a series of decisions (e.g., how to properly link crashes that are not mileposted to different sides of the roadway). The product of the significant data-preparation effort was an analysis file containing 1993–94 crash counts and 1992 AADT and other roadway characteristic data (e.g., the presence of a curb) linked to directional segments of 1992 guardrail for three highway classes: urban freeways, urban multilane divided roads, and urban multilane undivided roads. Only these classes contained sufficient mileage of both guardrail-only sections and curb–guardrail combination sections, and even in these classes the total directional mileage was limited (e.g., only 15 total miles of curb–guardrail combination sections on urban freeways). Two types of analysis were conducted: simple comparison of guardrail versus curb–guardrail crash rates per million vehicle-miles of passing vehicles within each of the three roadway classes, and regression modeling (i.e., Poisson and negative-binomial) in which AADT and other factors were better controlled for. Details of both analyses are presented in Appendix E. The crash rates developed for total SV crashes, injury- producing SV crashes, and SV rollover crashes are shown in Table 2 in Appendix E for all three roadway classes. How- ever, due to the small number of such crashes, only the rates related to urban freeways appeared to be somewhat mean- ingful; these are shown in Table 15. For the urban freeways, it appeared that the total run-off- road rate was slightly lower on guardrail-only segments than on curb–guardrail segments (0.175 versus 0.195 crashes per million vehicle-miles passing). This may have been due to the presence of the curb as another object to strike on the roadside, or to other factors that were not accounted for in these analyses (e.g., speed limit). Perhaps of more impor- tance, but with the same caveats, the injury crash rate in such crashes was also slightly higher where there was a curb pres- ent in addition to the guardrail. Finally, the rollover rate was

essentially the same for guardrail-only and curb–guardrail sections based on the simple rates. However, the comparison of rates such as these can be misleading unless the rates are from highway segments with essentially the same AADT, because the relationship between crashes and AADT is not linear in nature. The sec- ond analysis, statistical modeling, was intended to account for this issue. Poisson and negative-binomial models were developed to predict both SV crash and SV injury crash fre- quency on urban freeways as a function of a number of pre- dictor variables. Unfortunately, rollover crashes could not be analyzed separately due to the small sample size. Pre- dictor variables analyzed included curb presence, segment length, AADT, horizontal curve presence, speed limit, and guardrail offset. Tables 3 and 4 in Appendix E provide the detailed results. Because the Poisson results were similar to the negative-binomial results, and since the latter is consid- ered more appropriate, only the negative-binomial results are summarized here. In almost all cases, the predictor variables in the model of both total SV and injury SV crashes exhibited logical behav- ior (e.g., crashes increased with AADT and segment length and decreased with increasing guardrail offset). Of most interest, the presence of a curb with the guardrail signifi- cantly increased both total and injury SV crashes when all other factors were held constant. The total SV crash model predicted 0.1525 crashes per mile on average, when the inde- pendent variables were held at their means. Crashes increased by 0.0640 per mile (42%) when a curb was present or when a curb was added to a guardrail. The injury crash model, which was considered to be a surrogate of rollover crashes, predicted 0.0416 injury crashes per mile, increasing by 0.0238 (57%) when a curb was added. In summary, both the simple rate comparisons and the Poisson and negative-binomial models indicated that on urban freeways, segments with both guardrails and curbs were more likely to have both SV crashes and injury crashes than segments with only guardrails. While it was not possible to control for all potentially confounding variables, the fact that the models statistically control for exposure variables (e.g., segment length and AADT) and geometric/design variables 52 (e.g., curves and speed limits) strengthens these findings. The injury-crash model was considered to be a limited surrogate for a model of rollover crashes. The presence of a curb appeared to increase the crash frequencies on these urban freeway segments. Curb-Crash Severity Modeling The analyses described in the preceding two sections were related to rollover, one of the most important factors predict- ing crash severity. The analysis in this task was designed to provide additional information on curb-crash severity in both rollover and nonrollover crashes. As noted previously, it is difficult to study either crash occurrence or severity of curb impacts since the vehicle almost always overrides the curb, and both the occurrence of a reported crash and the resulting severity are often defined by what is behind the curb. Unfor- tunately, there was no good inventory of the area behind the curb in even the best databases (e.g., the HSIS roadway inventory files). The goal of this task was to compare the severity of all SV curb crashes with the severity of SV noncurb crashes (i.e., crashes with other roadside objects) to determine whether they differed under similar conditions. Since curb and non- curb crashes do not always occur under similar conditions, conditions were controlled through regression-type model- ing. To ensure that the curb was related to the subsequent injury or rollover, a database was needed that included a sequence of events. To equalize the roadside behind the curb with the roadside for noncurb crashes to the extent possible, a subsample was needed of noncurb crashes that occurred in areas similar to the curb crashes (i.e., crashes occurring on a “curb-type” roadway, but without a curb present in the crash). These requirements led to the use of the 1996-97 Michigan HSIS database, which contained both crash and roadway inventory information. A more detailed description of this effort can be found in Appendix F. The curb crashes included in the dataset were SV crashes involving a vehicle striking a curb as the first or first mean- ingful event in the sequence of events. Note that first mean- ingful included curb impacts as a second, third, or fourth event if all of the preceding events were nonobject/nonrollover events such as “loss of control” or “run off road.” The noncurb crashes occurred on segments of roadway with a curb present on at least one side of the roadway (i.e., opposite shoulder or median) according to the roadway inventory data, but not where the crash occurred, based on the absence of curb in the sequence of events. Crashes in which curb impacts were noted as an event following an impact with another fixed object or a rollover were deleted from the data set. In all cases, the analyses were restricted to roadways with posted speed limits of 40 mph (65 km/h) or greater. “Overturn” was captured as an event, and by definition, followed the curb impact in the curb-crash set. Other variables captured for analysis included the speed limit, assigned operating speed, Guardrail only Curb–guardrail Total Mileage 186.64 15.01 Average AADT 45,247 78,717 Total MVMT per Side 3064.1 442.1 Total SV Crash Freq. 537 87 Total SV Crash Rate 0.175 0.197 SV Injury Crash Freq. 139 29 SV Injury Crash Rate 0.045 0.066 SV Rollover Crash Frequency 31 5 SV Rollover Crash Rate 0.010 0.011 TABLE 15 Descriptors, crash frequencies, and crash rates per million vehicle-miles passing for guardrail-only and curb–guardrail segments on urban freeways in Michigan (1992 inventory data and 1993–94 crash data)

functional class, weather and light conditions, road surface, right shoulder type, highway area type, vehicle type, curve code, and terrain. As detailed in Appendix F, when one is attempting to model differences in the full distribution of crash severity (for example, by use of the KABCO injury scale: K = killed; A = severe injury; B = moderate injury; C = minor injury; and O = no injury) as a function of other variables (such as curb/noncurb or speed limit), the most appropriate model is an ordinal regression model. Two common forms are the ordered logit and probit. In this case, the logit form proved to be most appropriate. Models predicting severity were developed using the above set of variables. The primary model included speed limit and an urban/rural variable based on functional class as two of the independent predictors. A subsequent model used assigned roadway operating speed instead of these two variables, since assigned operating speed was a direct function of speed limit within functional class. Both of these models contained roll- over as a predictor and thus allowed controlling for rollover in examining curb versus noncurb severity. The speed limit–urban/rural model indicated that the effect of hitting a curb on injury severity was negative (i.e., it lowered injury severity), although this effect was only mar- ginally significant (at the 8% level). The model, which con- trolled for many other variables, showed that injury severity was higher in the following cases: • The vehicle rolled over; • The crash occurred on an urban rather than rural roadway; • The weather was clear or cloudy, not foggy, raining, snowing, sleeting/hailing, or severely windy; • The road surface was dry, rather than wet, muddy, snowy, slushy, or with debris; • The vehicles involved were trucks, buses, motorcycles, motor scooters and mopeds, rather than passenger cars, vans, or pickups; • The crash occurred on level terrain; • The crash occurred on a curve; or • The posted speed limit was at the higher end of the 40–65 mph range. The results of the second model in which assigned operat- ing speed replaced speed limit and rural/urban variables indi- cated that many of the same predictors were significant. However, in this case, while assigned operating speed was a significant predictor, curb presence was no longer significant. This model implied that there was no difference in crash severity between the curb and noncurb crashes. In conclusion, although the two models differed somewhat in their estimates of the effect of curbs on severity, both implied that in locations where curbs might be located, SV crashes involving curbs were clearly no more severe than crashes involving other roadside objects. Rollover was an important predictor of injury, perhaps the most important. 53 When the rollover variable and all other variables except curb presence were held constant, curb impacts were slightly less severe, or at least no more severe, than crashes with other objects. As noted earlier, there is no guarantee that the roadsides for these curb and noncurb crashes were similar, and if not, the severity difference found, or the lack thereof, may have been confounded by these unknown differences. However, given that this analysis was restricted to the most similar locations possible, those with a curb on at least one side of one roadway, the conclusion that curb crashes are at least no more severe, and probably less severe, than noncurb SV crashes appeared to be supported by the data. Nature of Curb Impacts One of the goals of the crash-data analysis effort was to develop or extract information from real-world crash data that might be useful in defining inputs to the simulation and crash-testing analyses. This was a two-part task. Crash Reconstruction Data The first part of this effort involved the analysis of NASS- CDS data to extract information on the specific nature of curb- related impacts in the real world (e.g., angle of impact, speed, tracking/nontracking). NASS-CDS was the only national data- base of crash reconstruction data where such detail was cap- tured. Police data did not include such information. Two sources of CDS data were used: basic data downloaded from the NHTSA website and enhanced data obtained from TTI. Details of the data, analyses, and results are found in Appendix G. Initially, NASS-CDS data for the 1997–99 period were downloaded from the NHTSA web site for analysis. The 11 separate files for each year (e.g., vehicle exterior file and event file) were combined into usable vehicle-based analysis files that allowed examination of the sequence of events for each vehicle in a crash and determination of when the curb was struck and what occurred after that impact. The NASS- CDS data contained up to 22 events (e.g., “hit curb”) for each vehicle involved in a crash. However, detailed information such as impact speed and impact angle was only recorded for one event in each crash, the event that caused the largest change in velocity. Data on the direction of force and defor- mation extent were recorded for the events that caused the highest and second-highest changes in velocity. Examination of the data after preparation indicated a major problem in the sample of curb-related cases: no impact angle or speed data were present, probably because CDS places higher priority on reconstruction of vehicle-to-vehicle impacts than fixed-object impacts. Of the 473 cases in which a curb impact was one of the events, including 32 cases in which the curb was the highest-change-in-velocity event,

none included reconstructed impact speed and angle data. Examination of the “direction of force” variable also indi- cated that it could not be a measure of “angle of impact.” This was verified in subsequent conversations with a NASS inves- tigation supervisor, who indicated that the data might be used as an indicator of tracking/nontracking vehicles, although that also proved later to be somewhat questionable. Because of these initial data problems, the researchers requested and received enhanced CDS data developed by Dr. Roger Bligh of TTI for NCHRP Project 17-11, “Deter- mination of Safe/Cost Effective Roadside Slopes and Asso- ciated Clear Distances.” In that project, TTI had NASS crash investigators capture additional data at selected CDS crash sites, and then reconstructed encroachment speed, angle, and tracking/nontracking information where possible. They also developed a confidence rating for the speed and angle recon- structions (i.e., 1 as low confidence and 10 as high confi- dence). TTI staff provided a set of 21 cases in which a curb had been struck for use in this study. All these cases were SV run-off-road (ROR) collisions; and all occurred on roadways with speed limits of 45, 50, or 55 mph (72, 80 or 89 km/h). Since the widest shoulder width was less than 6 ft, most of the curbs were apparently near the travel lane. For that rea- son, the encroachment data were expected to provide some indication of the speed and angle distributions for the curb impacts. In addition, whether the vehicle was tracking or not when it left the roadway was considered to be a good indica- tor of tracking during curb impact. Table 16 presents the reconstructed encroachment speed data, based on a very small sample of 14 curb-related crashes; since the encroachment speed was “unknown” in 7 of the 21 crashes, only these 14 crashes were relevant. In addition to the raw frequencies for all cases and a subset of cases with higher confidence ratings (i.e., 5 or higher), percentages within speed categories are presented for the total unweighted data, the total unweighted subset of cases with higher confidence, the weighted full sample and the weighted high-confidence subset. The weights in the latter two cases are those provided in the CDS data for each NASS-CDS case. As shown in the table, the encroachment speeds for this small sample of cases, which are estimates of the curb-impact 54 speeds, ranged from 15 to 61 mph. Except for the Total Weighted group, which was almost totally influenced by one case with a weight of 10,939, the distributions were some- what similar. Approximately 50% of the cases had encroach- ment speeds of between 35 and 45 mph. Similar tables related to reconstructed encroachment angle, combined speed and angle data, and tracking/nontracking status of the vehicle can be found in Appendix G. Both samples of NASS-CDS data available for use in these analyses of speed, angle and tracking/nontracking were very small. Thus, the results must be viewed with caution. This is particularly true of some of the weighted results in the TTI data, which were significantly affected by two high- weight cases. Given these important caveats, based on the data available, the following observations can be made: • Curb impact speeds ranged from 15 to 61 mph. Approx- imately 50% of the cases had encroachment speeds between 35 and 45 mph. • Curb impact angles ranged from 6 to 31 degrees. In the majority of the unweighted cases (70%), the angles were 15 degrees or less, with 50% between 11 and 15 degrees. The distribution of impact angles in the weighted data was highly dependent on the inclusion of the high-weight cases, with 26 to 83% of the cases having angles less than 15 degrees and 20 to 80% between 11 and 15 degrees. • According to the TTI (reconstructed) data, 77% of the vehicles in the unweighted sample and 97% of the vehi- cles in the weighted were tracking; while in the NASS- CDS data, 51% (unweighted) and 56% (weighted) were tracking based on direction of force. Analysis of Extreme versus Nonextreme Crashes The second part of this task involved analysis of the NASS-GES, Michigan, and Illinois data to determine if cer- tain curb-related crash conditions might distinguish extreme crashes (those involving fatal or incapacitating injury) from nonextreme crashes (those involving property damage only [PDO]). Three analyses were conducted in this effort: Unweighted data Weighted data Speed (mph) All cases High-confidence All cases High-confidence 15-20 1 7.1% 0 0.0% 96 0.8% 0 0.0% 20.1-25 0 0.0% 0 0.0% 0 0.0% 0 0.0% 25.1-30 2 14.3% 1 9.1% 11205 92.3% 266 25.0% 30.1-35 1 7.1% 1 9.1% 24 0.2% 24 2.3% 35.1-40 4 28.6% 4 36.4% 361 3.0% 361 34.0% 40.1-45 3 21.4% 2 18.2% 272 2.2% 225 21.2% 45.1-50 0 0.0% 0 0.0% 0 0.0% 0 0.0% 50.1-55 1 7.1% 1 9.1% 82 0.7% 82 7.7% 55.1-61 2 14.3% 2 18.2% 106 0.9% 106 9.9% Unknown 7 0 7 0 Total 21 100.0% 11 100.0% 12153 100.0% 1064 100.0% TABLE 16 Encroachment speed distributions from the TTI NASS-CDS sample

• A comparison of NASS-GES curb-related crashes result- ing in severe injury with curb-related crashes resulting in no injury, • A comparison of NASS-GES severe curb-related crashes with severe SV ROR crashes not involving a curb, and • An analysis of Michigan and Illinois severe and non- severe curb-related crashes. Details of the analyses are given in Appendix H. NASS-GES Severe and Nonsevere Curb-Related Crashes. The GES sample was drawn from the same police agencies as the NASS-CDS data described earlier, but the sample was much larger, approximately 50,000 cases per year. All 1995–99 GES crashes in which the curb was the FHE were divided into two groups: (1) all crashes involving fatal or incapacitat- ing injury, approximately 10 to 15% of the sample, and (2) all PDO crashes, approximately 50% of the sample. The crashes in each group were categorized by roadway class and speed limit (i.e., Interstate highways, non-Interstate highways with speed limits of 40 to 50 mph, and non-Interstate highways with speed limits greater than 50 mph). The sample sizes for these categories are shown in Table 17. The severe crashes were compared to the PDO crashes within the three roadway types for variables related to crashes (e.g., relationship to junction), vehicles (e.g., vehicle body type), and roadways (e.g., roadway profile). These analyses were conducted using only unweighted GES data because severity was the predominant weighting variable used in weighting. However, to verify the unweighted results, a set of analyses was conducted of the severe curb and non- curb crashes using weighted data. These analyses indicated that the overwhelming majority of the variables analyzed had very similar distributions for the weighted and nonweighted data, for both the severe and PDO crashes. Generally, the dis- tributions were within 2% of each other, and those outside this range exhibited differences of less than 5%. Therefore, using the nonweighted data appeared to be an appropriate method of comparison. Some consistent findings were noted and are included in the summary at the end of this subsection. NASS-GES Severe Curb-Related and Noncurb-Related SV Crashes. This second GES analysis compared extreme curb-related crashes to the larger group of extreme ROR crashes not involving a curb as the FHE. Again, the goal was to see if these high-injury curb and ROR crashes differed in 55 ways that might provide guidance to the simulation, crash testing, and policy development efforts. The data set used was similar to the one described above. The curb-related group comprised SV crashes in which (1) a curb impact was the FHE, (2) the posted speed limit for the roadway was 40 mph (65 km/h) or greater, and (3) the most severe injury was either fatal or incapacitating. The noncurb group included SV crashes meeting the same criteria except the FHE was not a curb impact. These sets of crashes were first characterized as Inter- states and non-Interstates, and the non-Interstate category was subdivided into multilane divided highways and multilane undivided highways to try to isolate groups more likely to have the same exposure to curb presence. While a number of different categorizations of the data were used in these comparisons, the findings did not add a significant amount of information to that learned from the earlier analysis of severe versus nonsevere curb crashes. It was difficult to identify clear findings because the curb and noncurb crashes might well be occurring at different types of locations (i.e., the curb locations could be somewhat dif- ferent from locations of crashes where no curb is involved). In addition, the freeway-related findings were based on very small samples of curb-related crashes. The more con- sistent patterns are included in a summary at the end of this subsection. Michigan and Illinois Severe and Nonsevere Curb-Related Crashes. Analyses similar to those described above were con- ducted with the 1996–97 Michigan and Illinois HSIS data. Criteria similar to those described for the GES analyses were employed: • A crash either involved at least one vehicle that struck a curb somewhere in the sequence of events or occurred on a segment of roadway equipped with curbs accord- ing to the roadway inventory data; • Either the FHE was an impact with a curb or the second, third, or fourth harmful event was an impact with a curb and the preceding events were nonimpact events such as “uncoded or errors,” “loss of control,” “ran off road left,” or “ran off road right;” • The posted speed limit was at least 40 mph; and • The maximum injury in the crash was either a fatality or an incapacitating injury (K or A on the KABCO injury scale used by most police departments) for the severe impacts or PDO for the nonsevere impacts. Roadway class and speed limit Fatal & severe injury crashes PDO crashes Interstate Highway (All Posted Speeds) 10 64 Non-Interstate Highway (Posted 40-50 mph) 105 428 Non-Interstate Highway (Posted over 50 mph) 17 113 TABLE 17 Sample sizes for NASS-GES analysis of extreme crashes in which the curb was the FHE

These severe and nonsevere curb crashes were categorized further as Interstate and non-Interstate crashes. The sample sizes for severe and nonsevere curb crashes on Interstates and non-Interstates are presented in Table 18. Because of the extremely small sample size of Interstate crashes in Illinois, those were not analyzed. Severe curb crashes differed from the nonsevere crashes on Michigan Interstates by occurring more often on ramps, in good weather, and involving alcohol use and motorcycles. Curbs were more often the MHE in the nonsevere Interstate crashes, with rollover and other impacts being the MHE in the severe crashes. The non-Interstate findings were some- what similar. Curb impacts on these roads were more likely to be in urban areas, regardless of severity. Severe crashes differed from nonsevere crashes in both states by occurring more often in clear weather on dry roads. The Michigan data again indicated that the curb was less likely to be the MHE in the severe crashes and that more alcohol use and more motorcycles were involved in the severe crashes. Illinois data for the non-Interstates indicated that the severe crashes occurred slightly more often at night. Comparison of Findings from the NASS-GES and State Analyses. The analyses described above examined a wide variety of crash-related factors that might differentiate among severe curb crashes, nonsevere curb crashes, and severe non- curb crashes. Both NASS-GES, a national database, and state- level data from Michigan and Illinois were examined. While there were some subtle differences among the results, there were some rather consistent findings related to curb design and placement: • Overturn was an important variable in terms of severe injury causation. Curb designs or placement that decreases the probability of overturn are clearly important. • Impact speed and angle were important; more severe curb-related crashes occurred at higher speeds, on grades, and on curved alignments. • Both in comparison with other curb crashes (in GES and state data) and in comparison with other SV ROR crashes on freeways, severe curb crashes were more often related to ramps. This could simply be because curbs were more likely to be located on ramps than on other road segments of freeways. However, it does underline the need for for- giving curb designs on interchange ramps. 56 • Severe curb crashes on Interstates and higher-speed non-Interstates were more likely to be in urban areas. This could reflect such factors as high-speed roadside encroachments at which more barrier curbs are present and the higher severity of curb crashes on interchange ramps, more ramps being located in urban areas. Design and placement may therefore be even more critical on higher-speed roads in urban areas. • Severe curb crashes were somewhat more likely to occur in clear weather on dry roads than less severe crashes were. • There was little difference between the curb and non- curb groups with respect to violations cited or whether the crash was considered speed related. The pattern of which of the groups had higher proportions varied by roadway type. However, the Michigan data appeared to indicate more alcohol use in the severe crashes. • There were no major differences between the frequency of rollovers in the severe curb-related and SV ROR crashes. The percentage of rollover was relatively high in both groups (18% and 70%, respectively). As expected, the mechanism for the rollover differed between the two groups, being the curb in the curb-related crashes. • Curbs were problematic for motorcycles. Summary of Crash and Inventory Data Analysis This section has described the analysis efforts involving real-world crash data that were included in this project. The goals of these analyses were (1) to better characterize safety problems associated with curb and curb–barrier combina- tions on high-speed roadways, and (2) to provide leads to the crash testing and simulation efforts that were conducted in other parts of this project. All efforts were ultimately aimed at the development of the design guidelines. The major findings concerning extent of the problem, curb-crash characteristics, and leads to simulation and crash testing efforts include the following: • Curb-related crashes on roadways with speed limits of 40 mph (65 km/h) and above represented a very small percentage of either total fatal crashes (1%) or all crashes (0.5%). The importance of the curb problem stems from the potential for rollover following impact. Michigan Illinois Severe (K+A) Nonsevere (PDO) Severe (K+A) Nonsevere (PDO) Interstate 17 185 2 26 Non-Interstate 63 1171 37 332 Total 80 1356 39 358 TABLE 18 Sample sizes of severe and nonsevere curb crashes in 1996 and 1997 Michigan and Illinois HSIS Data

• Curbs were very seldom the MHE in fatal crashes (5%) but much more often the MHE in total curb-related crashes (53%). This implies that curb impacts can cause enough property damage to result in a reportable crash, but that a fatality is more likely to result from a rollover. • SV curb-related crashes were clearly no more severe than crashes involving other roadside objects. Indeed, among crashes under similar conditions and controlling for rollover occurrence, curb impacts were slightly less severe than crashes with other objects. • Severe curb crashes were often related to ramps. • Severe curb crashes on Interstates and higher-speed non-Interstates were often in urban areas. • It did not appear that environmental or driver factors played a major role in curb crashes. Severe curb crashes were likely to occur in clear weather on dry roads. There was little difference between the curb and noncurb groups with respect to violations cited or whether the crash was considered “speed related.” As might be expected from other research on alcohol use and crash severity, the data from Michigan did appear to hint at more alcohol use in the severe curb crashes than in the less severe crashes. • Severe curb impacts involved passenger cars and motor- cycles more often than SUVs or pickups. • Rollover is a factor of interest in improving curb design and placement, since it clearly differentiated between a severe and nonsevere crash. Rollover after impacts with curbs appeared to be a relatively low-frequency occur- rence, but it remains a problem worthy of design atten- tion due to the severity of crashes involving rollovers, as demonstrated by the higher rollover percentages in fatal curb-related crashes. • On urban freeways, segments with both guardrails and curbs were likely to have both more SV curb–guardrail crashes and more SV curb–guardrail injury crashes than segments with only guardrails. Assuming this relation- ship holds for other roadway classes, which it should for at least rural Interstates, it clearly supports attention to the design and placement of curbs in combination with guardrails. • Even though a relatively large number of crash recon- structions are developed each year in the NASS-CDS system (i.e., 5,000 per year), there was very little infor- 57 mation available on key issues, such as vehicle impact speed or angle for minor fixed objects like curbs; the need to weight such small samples resulted in interpre- tation problems. • Based on an enhanced sample of 21 curb-related cases on higher-speed roads, it appeared that there was a wide range of impact speeds (i.e., 15 to 61 mph), with approx- imately 50% between 35 and 45 mph; that there was a wide range of impact angles (6 to 31 degrees) with a sig- nificant proportion falling between 11 and 15 degrees; and that 77% (unweighted) or 97% (weighted) of the vehicles in these crashes were tracking, depending on whether the unweighted or weighted data are used. VEHICLE CURB TRAVERSAL SIMULATIONS AND TESTS The kinematic behavior of a vehicle traversing a roadside curb is the primary focus of this section. The modified NCAC finite-element model of the C2500 pickup truck was used to investigate the vehicle’s response when crossing a number of different curb types at various impact conditions. The results of the simulations were verified through a series of live-driver curb traversal tests with a C2500. Curb Traversal Simulations The case of a vehicle impacting a curb in a tracking man- ner was investigated using LS-DYNA. This section is a brief overview of the methods and results of the FEAs; much more detail can be found in a dissertation by Plaxico (52). Parametric Study A parametric study was conducted using different curb types, impact speeds, and impact angles, as shown in Table 19. The curb types included in the study were AASHTO Types A, B, C, D, and G and the 100-mm New York curb, which is referred to in the tables and figures as “New York” or “NY.” These curbs are illustrated in Figure 29. Two impact speeds were used: 70 km/h, which corresponds to the intermediate speed range of interest (i.e., 60 to 80 km/h), and 100 km/h, Impact speed = 70 km/h Impact speed = 100 km/h Angle of impact: Angle of impact: Curb type 5° 15° 25° 5° 15° 25° A T T T B T T T T T T C T T T T T T D T T T T T T G T T T T T T New York T T T T T T TABLE 19 Simulation matrix of impact speed, angle of impact, and curb type for the 2000-kg pickup impacting in a tracking manner

which corresponds to the high speed range (i.e., greater than 80 km/h). Three angles of impact were investigated: 5, 15, and 25 degrees. Impact angles of 5 and 15 degrees repre- sented the more probable range, while the 25-degree impact was consistent with NCHRP Report 350 impact conditions for longitudinal barriers. The vehicle used in the simulations was the 2000-kg C2500 pickup truck model developed by NCAC with modifications made by WPI. Only one roadway cross-section, shown in Figure 30, was used in the parametric study. A typical two-lane cross-section in which a curb and a barrier may be installed together was chosen: a road surface with a 2% cross slope and 1.1-m wide gutter section with a 4.5% slope. Although a 4% backfill would have been typical, a level roadside (i.e., 0% slope) was used in this study since it was more conservative in terms of the potential for the vehicle to override a barrier. The road- way and curb were modeled using shell elements with rigid material properties, and the tire-ground interaction was sim- ulated by using a surface-to-surface contact definition that included friction. The friction values used were based on results from physical tests in which a C2500 pickup truck was pulled at low speed over an asphalt surface and a con- 58 crete surface with its wheels locked and the force required to pull the truck was measured using a hydraulic scale. Data collected from the simulations included bumper tra- jectories and vehicle paths; acceleration-time histories; yaw, pitch and roll-time histories; yaw, pitch, and roll angle rate- histories; sequential snapshots; and TRAP results (i.e., occu- pant risk values). Results Bumper trajectories and vehicle paths. Figure 31 shows the bumper height of the pickup truck during impact with the different curbs for each impact angle studied. From the results of the simulations, the following observations were made regarding the potential for barrier override. Influence of lateral offset. For the 150-mm curbs (i.e., AASHTO types A, B, and D), there was a potential for bar- rier override if the barrier was positioned within 8 m behind the curb. For the 100-mm curbs (i.e., AASHTO types C and G and the NY curb), the potential for barrier override appeared 15 0 m m 25 m m R5 0 m m 44 m m (B) 50 mm 75 m m 10 0 m m R25 mm 125 mm 50 mm 130 mm 100 mm 10 0 m m 300 mm (D) (NY) (C) 150 mm R75 mm R7 5 m m (A) 150 mm 10 mm 305 mm (G) R1 0 m m 15 0 m m 15 0 m m 10 0 m m Figure 29. Curb types used in this study. Figure 30. Roadway cross-section used in this study.

to be less if the barrier was positioned between 2 m and 3 m or more than 8 m behind the curb. However, in the case of the 25-degree impacts, the trajectory of the front bumper continuously increased over a lateral distance of approxi- mately 4 m behind the curbs. Influence of impact conditions. The trajectory of the front bumper was nearly independent of vehicle speed but slightly dependent on impact angle (it increased with increase in impact angle; less so for the 150-mm curbs). For 100-mm curbs the potential for barrier override was minimal for impact angles of 5 and 15 degrees. For a given impact speed and angle, the mode of vehicle trajectory was similar for all curb types (for a given impact speed and angle, the maximum bumper trajectory occurs at approximately the same point, regardless of curb type). Influence of curb type. The maximum value of bumper tra- jectory was dependent on curb height (i.e., increased with increase in curb height). It was also slightly dependent on slope of curb face, with some discrepancy in the results from the AASHTO curb type A analyses. The accelerations and angle-displacement rates computed at the center of gravity of the vehicle model were extracted 59 from the results of the FEAs and input into TRAP. From these data, occupant risk factors were computed based on OIVs and ridedown accelerations and were found to be minor, as expected. The primary purpose for using TRAP, however, was to obtain information that would aid in quantifying vehi- cle stability regarding the various curb types. The stability of the vehicle during and after interaction with curbs may be adversely affected by wheel interaction with the curbs. For example, the front wheels of a vehicle may undergo abrupt steering during impact with a curb which may eventu- ally lead to spin-out or overturn of the vehicle. TRAP pro- vides information based on maximum accelerations, maximum angle displacements, and maximum displacements rates that may be useful in discerning vehicle instability. The results from TRAP for each analysis in the study matrix are pre- sented in Table 20. Acceleration-Time histories. An Acceleration Severity Index (ASI) value was computed from each analysis. These values were relatively low concerning occupant risk during impact, but they did give an indication of the overall acceleration response of the vehicle during vehicle-curb interaction, which may be regarded as some measure of difficulty for a driver to maintain control of the vehicle. For example, a higher ASI Figure 31a. Bumper height with respect to lateral distance behind curb and curb type for C2500 pickup crossing the curb at an angle of 5 degrees at 70 km/h (top) and 100 km/h (bottom). Figure 31b. Bumper height with respect to lateral distance behind curb and curb type for C2500 pickup crossing the curb at an angle of 15 degrees at 70 km/h (top) and 100 km/h (bottom).

value indicates that the vehicle experienced higher accelera- tions, which could affect the driver’s ability to maintain con- trol of the steering and braking of the vehicle during impact. Figures 32 and 33 show a comparison of the ASI for each analysis; note that curb types A, B, and D are 150-mm curbs and curb types C, G, and NY are 100-mm curbs. Figures 32 and 33 point to the following conclusions about ASI values: • They increased as impact velocity increased. • They increased as impact angle increased. • They increased as the curb height increased. • They increased as the slope of the curb face increased. Yaw, pitch, and roll. Figure 34 shows the maximum angu- lar displacements and maximum angular displacement rates from each analysis case. The following observations were made from the analyses: • Roll angles were minimal in all cases (i.e., less than 8 degrees). They appeared to be unaffected by the slope of the curb face, especially at higher impact speeds, and almost unaffected by impact speed. The roll angle 60 increased as curb height increased and decreased as impact angle increased. • Roll rates were also independent of the slope of the curb face, and they were minimally affected by impact speed. The roll rate increased as curb height increased and as impact angle increased. The influence of impact angle on roll rates was much more pronounced for cases involving 150-mm curbs than for cases involving 100-mm curbs. • Pitch angles were minimal in all cases (i.e., less than 3.5 degrees). They appeared to be independent of impact speed and slope of the curb. The pitch angles increased slightly as curb height increased. • Pitch rates were independent of the impact speed and slope of the curb face. For the 150-mm curbs, pitch rates varied significantly with respect to impact angle; for the 100-mm curbs, the pitch rate was much less influenced by impact angle. • Yaw angles were primarily affected by the steer angle of the front wheels after impact with the curb. As the front wheels steered out, usually to the right, during wheel- curb interaction, the yaw angle increased and was typi- cally greatest at the end of the analysis. As the vehicle tra- versed the curb, the resulting yaw angle of the vehicle could lead to an impact with the barrier at a higher or lower impact angle than the original encroachment angle. • Yaw angles were independent of curb height and impact speed but increased as the slope of the curb face increased. For the 150-mm curbs, the yaw angle ranged from 8 to 28 degrees and varied erratically with respect to the impact angle. For the Type C curb, the yaw angle ranged between 9 and 24 degrees; for the G curb, the angles were very low (3 to 10 degrees) except for the high-speed, high-angle impact for which the maximum yaw angle was 22 degrees; for the NY curb, the angles were also very low (i.e., 3 to 6 degrees, and negative 8 degrees in one case) except for the high-speed, high- angle impact for which the maximum yaw angle was 18 degrees. • Yaw rates increased as the height of curb increased and as the slope of the curb face increased. For the 100-mm curbs, the yaw rate was independent of impact speed but increased slightly as the impact angle increased. For the A curb, there was no discernable effect of the impact angle on the yaw rate; for the B curb, the yaw rate var- ied significantly and erratically with respect to impact speed, while the impact angle had minimal influence except for the high-speed, high-angle impact; for the D Curb, the yaw rate increased as impact angle increased and increased slightly as impact speed increased. Summary The FEA program LS-DYNA was used in a parametric study to investigate the influence of several factors regarding Figure 31c. Bumper height with respect to lateral distance behind curb and curb type for C2500 pickup crossing the curb at an angle of 25 degrees at 70 km/h (top) and 100 km/h (bottom).

vehicle stability and trajectory when traversing curbs. The variables used in the study included curb height and shape, impact speed, and impact angle. The results of the study indicated that the trajectory of the front bumper was only slightly affected by impact speed, impact angle, or the slope of the curb face. The most signif- icant factor influencing trajectory was the height of the curb. Based on the range of impact conditions considered in this study, the trajectory of a 2000-kg pickup truck traversing curbs with a height of 100 or 150-mm was considered suffi- 61 cient to override a standard strong-post guardrail placed at 0.5 to 8 m behind the curb. Acceleration and angular rate data collected at the center of gravity of the vehicle model during analysis were used as inputs to TRAP. The results indicate that ASI values were proportional to impact speed, impact angle, curb height, and the slope of the curb face. This suggests that a driver was much less likely to lose control while traversing a lower curb with a more mild, sloping face (e.g., the New York curb) than while traversing a taller, steep-faced curb such as the AASHTO Type A or B. The analysis showed that vehicle impacts with roadside curbs could often result in the driver losing control of the vehi- cle. There were many factors that influenced vehicle behavior during curb traversal, such as abrupt steering caused by the interaction of the front wheels with the curb, loss of contact between the tires and the ground, excessive vehicle accelera- tions, and excessive roll, pitch, and yaw during impact. While each of these factors may lead to loss of control of the vehicle, total loss of control is unlikely except in extreme cases. A more important issue may be the effects that these factors precipi- tate when curbs are placed in combination with roadside hard- ware (e.g., guardrail, crash cushions, or breakaway poles). Vehicle behavior in impacts with curb–guardrail combinations are discussed later in this section. 15 7 73 4 01 1 36 2170 0 20 6 5 3 1 27 4 77 0 50 3 40 9 25 9 41 4 11 1 77 3519 0 22 5 4 2 4 12 0 82 3 69 0 38 0 5 5 08 1 00 2 57 1785 0 11 6 9 2 4 20 2 46 1 28 3 47 2 15 7 38 3 88 1 42 5238 0 19 6 6 3 3 25 2 67 5 54 4 47 2 25 6 59 5 47 1 92 3453 0 25 5 4 2 8 26 9 116 0 34 1 44 3 5 4 44 2 72 1 07 2523 0 19 7 6 2 3 21 4 62 4 27 8 34 6 15 6 79 4 65 1 39 2517 0 22 5 0 2 6 20 0 80 7 39 9 37 2 25 14 93 10 00 2 84 4284 0 29 4 2 2 4 23 1 97 5 71 6 57 1 5 1 50 1 25 0 90 1506 0 10 7 4 2 2 11 1 45 8 17 9 22 7 15 3 58 2 63 1 31 1990 0 14 5 4 2 6 8 4 63 4 34 4 32 7 25 7 56 5 57 1 75 3349 0 21 5 2 2 7 28 1 100 7 38 8 39 7 5 2 55 1 40 0 87 2115 0 14 7 1 1 8 7 8 45 3 13 4 26 9 15 5 78 4 51 1 17 2443 0 19 5 3 2 8 24 6 72 9 37 7 37 6 25 11 41 7 19 2 54 3772 0 26 4 2 2 4 23 8 95 8 57 5 51 1 5 1 30 0 97 0 63 1318 0 09 6 0 1 6 12 6 37 9 12 2 18 9 Impact conditions Max. vertical acceleration (G’s) Max. angle displacements (degrees) Max. angle disp. rates (deg/s) Curb type Speed (km/h) Angle (deg) 60 Hz filter 10 ms average 50 ms average Max. vertical impulse (N*s) ASI Roll Pitch Yaw Roll rate Pitch rate Yaw rate 5 6.31 3.21 1.05 1618 0.18 -6.4 3.0 22.6 68.2 36.2 35.2 A 70 70 B 100 70 15 0- m m cu rb s D 100 15 2.86 1.73 0.95 1557 0.11 -4.2 2.1 11.4 36.4 22.9 21.7 70 25 3.93 2.42 1.20 2411 0.16 -3.9 2.5 23.7 48.2 23.0 28.7 5 1.50 1.03 0.70 1594 0.09 -5.7 1.4 9.1 36.8 12.1 17.9 15 3.20 2.06 1.00 1990 0.15 -3.8 2.3 22.8 50.1 22.8 23.5 C 100 25 5.86 4.54 1.25 2500 0.18 -3.4 2.0 23.7 61.1 28.0 27.8 5 0.83 0.77 0.61 1097 0.07 -5.9 1.6 6.4 35.9 12.3 15.8 15 2.17 1.41 0.85 1811 0.09 -4.0 2.2 4.1 36.1 17.9 17.5 70 25 4.20 2.73 1.14 2319 0.14 -4.1 2.7 9.8 54.7 24.6 20.1 5 0.99 0.89 0.74 1589 0.08 -5.4 1.3 3.3 37.7 12.1 13.1 15 2.59 1.81 1.06 1973 0.12 -4.0 2.4 6.7 47.8 26.6 16.8 G 100 25 7.81 5.87 1.81 2585 0.19 -3.6 2.2 21.7 66.2 26.9 29.0 5 0.43 0.34 0.26 796 0.05 -4.9 1.2 4.0 27.8 12.2 15.8 15 1.30 0.84 0.63 1188 0.07 -3.8 2.1 3.0 32.3 14.5 12.5 70 25 3.10 2.00 1.11 2017 0.14 -3.7 2.3 -7.8 35.0 25.1 16.2 5 0.96 0.88 0.71 1477 0.08 -5.5 1.2 4.1 37.6 12.1 10.6 15 1.72 1.43 0.97 1626 0.10 -3.7 2.0 5.9 29.2 19.0 13.3 10 0- m m cu rb s NY 100 25 5.23 4.45 1.59 2313 0.17 -3.4 2.1 18.4 57.7 22.3 19.3 TABLE 20 Summary of results from TRAP for each analysis in the curb study matrix Speed = 100 Speed = 70 B A D C G NY Curb A SI Figure 32. ASI of C2500 pickup truck by curb type and speed at impact, based on FEA.

Live-Driver Curb Traversal Tests As described in Chapter 4, low-speed curb-traversal tests were conducted using a live driver and a C2500 pickup truck. These tests were primarily used to validate the finite-element model of the roadway and curb. Tests were performed on the Type B curb, Type G curb, and 6-in. vertical curb at impact angles of 10, 15, 25, and 90 degrees. Tables 21 through 23 summarize the results. The sequence of events that occurred in these tests is illus- trated in Figure 35, which is a series of snapshots of the 25-degree test with the B curb. The vehicle impacted the curb at approximately 25 km/h, striking it first with the right front wheel. The front right suspension was compressed by the impact, and the linkage provided by the stabilizer bar caused the left suspension to slightly compress as well. While the front wheels started to rebound, the vehicle began to roll, extending the back right suspension and compressing the left suspension. The front left suspension then started to com- press again, while the right one maintained a steady elonga- tion because it encountered the descending slope of the back- fill while the cabin rolled back. When the right back wheel impacted the curb, the right back suspension experienced a sudden compression. The impact force and rolling moment were transferred to the chassis, thus extending the two rear suspensions. During this phase, the relative rolling of the bed with respect to the pickup truck cabin was apparent. The lat- eral force caused by the impact of the back right wheel against the curb caused the vehicle to yaw towards the back- fill behind the curb. In one of the tests, the high-speed video showed that the left back wheel left the ground right after the rebound. The pitching moment due to the back right wheel impact compressed the front suspension. As described in Chapter 4, live-driver curb traversal tests were also conducted at moderate speeds, with the vehicle approaching the curb in both tracking and nontracking modes. 62 A number of tracking tests were first performed at 35 mph with the B, C, D, and NY curbs. The driver was able to achieve reasonably repeatable impact conditions and did not report any uncontrollable behavior of the vehicle, although at times he reported feeling strong shocks to the steering wheel. The vehi- cle traversed all the curbs without impacting any components other than the tires. Minor damage occurred to the tread of the tire impacting the curb, including plugs torn from the tire. Tables 24 to 27 summarize the results of these tests. Nontracking tests were then conducted at 35 mph on the same curbs. Two nontracking scenarios were used: (1) over- steering and (2) understeering. During these tests, extreme trajectories and roll angles were often recorded in the impacts, and the anti-rollover outrigger was engaged twice, prevent- ing the physical rollover of the vehicle. The vehicle was not damaged during the testing of the NY and C curbs; but, dur- ing the testing of the B and D curbs, damage to the wheels and the steering system was reported. In particular, bending of the rims was noticed each time they came into direct con- tact with the concrete curbs. Tire blow-out (i.e., tire failure by debedding and subsequent sudden air loss) was recorded in four cases. After one test with the D curb, the neutral posi- tion of the steering wheel was 180 degrees off-center, appar- ently a result of damage to one or more parts of the steering system. The B and D curbs also suffered severe damage, including gouges, scrapes, and broken concrete from the impact of the rims and rim flanges. Tables 28 through 31 summarize the results of the non- tracking tests. CURB–GUARDRAIL SIMULATIONS AND TESTS Finite-element simulations and full-scale crash tests were also used to investigate the response of a 3/4-ton pickup truck 0 5 10 15 20 25 30 0.24 0.28 0.20 0.16 0.12 0.08 0.04 D Curb B Curb Impact Speed of 70 km/h Impact Speed of 100 km/h Impact Angle (degrees) Impact Angle (degrees) A Curb G Curb C Curb NY Curb 0 5 10 15 20 25 30 0.04 0.08 0.12 0.16 0.20 0.24 0.28 0.32 Figure 33. ASI of C2500 pickup truck by impact angle and each curb type at impact speeds of 70 km/h and 100 km/h, based on FEA.

63 Impact Angle (degrees) 0 5 10 15 20 25 30 0 1 2 3 Impact Angle (degrees) Impact Angle (degrees) Impact Angle (degrees) 100 km/h70 km/h Impact Angle (degrees) 70 km/h 70 km/h 100 km/h 100 km/h D CurbB CurbA Curb G CurbC Curb NY Curb Impact Angle (degrees) 0 5 10 15 20 25 300 1 2 3 4 5 6 7 8 0 5 10 15 20 25 30 0 1 2 3 4 5 6 7 8 0 5 10 15 20 25 30-20 -10 0 10 20 30 0 5 10 15 20 25 30 -20 -10 0 10 20 30 0 5 10 15 20 25 30 0 0.50 1 1.50 2 2.50 3 3.50 M ax Y aw A ng le (de gre es) M ax P itc h A ng le (de gre es) M ax R ol l A ng le (de gre es) M ax Y aw A ng le (de gre es) M ax P itc h A ng le (de gre es) M ax R ol l A ng le (de gre es) Figure 34. Maximum roll, pitch, and yaw angle displacements of C2500 pickup truck by curb type and speed at impact.

impacting curb–barrier systems in which the barrier was a modified G4(1S) guardrail with wood blockouts. Curb–Guardrail Simulations LS-DYNA was used to analyze various curb–guardrail systems subjected to impact by the modified C2500 pickup truck model under three different impact conditions: • 100 km/h and 25 degrees (i.e., NCHRP Report 350 Test 3-11), • 85 km/h and 25 degrees, and 64 • 70 km/h and 25 degrees (i.e., NCHRP Report 350 Test 2-11). The study included the modified G4(1S) guardrail installed in combination with five curb types: AASHTO Types B, C, D, and G, and the 100-mm New York curb. Parametric Study A parametric study was again performed, this time vary- ing the impact speed, curb type, and offset from the guardrail. The impact angle was 25 degrees in all simulations. Impact angle 10° 15° 25° 90° Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 73 58 39 48 86 59 50 43 102 54 38 76 110 115 97 93 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 59 43 81 50 39 30 84 61 32 52 85 46 95 69 118 99 Maximum vertical acceleration (g): C.G. of truck Bed of truck 0.9 2.5 1.0 2.5 1.1 3.7 2.0 6.2 Impact angle 10° 15° 25° 90° Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 73 103 62 50 67 103 65 59 55 115 75 60 111 106 82 74 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 57 75 85 46 54 64 82 65 47 58 81 51 54 69 115 103 Maximum vertical acceleration (g): C.G. of truck Bed of truck 1.2 3.5 1.4 3.9 1.8 3.9 1.3 4.6 Impact angle 10° 15° 25° 90° Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 102 53 88 60 119 85 90 57 132 87 115 88 131 128 104 113 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 89 62 48 175 75 70 85 244 64 41 125 256 90 87 189 114 Maximum vertical acceleration (g): C.G. of truck Bed of truck 1.1 3.8 1.6 5.4 2.0 7.4 2.5 6.8 TABLE 21 Type B curb low-speed, live-driver test summary TABLE 22 Type G curb low-speed, live-driver test summary TABLE 23 Vertical 6-inch curb low-speed, live-driver test summary

65 Figure 35. Sequential views of low-speed Type B curb impact at 25 degrees. Impact angle 15° 25° Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 137 136 93 133 100 113 91 103 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 111 99 75 96 93 120 64 98 Maximum acceleration at C.G. (g): x y z +1.7 -1.5 +4.0 1.3 -1.4 +2.8 Stability Good Adequate Impact angle 15° 25° Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 80 123 58 65 85 57 54 113 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 108 82 53 52 49 105 73 94 Maximum acceleration at C.G. (g): x y z -1.4 +0.9 -2.0 +1.3 -1.2 +2.9 Stability Excellent Excellent Impact angle 15° 25° Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 130 101 92 105 109 111 95 136 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 95 131 79 104 91 141 83 102 Maximum acceleration at C.G. (g): x y z -1.5 +2.8 +2.1 +1.8 +1.9 +4.3 Stability Excellent Adequate TABLE 24 Type B curb moderate-speed, live-driver test summary TABLE 26 Type D curb moderate-speed, live-driver test summary TABLE 25 Type C curb moderate-speed, live-driver test summary

66 The AASHTO curbs used in this study were the types most commonly used. Although according to the survey discussed in Chapter 3 many states did not use AASHTO curbs, most of them used curbs that were at least similar to one of these four types (i.e., B, C, D, and G) or Type A. The Type A curb was excluded from the curb–guardrail study because the results of the curb traversal study involving this curb were inconclusive. Three curb placement scenarios were investigated. One scenario involved each of the curbs placed behind the face of the barrier with the front of the curb flush with the front of the W-beam where possible. This scenario was consistent with the recommendations of the FHWA memorandum of Feb 28, 1992, and was expected to provide useful information to the states about the performance of these currently advocated curb–barrier combinations (27). Two other curb-placement scenarios were investigated to determine the effects of curbs placed in combination with guardrails where the offset dis- tance from curb to barrier is greater than zero. Since offset curb–barrier combinations are more common along low- to Impact angle 15° 25° Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 126 96 81 96 67 112 79 99 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 101 126 93 64 81 89 57 104 Maximum acceleration at C.G. (g): x y z +1.7 +1.6 +2.0 +1.3 +1.2 -1.9 Stability Excellent Excellent Average values for all tests Over- steering Under- steering Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 131 47 68 67 76 98 98 23 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 87 73 70 86 110 101 30 11 Maximum acceleration at C.G. (g): x y z +2.7 +5.5 ±4.0 +1.8 +4.0 +3.7 Stability Adequate to Poor Adequate Average values for all tests Over- steering Under- steering Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 108 102 77 62 54 100 74 141 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 82 92 82 87 97 92 46 75 Maximum acceleration at C.G. (g): x y z +3.4 +2.2 +2.5 +1.6 +2.0 +2.0 Stability Poor Good Average values for all tests Over- steering Under- steering Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 63 10 54 58 195 77 91 87 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 71 93 40 56 123 86 52 99 Maximum acceleration at C.G. (g): x y z +3.5 +3.5 +1.7 +3.1 +2.3 +1.4 Stability Poor Good Average values for all tests Over- steering Under- steering Maximum compression (mm): Front right wheel Front left wheel Back right wheel Back left wheel 90 84 70 57 47 77 74 131 Maximum extension (mm): Front right wheel Front left wheel Back right wheel Back left wheel 51 105 68 78 83 83 60 61 Maximum acceleration at C.G. (g): x y z +1.5 +2.2 +1.3 +1.5 +1.9 +2.5 Stability Good Excellent TABLE 27 NY curb moderate-speed, live-driver test summary TABLE 28 Type B curb moderate-speed, nontracking test summary TABLE 29 Type C curb moderate-speed, nontracking test summary TABLE 30 Type D curb moderate-speed, nontracking test summary TABLE 31 NY curb moderate-speed, nontracking test summary

moderate-speed roadways (i.e., less than 80 km/h), analyses of such combinations were primarily conducted for NCHRP Test Level 2 conditions (i.e., 70 km/h), although a select number of impacts with certain combinations were investi- gated at higher speeds. The placement of the curbs in those analyses was based on the results of the curb traversal study discussed previously, with consideration given to the clear zone distances that were required for typical roadways. The backfill and the roadway terrain in the computer model simulations had zero slope. For design speeds of 70 to 80 km/h, the Roadside Design Guide states that the clear zone distance should range from 3.5 m for roadways with an average daily traffic (ADT) volume of less than 750 vehicles to 6.5 m for roadways with an ADT of greater than 6,000 vehicles (2). For design speeds of 100 km/h the clear zone distance ranges from 5 to 8.5 m, depending on ADT. Based on the bumper trajectory plots obtained from the curb traversal study, a vehicle impact speed of 70 km/h and angle of 25 degrees will result in the height of the front bumper continuously increas- ing from the time of wheel contact with the curb to a lateral offset distance of approximately 4 m behind the curb. The bumper will be higher than the top of the guardrail until the vehicle reaches a lateral distance of 5 m behind the curb. Since the middle value of the clear zone distance is approxi- mately 5 m, offset distances of 5 m or greater were not inves- tigated since the guardrail would not have been warranted outside the clear zone area. In those cases, offset distances of 2.5 and 4.5 m were investigated under impact conditions con- sistent with NCHRP Report 350 Test 2-11. For the case of the pickup traversing a curb at 100 km/h and 25 degrees, the bumper trajectory plots from the curb tra- versal study indicated that the bumper height continuously increased after wheel impact with the curb until the vehicle reached a lateral distance of approximately 6 m behind the 67 curb. The bumper remained higher than the guardrail for a lateral distance of approximately 8 m in this case, with the maximum height occurring between 4 and 6 m. Computer- simulated impacts with curb–barrier systems at an offset dis- tance of 4 m were investigated under impact conditions con- sistent with NCHRP Report 350 Test 3-11. The performance of certain curb–barrier systems was also investigated at 85 km/h, which represented the upper speed range for intermediate- speed roadways (i.e., 60 to 80 km/h). Table 32 is a matrix of the simulations performed. The backfill area behind the curbs was modeled with rigid elements using a dynamic coefficient of friction of 0.82 between the tires of the vehicle and the ground surface. It should be noted that the interaction between the tires and ground in these analyses may not accurately represent cases where the backfill is composed primarily of soft soil. Data collected from the simulations included sequential snapshots of the impact event; acceleration-time histories; yaw, pitch, and roll time histories; W-beam tensile force- time histories; and TRAP results (i.e., occupant risk). Much more detail on the analyses and results can be found in Plaxico’s dissertation (52). Results At the beginning of each simulation, the vehicle was aligned to impact post 14 of the guardrail system. This point is 2.4 m upstream of a splice connection. The exact impact point can vary when a barrier is offset from a curb, depending on the yaw angle of the vehicle after impact with the curb. It is important to note that vehicle impact into roadside barriers is highly nonlinear, which means that small varia- tions in the system may lead to very different results. Such Offset distance from barrier to curb Curb type 0 m 2.5 m 4 m B C D G Simulation Test 2-11: Impact speed 70 km/h, Angle of 25 degrees NY B C D G Simulation Test 3-11: Impact speed 100 km/h, Angle of 25 degrees NY B Simulation: Impact speed 85 km/h, Angle of 25 degrees C TABLE 32 Simulations of impact tests with a curb and G4(1S) guardrail system

variations may include impact conditions, impact location on the barrier, vehicle suspension properties, soil conditions, barrier connections, and barrier component properties, to name only a few. Because of the nature of these factors, the results of the FEAs should only be viewed as a tool for assess- ing the performance of the system; they only represent a pos- sible outcome for the conditions specified. For example, in many cases the trajectory of the vehicle during interaction with the barrier causes the tires to impact higher than normal against the W-beam rail. With the wheels in this position, the connec- tion of the W-beam to the post becomes a critical factor. If the connection between the W-beam and post does not fail quickly enough during impact, the posts may pull the W-beam down to a point that allows the wheels of the vehicle to ride up the rail and launch the vehicle, as was the case involving the sim- ulation of the modified C2500R impacting an AASHTO C curb at 100 km/h and 25 degrees with the guardrail posi- tioned at 0-m offset from the curb, as shown in Figure 36. A similar event also occurred in a recent crash test per- formed at the Midwest Roadside Safety Facility in Lincoln, Nebraska, which was documented in a test report by Polivka et al. (29). That test involved a modified G4(1S) guardrail with a 102-mm curb placed underneath the rail behind the face of the W-beam under impact conditions corresponding to NCHRP Report 350 Test 3-11. A section of the guardrail in the impact region incorporated two layers of W-beam (i.e., nested W-beams) to reduce the potential for rupture. Conse- quently, this resulted in four layers of W-beam at the splice connections, which required a much higher force to pull the head of the bolt through W-beam slots in the connection of the rail to the posts. As a result of the stronger connection, the W-beam rail was pulled down and the vehicle launched into the air. Although the vehicle experienced extreme tra- jectory during the impact, the vehicle remained upright and came down on the front side of the guardrail and satisfied all requirements of NCHRP Report 350. The repeatability of such an event is questionable due to the instability of the vehicle during impact with the system; slight changes in either the system or impact conditions may lead to drastically different results. 68 Vehicle Kinematics. Sequential snapshots of the impact event provided a qualitative means of evaluating the general behavior of vehicle interaction with the guardrail as well as the important safety issues regarding vehicle kinematics, such as barrier override, barrier underride, vehicle overturn, and vehicle redirection. Table 33 summarizes the results of these evaluations. The conclusions reached are given in the following paragraphs. For impacts at 70 km/h and 25 degrees, all five curb types were analyzed. The following results were notable: • In cases involving the barrier positioned at 0-m offset from the curb, it appeared that the vehicle remained very stable throughout the impact event and barrier damage appeared to be minimal, regardless of curb type. Although the scenario with the 150-mm AASHTO type D curb resulted in the bumper getting above the rail dur- ing redirection, the potential for override of the barrier appeared minimal. • In cases involving the barrier positioned at 2.5-m offset from curb types B, C, D, and G, the sequential views of the impact events suggested that the vehicle would experience moderate roll angle during impact and a rel- atively high yaw rate, the front of vehicle redirecting out of the system before the rear of the vehicle contacted the rail. Also, while for cases involving 150-mm curb types the bumper of the vehicle climbed above the rail, there was little possibility of override in these cases. The impact scenario involving the 100-mm New York curb resulted in very stable redirection, although the yaw rate appeared somewhat high in this case as well. • In cases involving the barrier positioned at 4.0-m offset from the curbs, the vehicle remained very stable through- out the impact event and barrier damage appeared to be minimal, regardless of the type of curb used in conjunc- tion with the guardrail. However, the vehicle appeared to experience a high yaw rate during redirection, which could increase risk of occupant injury. For impacts at 85 km/h and 25 degrees, only two curb types, the type B and C curbs, were used in these curb–barrier sce- narios. These cases were analyzed in order to assess the per- formance of the curb–barrier systems at speeds correspond- ing to the upper bound of the moderate-speed range (i.e., 60 to 80 km/h) and the lower bound of the high-speed range (i.e., >80 km/h). The following results were observed: • In the cases involving the barrier positioned at 0.0-m off- set from the curbs, the sequential views of the impact sug- gested that the vehicle would remain relatively stable dur- ing impact. There was a slight pitch of the vehicle when the rear wheels contacted the 150-mm Type B curb. • In the cases with the barrier positioned at 2.5-m offset from the curb, the analyses terminated prematurely due to numerical problems in the calculations that were related to contact between the W-beam rail and truck Figure 36. FEA simulation of C2500 pickup impacting guardrail with C curb under rail.

TABLE 33 Curb–guardrail FEA for vehicle override, underride, rollover, and redirection Offset distance Impact speed Curb type Over- ride Under- ride Roll- over Redirection comments B - - - Stable redirection C - - - Stable redirection D - - - Slight bumper trajectory, stable redirection G Analysis Not Conducted 70 km/h NY - - - Stable redirection B - - - Slight pitch 85 km/h C - - - Stable redirection B - - Possible Excessive pitch C Likely - Likely Excessive trajectory D - - Possible Excessive pitch G - - Possible Excessive pitch 0.0 m 100 km/h NY - - - Moderate pitch, stable redirection B - - - Moderate roll angle, high yaw rate, bumper above rail C - - - Moderate roll angle, high yaw rate, slight bumper trajectory D - - - Moderate roll angle, high yaw rate, bumper above rail, tie rod breaks G - - - Moderate roll angle, high yaw rate, bumper above rail 70 km/h NY - - - Stable redirection, high yaw rate Likely - - B Analysis terminated prematurely as bumper started over rail. Excessive roll angle, bumper above rail Likely - - 2.5 m 85 km/h C Analysis terminated prematurely as bumper started over rail. Excessive roll angle, bumper above rail 100 km/h G Likely - Likely Bumper over rail, truck rollover B - - - Analysis terminated during redirection C - - - Stable redirection D - - - Stable redirection G - - - 70 km/h NY Analysis Not Conducted B - - - Stable redirection, high yaw rate 85 km/h C - - - Stable redirection, high yaw rate B Likely - - Override C Likely - - Override D Analysis Not Conducted G Likely - - Override Possible - - 4.0 m 100 km/h NY Analysis terminated prematurely during redirection Excessive trajectory

fender. The analyses did continue long enough, how- ever, to conclude that there was a potential for excessive roll of the vehicle during impact and that the bumper was likely to get over the W-beam rail. Furthermore, the momentum of the truck combined with the excessive trajectory of the bumper was sufficient to cause barrier override. • In the cases involving the barrier positioned at 4.0-m offset from the curb, the sequential views of the impact events suggested that the vehicle would remain stable but was likely to experience a high yaw rate during redirection. For impacts at 100 km/h and 25 degrees, all five curb types were analyzed with the barrier offset 0.0 m and 4.0 m from the curb. Only the Type G curb was analyzed for a barrier off- set 2.5 m from the curb. The following results were observed: • The sequential views of the simulated impact events involving the barrier positioned at 0.0-m offset from the curbs indicated that rollover of the vehicle was possible for each curb–barrier scenario involving the types B, C, D, and G curbs due to excessive pitch of the vehicle dur- ing redirection. Although the vehicle did not roll over in the simulations, the amount of damage to the front impact-side wheel during impact and the position of the front wheels during redirection became a critical factor regarding vehicle stability when the pitch angle of the vehicle was excessive during redirection. In the simula- tions, the wheels remained undamaged and in straight alignment during redirection. There was one case of bar- rier override involving the Type C curb. In this analysis, a wheel snag against a guardrail blockout early in the impact event caused the tie rod to break. The front wheel on the impact side of the vehicle then rotated 90 degrees toward the guardrail. The W-beam rail was pushed down and the vehicle launched over the guardrail. • The impact scenario involving the 100-mm New York curb at a 0.0-m offset from the barrier resulted in mini- mal trajectory of the vehicle with only moderate pitch and a relatively stable redirection. • In the case involving the barrier positioned at 2.5-m offset from the Type G curb, the trajectory of the truck was excessive during impact and, although the trajec- tory of the front bumper and the momentum of the vehi- cle appeared sufficient to cause the vehicle to override the barrier, the guardrail redirected the vehicle away from the system. The vehicle then proceeded to roll over onto its side. • In the cases involving the barrier positioned at 4.0-m offset from the curb, the sequential views of the impact events suggested that barrier override was likely regard- less of curb type. Note: the analysis involving the 100-mm New York curb resulted in premature termination due to numerical problems in the calculations that were related to contact between the front tire and the W-beam, but at 70 the time the analysis was stopped the trajectory and roll angle of the truck was excessive enough to suspect bar- rier override, the likelihood of rollover, or both. Vehicle Angular Displacement. The roll, pitch, and yaw angle displacement-time history data were collected at the center of gravity of the vehicle during the impact event. Table 34 summarizes the vehicle’s angular position at the time of impact with the guardrail, the maximum roll and pitch angle of the vehicle during the impact event, and the yaw angle of the vehicle as it exited the guardrail. The following observations are based on these data: • When the barrier was offset 2.5 m from the curb and the truck impacted the system at 70 or 85 km/h, the initial roll and pitch angle of the vehicle at the time of impact with the guardrail were typically both positive (i.e., away from the guardrail) with the exception of the NY curb. This resulted in the front bumper on the impact side of the vehicle being higher than normal at the time of impact and, according to a qualitative analysis of the sequential views of the impact, the bumper was above the rail during impact for each of these cases. The max- imum roll angle of the vehicle during impact was rela- tively higher in those cases as well. • In the cases involving the barrier offset a distance of 4.0 m from the curb and impact speeds of 70 and 85 km/h, the opposite was typically true, with both the initial roll and pitch angle of the vehicle being negative at the time of impact with the guardrail. In those cases the position of the front bumper on the impact side was relatively lower and, according to the sequential views, the bumper stayed below the top of the rail throughout the impact event. For the scenarios involving impact speeds of 100 km/h, the initial roll angle was typically either zero or positive, while the initial pitch angle was typically negative. In those cases the trajectory and momentum of the vehicle dominated and the primary result was vehi- cle override. • In all cases involving the barrier offset at distances of 2.5 m or 4.0 m from the curb, the curb caused the wheels of the truck to steer toward the guardrail while the vehi- cle traversed the curb, resulting in the vehicle impacting the guardrail at a steeper than normal angle. Conse- quently, for any given curb–barrier case, the impact angle became steeper as the offset distance increased. A steeper impact angle may increase the severity of the impact by increasing the potential for failure of the barrier and by increasing occupant risk factors. Tensile Force in the Guardrail. The maximum values of tensile force in the W-beam cross-section at two critical loca- tions (i.e., in the impact region of the guardrail and at the upstream anchor) as computed in the FEAs are summarized in Table 35. The cases involving the modified C2500R pickup

model impacting the guardrail at 100 km/h and 25 degrees with an offset distance of 0.0 m from curb to barrier are com- pared to the results of the modified C2500R pickup model impacting the guardrail under the same impact conditions without a curb present. In cases in which the rail forces were significantly higher when the curb was present than when it was not, there may be a potential for rupture. For the simu- lation of the guardrail without a curb present under NCHRP Report 350 Test 3-11 conditions, the maximum force in the guardrail occurred in the impact region and was 209 kN and the maximum anchor force was approximately 179 kN. The following conclusions were reached: • The results from the analyses of vehicle impact with the guardrail under Test 2-11 conditions involving each of the different curb types indicated that rupture of the guardrail was not likely to occur regardless of the offset location of the barrier with respect to the curb. 71 • The results from the analyses of vehicle impact at 85 km/h at 25 degrees indicated that rupture of the guardrail was not likely to occur for offset distances of 0 m or 4 m. When the guardrail was placed 2.5 m behind the curb, the tension in the rail reaches magnitudes that may be critical; however, there was also bumper override in those cases. • The analyses of vehicle impact with the guardrail under Test 3-11 conditions involving each of the different curb types located at 0-m offset (i.e., under the W-beam rail) resulted in significantly higher forces in the rail and anchor then when the curb was not present. In all cases, however, there appeared to be potential for excessive anchor movement and rail rupture during impact. The maximum rail forces under Test 3-11 conditions for curb–barrier offset distances greater than 0.0 m are not shown in the table because the predominate outcome in all those cases was barrier override. Impact angle with guardrail (degrees) Max. angular displacement in impact (degrees) Offset distance Impact speed Curb type Roll Pitch Yaw Roll Pitch Yaw B 0.0 0.0 -25.0 -1.9 -6.4 21.0 C 0.0 0.0 -25.0 -7.0 -3.7 21.0 D 0.0 0.0 -25.0 2.2 3.5 20.2 G Analysis not conducted 70 km/h NY 0.0 0.0 -25.0 -4.3 -2.1 21.3 B 0.0 0.0 -25.0 5.4 -7.6 19.385 km/h C 0.0 0.0 -25.0 8.2 -3.3 18.5 B 0.0 0.0 -25.0 -18 -14.2 22.4 C 0.0 0.0 -25.0 31.3 6.0 29.5 D 0.0 0.0 -25.0 -12.5 -14.3 24.2 G 0.0 0.0 -25.0 -11.4 -21.6 23.0 0.0 m 100 km/h NY 0.0 0.0 -25.0 -10.9 -9.1 23.5 B 0.27 0.44 -25.8 -11.9 -3.2 13.7 C Data not recorded due to input error D 0.89 1.13 -26.8 -11.4 -5.2 18.9 G 3.48 0.16 -26.2 -14.1 -6.3 19.9 70 km/h NY 2.87 -0.17 -26.0 -8.4 -5.2 15.8 B 1.22 1.33 -25.7 - - - 2.5 m 85 km/h C 2.92 0.55 -26.3 - - - B -1.95 -1.14 -28.8 5.1 -2.8 NA C -3.39 -2.48 -28.0 -7.6 -2.7 17.7 D -1.80 -1.55 -29.7 5.6 -2.9 19.2 G 0.49 -0.85 -26.8 4.4 -3.4 14.6 70 km/h NY Analysis not conducted B -1.63 -0.81 -27.8 -10.8 -2.0 18.9 85 km/h C -0.82 -1.78 -28.1 -6.3 -3.2 17.0 B 0.0 -0.49 -28.7 -19.6 -6.2 NA C -0.06 -1.42 -27.6 -6.7 -3.5 NA 4.0 m 100 km/h G 2.21 -0.93 -27.5 -45.1 3.5 NA NY 1.84 -0.95 -27.5 -15.2 -3.1 NA TABLE 34 Angular displacement-time history data collected at the center of gravity of the vehicle in the curb–guardrail FEAs

• For cases involving the guardrail positioned at 0.0-m off- set from the curb, the maximum tension in the W-beam rail ranged from 107% to 111% and the maximum force at the upstream anchor was as high as 117% of the val- ues computed in the analysis of the guardrail without a curb present. TRAP Results. Table 36 summarizes the TRAP results, including the OIV, occupant ridedown acceleration (ORA), and maximum 50-m/s moving average acceleration for each curb–guardrail scenario. The OIV in all cases was below the maximum limit of 12 m/s, as required in NCHRP Report 350. For the curb–barrier scenarios in which the barrier was off- set at 2.5 m or 4.0 m from the curb, the data analysis began at first tire contact with the curb. In some of these cases, occupant impact occurred prior to vehicle impact with the barrier (e.g., Type D curb, 70 km/h impact speed, 2.5-m off- set), which resulted in very low values of OIV. The longitudinal ORA values were below the maximum limit of 20 Gs required in NCHRP Report 350 for the cases 72 of 0.0-m offset distance from curb to barrier at all three impact speeds. Seven of the cases for which the offset dis- tance was greater than zero resulted in longitudinal ORA val- ues exceeding 20 Gs: • 150-mm B curb, impact speed of 85 km/h and offset dis- tance of 4.0 m; • 150-mm B curb, impact speed of 100 km/h and offset distance of 4.0 m; • 100-mm C curb, impact speed of 85 km/h and offset dis- tance of 2.5 m; • 100-mm C curb, impact speed of 100 km/h and offset distance of 4.0 m; • 100-mm G curb, impact speed of 70 km/h and offset dis- tance of 2.5 m; • 100-mm G curb, impact speed of 70 km/h and offset dis- tance of 4.0 m; and • 100-mm G curb, impact speed of 100 km/h and offset distance of 4.0 m. Maximum tensile force in W-beam rail Impact region Upstream anchor Downstream location Offset distance Impact speed Curb type (kN) Force/ 209 (kN) Force/ 179 (kN) Force/ 147 B 127 0.61 - - 71.2 0.48 C 127 0.61 124 0.69 87.8 0.60 D 128 0.61 127 0.71 82.9 0.56 G Analysis not conducted 70 km/h NY 135 0.65 131 0.73 76.0 0.52 B 165 0.79 141 0.79 117 0.8085 km/h C 170 0.81 142 0.79 122 0.83 B 232 1.11 - - 182 1.24 C 226 1.08 202 1.13 175 1.19 D 243 1.16 210 1.17 183 1.24 G 223 1.07 - - 174 1.18 0.0 m 100 km/h NY 231 1.11 198 1.11 178 1.21 B 95.0 0.45 88.7 0.50 68.6 0.47 C Data not recorded due to input error D 128 0.61 120 0.67 82.1 0.56 G 123 0.59 118 0.66 77.8 0.53 70 km/h NY 132 0.63 119 0.66 77.7 0.53 B 185 0.89 - - 91.0 0.62 2.5 m 85 km/h C 205 0.98 177 0.99 102 0.69 B 101 0.48 89.4 0.50 66.1 0.45 C 114 0.55 113 0.63 76.5 0.52 D 97.5 0.47 - - 65.1 0.44 G 130 0.62 116 0.65 78.8 0.54 70 km/h NY Analysis not conducted B 171 0.82 143 0.80 103 0.70 4.0 m 85 km/h C 171 0.82 148 0.83 120 0.82 TABLE 35 Maximum tensile force values in the W-beam rail within the impact region and at the upstream anchor, based on FEA

Summary The results of the pickup truck model impacting the curb– barrier combination at 0-m offset distance (i.e., curbs under the face of the barrier) at speeds of 70 km/h and 85 km/h indi- cate that the vehicle would remain stable throughout the impact event and that barrier damage would be minimal regardless of the type of curb used. The bumper of the pickup was above the rail during redirection in one case involving the 150-mm AASHTO Type D curb, but the potential for override of the barrier was considered minimal. At the higher impact speed of 100 km/h the analyses pro- vided mixed conclusions. In one case involving the 100-mm high Type C curb, the vehicle vaulted over the guardrail, whereas vaulting was not a serious issue in the other cases. The difference in this particular case was attributed to a wheel snag against a blockout early in the impact event; this affected the way the vehicle interacted with the barrier throughout the remainder of the event. Wheel snag is com- mon in impacts with strong-post W-beam guardrails, and similar results are possible for cases involving any of the 73 curb types. It was also concluded that vehicle stability may be an issue during redirection due to the high pitch angles of the vehicle when exiting the system. Furthermore, the tensile forces in the W-beam were high during impact, indicating potential for rail rupture at the splice connections, especially for cases involving the 150-mm curbs. The most promising combination involved the 100-mm New York curb, which resulted in safe redirection of the vehicle, although the ten- sile forces in the rail were somewhat high. The results of the FEAs regarding higher-speed impact indi- cated that the roll angle and pitch angle of the vehicle after tra- versing curbs had a significant influence on the kinematics of the vehicle during impact with the guardrail for cases involv- ing offset distances of 2.5 m and 4.0 m. The potential for over- ride was increased when the roll angle of the vehicle was pos- itive (i.e., roll away from the barrier) at the time of impact with the guardrail. When the roll angle of the vehicle was negative (i.e., roll toward the barrier) at the time of impact with the guardrail, rollover became a likely outcome. At impact speeds of 70 km/h into curb–guardrail systems at offset distances of 2.5 m and 4.0 m, there was very little Impact conditions OIV ORA Max. 50-m/s moving average Curb type Speed (km/h) Offset distance (m) x-dir (m/s) y-dir (m/s) x-dir (Gs) (Gs) (Gs) (Gs) (Gs) y-dir x-dir y-dir z-dir 0.0 4.1 -3.6 -6.0 4.7 -4.6 3.3 2.0 2 5 3 5 -2 5 -15 1 19 4 4 6 -10 0 7 470 4 0 2 0 -4 5 13 6 -19 2 -6 3 8 3 -6 7 0 0 4 2 -4 1 8 1 10 6 -4 2 5 7 4 2 2 5 - - - - - - -85 4 0 0 1 -2 6 31 1 29 0 -14 7 10 1 -9 0 0 0 5 5 -5 0 -11 0 14 9 -5 4 7 6 3 3 B 100 4 0 3 6 0 3 -40 0 -49 9 -13 1 9 6 -14 6 0 0 4 3 -4 1 -6 6 6 7 -4 6 3 7 -2 0 2 5 -0 1 1 6 -12 7 17 3 -5 6 5 8 -7 770 4 0 0 3 -1 6 13 3 14 4 -3 9 7 2 5 1 15 0- m m cu rb s D 100 0.0 5.9 -4.8 -14.0 15.9 -5.4 7.1 3.5 0 0 4 2 -4 2 -6 3 7 5 -4 0 3 8 -1 7 2.5 - - - - - - - 70 4.0 1.6 1.4 14.4 13.8 6.9 6.3 6.8 0.0 4.1 -4.3 -12.9 12.6 -4.1 5.5 2.3 2.5 6.1 -3.6 -25.2 -22.0 -9.2 8.5 -12.5 85 4.0 0.7 -1.7 -20.0 16.9 -6.9 5.8 6.7 0.0 5.7 -5.0 8.7 7.4 -5.3 6.0 -3.9 C 100 4.0 5.0 -3.8 -40.0 -49.9 -6.5 5.8 -4.2 0.0 - - - - - - - 2.5 6.0 -2.4 -26.6 17.2 -6.6 5.2 -8.2 70 4.0 1.1 -2.6 21.2 -16.8 -8.5 5.6 6.9 0.0 4.8 -5.3 -11.6 14.8 -5.0 7.0 2.5 G 100 4.0 6.3 -4.9 26.2 -29.2 13.4 -9.6 -11.5 0.0 4.7 -4.2 -5.1 5.7 -4.7 4.1 1.5 2.5 5.8 -4.5 -11.0 10.9 -4.4 6.4 -5.1 70 4.0 - - - - - - - 0.0 5.0 -5.2 -8.2 13.1 -5.0 5.7 2.4 10 0- m m cu rb s NY 100 4.0 5.3 -5.6 -17.0 21.1 -10.4 9.3 6.7 TABLE 36 Occupant risk factors computed using TRAP and the results from the FEAs of the curb–barrier impact study

probability of barrier override; but ORAs during redirection were relatively high. In one case involving the 100-mm Type G curb, the longitudinal ORAs exceeded the maximum value of 20 Gs allowed in NCHRP Report 350. At the intermediate speed of 85 km/h the results from the finite element simula- tions indicated the potential for a pickup truck to override a standard strong-post W-beam guardrail located at 2.5-m off- set distance from both 150-mm and 100-mm curbs. At an off- set distance of 4 m from curb to barrier, the guardrail redi- rected the vehicle at an impact speed of 85 km/h. The ORAs of the vehicle during redirection were considered high, and the Type B curb resulted in excessive ORAs (i.e., greater than 20 Gs). Table 37 provides a summary of the results of the curb– barrier impact study regarding success or failure of the sys- tem in each case, based on the information obtained from the analyses. Analyses were not conducted for all combinations 74 of impact speed, curb type, and offset distance because of limited funds. Full-Scale Crash Tests of Curb–Guardrail Combinations As discussed in Chapter 4, full-scale crash tests were con- ducted of selected curb–guardrail combination scenarios to complement the FEA results. A series of full-scale crash tests, each conforming to the recommendations in NCHRP Report 350 Test 11 for longitudinal barriers, was performed to vali- date the design chart described in Chapter 6. The barriers for all tests were the AASHTO Standard G4(1S) or SGR04a guardrails, modified by the use of recycled plastic blockouts instead of wood blockouts. The impacting vehicle for each test was a 2000P vehicle (i.e., 3/4-ton pickup truck). Testing Offset distance from barrier to curb Impact speed Curb type 0 m 2.5 m 4 m B ✓ ✗ - high long. ORA - high lateral ORA ✓ - high lateral ORA C ✓ ✓ ORA ✓ D ✓ ✓ - high lateral ORA ✓ G N/A ✗ -excess lateral ORA - high lateral ORA ✗ -excess lateral ORA - high lateral ORA 70 km/h NY ✓ ✓ N/A (assumed ✓ ) B ✓ ✗ - override ✗ -excess lateral ORA - high lateral ORA 85 km/h C ✓ ✗ -excess lateral ORA - override - high lateral ORA ✓ - high long. ORA - high lateral ORA B ✓ - high pitch angle - high rail forces N/A ✗ - override -excess long. ORA - high lateral ORA - high roll angle C ✗ - override - rollover -excess lateral ORA - high trans. ORA N/A (assumed ✗ ) ✗ -excess lateral ORA - override - high lateral ORA - high roll angle D ✓ - high pitch angle - high rail forces N/A N/A G ✓ - high pitch angle - high rail forces ✗ - rollover - override -excess lateral ORA - high lateral ORA ✗ - override -excess lateral ORA - high lateral ORA - high roll angle 100 km/h NY ✓ - high rail forces N/A ✓ - high trajectory - high roll angle - high long. ORA - high lateral ORA long. = longitudinal. N/A = not analyzed. trans. = [PI to supply] TABLE 37 Summary of curb–barrier impact study regarding success () or failure () of the system based on the results of the FEAs

examined two types of curbs, three nominal impact speeds and three curb–barrier offset distances. Inadvertently a fourth variable was introduced in the test matrix, barrier height rel- ative to the curb approach. The test matrix was subsequently expanded to compensate for this added variable. The test matrix is shown in Table 38. Test Results Figures 37 through 43 summarize the full-scale crash tests. Each figure shows the theoretical OIVs and ridedown accel- erations in the longitudinal and lateral directions, theoretical head impact velocity (THIV), post-impact head deceleration (PHD), ASI, and maximum roll, pitch, and yaw angles. Test 52-2556-001, conducted by E-TECH Testing Ser- vices, Inc., is summarized in Figure 37. In this test, vehicle contact with the test article occurred 2.5 m upstream of the connection splice at the 15th post in the installation. The bumper was forced back crushing the front right fender and wheel well. The entire front right corner of the vehicle came to bear against the W-beam guardrail. The blockouts sup- ported the W-beam and began loading the posts laterally. The W-beam flattened out forming a ribbon that engaged the vehicle. As the W-beam deflected laterally it developed ten- sion that forced the vehicle to yaw counterclockwise. Maxi- mum dynamic deformation of the guardrail was 0.5 m, and the permanent deformation was 0.4 m. The tire and rim forced posts 15 and 16 to deform. The vehicle engaged two W-beam rails before losing contact with the installation. The exit trajectory of the vehicle center of gravity was 14 degrees relative to the installation centerline when the vehicle lost con- tact with the barrier. The exit velocity of the vehicle was 41.3 km/h. The emergency braking system was applied after loss of contact and the vehicle skidded to a stop 34 m down- stream and 11 m left of its position at impact. The furthest piece of debris, a 6.4-kg blockout, ended up 4 m downstream and 7 m to the rear of its position at impact; the pickup sustained major dents in the bumper, right front fender, and passenger door, and the grill and right front headlight were broken. There was no windshield contact or damage and negligible defor- mation of the vehicle interior. E-TECH Test 52-2556-002 is summarized in Figure 38. In this test, vehicle contact with the test article occurred 2.6 m 75 upstream of the connection splice at the 14th post in the installation. The vehicle traversed the curb, forcing the tires to lose contact with the ground. The right corner of the bumper came into contact with the top ridge of the W-beam guardrail. The blockouts supported the W-beam and began loading the posts laterally. The front end of the vehicle rose up over the guardrail and the guardrail flattened and came to bear against the right front wheel. All four wheels became airborne, and, as the W-beam deflected laterally, it developed tension that forced the vehicle to yaw counterclockwise and roll. The vehicle vaulted over the guardrail, rolled over, came back down on the downstream section of guardrail, and then righted itself on all four wheels. Maximum dynamic defor- mation of the guardrail was 0.6 m, and the permanent defor- mation was 0.4 m. The tire and rim forced posts 15 and 16 to deform in the initial impact, and the vehicle engaged two W-beam rails before losing contact. The vehicle subsequently damaged six downstream posts and one section of rail. The vehicle rolled to a stop 21 m downstream and approximately 6 m behind its position at impact. The furthest piece of debris, a 6.4-kg blockout, ended up 0.5 m downstream and 1.5 m to the rear of its position at impact. The vehicle sustained major dents in the bumper, right front fender, roof, hood, and pas- senger door, and the windshields, mirror, grill and right front headlight were broken. There was a maximum 330 mm defor- mation of the vehicle interior at the right windshield pillar. E-TECH Test 52-2556-003 is summarized in Figure 39. In this test, vehicle contact with the test article occurred 2.2 m upstream of the connection splice, just upstream of the 14th post in the installation. The vehicle traversed the curb, forc- ing the tires to lose contact with the ground. The bottom of the right corner of the bumper came into contact with the top edge of the W-beam guardrail. The blockouts supported the W-beam and began loading the posts laterally. The front overhang of the vehicle rose up over the guardrail and the guardrail came to bear against the right front wheel. All four wheels became airborne, and the front end of the vehicle passed over the guardrail. The rear end of the vehicle slid along the top of the downstream section of guardrail, and then the vehicle came to rest with the back tires on the guard- rail and the front wheels on the ground behind the guardrail. Maximum dynamic deformation of the guardrail was 0.4 m, and the permanent deformation was 0.3 m. The right front wheel forced posts 15 and 16 to deform in the initial impact, E-TECH test no. Nominal speed (km/h) Curb type Curb offset (m) Guardrail height (mm) relative to approach 52-2556-001 85 B 0.0 550 52-2556-002 85 B 2.5 550 52-2556-003 80 NY 2.5 550 52-2556-004 80 NY 4.5 550 52-2556-005 80 NY 4.5 650 52-2556-006 70 NY 2.5 650 52-2556-007 85 NY 2.5 650 TABLE 38 Full-scale crash test matrix

76 t = 0.000 sec t = 0.144 sec t = 0.288 sec t = 0.432 sec t = 0.576 sec t = final General Information Test Agency ............................................................... E-TECH Testing Services, Inc. Test Designation ...................................................... NCHRP 350 Test 3-11 (modified) Test No. ..................................................................... 52-2556-001 Date ............................................................................ 6/5/03 Test Article Curb Type ................................................................. AASHTO Type B Barrier Length ....................................................... 53.34 m (overall) Height (mm - relative to approach) ...................... 550 Setback (m - relative to curb) ............................... 0 Material and key elements ...................................... AASHTO SGR04a Guardrail with ............................................................................ SEW02a End Terminal equipped ............................................................................ Re-Block recycled plastic blockouts ............................................................................ of 50% HDPE / 50% PP Foundation Type and Condition ..................................... NCHRP 350 Strong Soil, dry Test Vehicle Type ........................................................................... Production Model Designation ............................................................... 2000P Model ......................................................................... 1998 GMC ............................................................................ 3/4 Ton Pickup Mass (kg) Curb .................................................................. 1975 Test inertial ....................................................... 1993 Impact Conditions Speed (km/h) ............................................................ 85.6 Angle (deg) ................................................................ 25 Impact Severity (kJ) .............................................. 100.6 Exit conditions Speed (km/h) ............................................................ 41.3 Angle (deg - veh. c.g.) .............................................. 14 Occupant Risk Values Impact Velocity (m/s) x-direction ........................................................ 4.9 y-direction ........................................................ -4.7 Ridedown Acceleration (g's) x-direction ......................................................... -8.1 y-direction ......................................................... -6.3 European Committee for Normalization (CEN) Values THIV (km/h) ............................................................. 24.1 PHD (g's) ................................................................... 8.8 ASI ............................................................................ 0.7 Post-Impact Vehicular Behavior (deg - rate gyro) Maximum Roll Angle .............................................. 6.5 Maximum Pitch Angle ............................................ -10.2 Maximum Yaw Angle .............................................. -52.0 Test Article Deflections (m) Dynamic .................................................................... 0.5 Permanent ................................................................. 0.4 Vehicle Damage (Primary Impact) Exterior VDS .................................................................... RFQ-3 CDC ................................................................... 01RFWE2 Interior VCDI ................................................................. AS0000000 Maximum Deformation (mm) ......................... Negligible Type B Curb SGR04a Guardrail Figure 37. Summary of curb–guardrail crash test 52-2556-001, B curb below guardrail.

77 t = 0.000 sec t = 0.144 sec t = 0.288 sec t = 0.432 sec t = 0.576 sec t = final General Information Test Agency ............................................................... E-TECH Testing Services, Inc. Test Designation ...................................................... NCHRP 350 Test 3-11 (modified) Test No. ..................................................................... 52-2556-002 Date ............................................................................ 6/18/03 Test Article Curb Type ................................................................. AASHTO Type B Barrier Length ....................................................... 53.34 m (overall) Height (mm - relative to approach) ...................... 550 Setback (m - relative to curb) ............................... 2.5 Material and key elements ...................................... AASHTO SGR04a Guardrail with ............................................................................ SEW02a End Terminal equipped ............................................................................ Re-Block recycled plastic blockouts ............................................................................ of 50% HDPE / 50% PP Foundation Type and Condition ..................................... NCHRP 350 Strong Soil, dry Test Vehicle Type ........................................................................... Production Model Designation ............................................................... 2000P Model ......................................................................... 1994 Chevrolet ............................................................................ 3/4 Ton Pickup Mass (kg) Curb .................................................................. 1919 Test inertial ....................................................... 2002 Impact Conditions Speed (km/h) ............................................................ 86.6 Angle (deg) ................................................................ 25 Impact Severity (kJ) .............................................. 103.5 Exit conditions Speed (km/h) ............................................................ N/A Angle (deg - veh. c.g.) .............................................. N/A Occupant Risk Values Impact Velocity (m/s) x-direction ........................................................ 5.5 y-direction ........................................................ -3.2 Ridedown Acceleration (g's) x-direction ......................................................... -10.8 y-direction ......................................................... 11.4 European Committee for Normalization (CEN) Values THIV (km/h) ............................................................. 22.4 PHD (g's) ................................................................... 14.7 ASI ............................................................................ 0.8 Post-Impact Vehicular Behavior (deg - rate gyro) Maximum Roll Angle .............................................. 472.1 Maximum Pitch Angle ............................................ 26.9 Maximum Yaw Angle .............................................. 20.3 Test Article Deflections (m) Dynamic .................................................................... 0.6 Permanent ................................................................. 0.4 Vehicle Damage (Primary Impact) Exterior VDS .................................................................... R&T-5/RFQ-4 CDC ................................................................... 01RFE03 Interior VCDI ................................................................. RF0000010 Maximum Deformation (mm) ......................... 330 Type B SGR04a Guardrail Figure 38. Summary of curb–guardrail crash test 52-2556-002, B curb offset 2.5 m from guardrail.

78 t = 0.000 sec t = 0.096 sec t = 0.192 sec t = 0.288sec t = 0.384 sec t = final General Information Test Agency ............................................................... E-TECH Testing Services, Inc. Test Designation ...................................................... NCHRP 350 Test 3-11 (modified) Test No. ..................................................................... 52-2556-003 Date ............................................................................ 7/21/2003 Test Article Curb Type ................................................................. New York T100 Barrier Length ....................................................... 53.34 m (overall) Height (mm - relative to approach) ...................... 550 Setback (m - relative to curb) ............................... 2.5 Material and key elements ...................................... AASHTO SGR04a Guardrail with ............................................................................ SEW02a End Terminal equipped ............................................................................ Re-Block recycled plastic blockouts ............................................................................ of 50% HDPE / 50% PP Foundation Type and Condition ..................................... NCHRP 350 Strong Soil, dry Test Vehicle Type ........................................................................... Production Model Designation ............................................................... 2000P Model ......................................................................... 1994 GMC ............................................................................ 3/4 Ton Pickup Mass (kg) Curb .................................................................. 1940 Test inertial ....................................................... 1994 Impact Conditions Speed (km/h) ............................................................ 80.0 Angle (deg) ................................................................ 25 Impact Severity (kJ) .............................................. 87.8 Exit conditions Speed (km/h) ............................................................ N/A Angle (deg - veh. c.g.) .............................................. N/A Occupant Risk Values Impact Velocity (m/s) x-direction ........................................................ 5.6 y-direction ........................................................ -3.0 Ridedown Acceleration (g's) x-direction ......................................................... -6.1 y-direction ......................................................... -4.3 European Committee for Normalization (CEN) Values THIV (km/h) ............................................................. 22.0 PHD (g's) ................................................................... 6.6 ASI ............................................................................ 0.6 Post-Impact Vehicular Behavior (deg - rate gyro) Maximum Roll Angle .............................................. -41.9 Maximum Pitch Angle ............................................ -32.5 Maximum Yaw Angle .............................................. 95.5 Test Article Deflections (m) Dynamic .................................................................... 0.4 Permanent ................................................................. 0.3 Vehicle Damage (Primary Impact) Exterior VDS .................................................................... RFQ-3 CDC ................................................................... 01RFWW1 Interior VCDI ................................................................. AS0000000 Maximum Deformation (mm) ......................... 74 Type T100 Curb SGR04a Guardrail Figure 39. Summary of curb–guardrail crash test at 80 km/h 52-2556-003, NY curb offset 2.5 m from guardrail, guardrail height 550 mm.

and the vehicle engaged two W-beam rails. The vehicle sub- sequently damaged two downstream posts and one section of rail. The vehicle slid to a stop 24 m downstream of its posi- tion at impact, straddling the rail. The vehicle sustained minor dents in the bumper and right front fender, a major dent in the bed on the driver side, major damage to the front right wheel and suspension, and a bent frame. There was no windshield contact or damage, and a maximum 74-mm defor- mation of the vehicle interior at the toe pan area on the pas- senger side. E-TECH Test 52-2556-004 is summarized in Figure 40. In this test, the vehicle contacted the curb and the suspension compressed at first and then extended during the traverse. The body of the vehicle was noticeably elevated, but the tires remained in contact with the ground. The vehicle bumper contacted the guardrail just upstream of the connection splice at Post 15. The right corner of the bottom surface of the bumper came into contact with the top edge of the W-beam guardrail. The blockouts supported the W-beam and began loading the posts laterally. The front overhang of the vehicle rose up over the guardrail and the guardrail came to bear against the right front wheel. All four wheels became air- borne, and the vehicle pitched up and passed over the guard- rail with relatively minor change in direction. The vehicle landed behind the guardrail and remained upright. The vehi- cle slid to a stop 36 m downstream and 3.8 m to the right of its position at impact. Maximum dynamic deformation of the guardrail was 0.4 m, and the permanent deformation was 0.3 m. The pickup sustained minor dents in the bumper and right front fender and major damage to the front right wheel and suspension, and the frame was bent. There was no wind- shield contact or damage, and negligible deformation of the vehicle interior. E-TECH Test 52-2556-005 is summarized in Figure 41. In this test, the vehicle bumper contacted the guardrail 0.6 m upstream of the connection splice at Post 15. The right cor- ner of the bumper came into contact with the W-beam guard- rail. The blockouts supported the W-beam and began loading the posts laterally. The W-beam flattened out, forming a rib- bon that engaged the vehicle. As the W-beam deflected later- ally, it developed tension that forced the vehicle to yaw coun- terclockwise. Maximum dynamic deformation of the guardrail was 0.6 m and the permanent deformation was 0.5 m. Posts 15 through 17 were deformed. The vehicle traversed four W-beam rails before losing contact with the installation. The vehicle exit angle was 12 degrees relative to the installation centerline when it lost contact with the barrier. The exit veloc- ity of the vehicle was 43.3 km/h. The emergency braking sys- tem was applied after loss of contact, and the vehicle skidded to a stop 27 m downstream and 2 m left of its position at impact. The pickup sustained minor dents in the bumper and right front fender and major damage to the front right wheel and suspension, and the frame was bent. There was no wind- shield contact or damage, and negligible deformation of the vehicle interior. 79 E-TECH Test 52-2556-006 is summarized in Figure 42. In this test, the vehicle bumper contacted the guardrail 1.9 m upstream of the connection splice at Post 15. The right cor- ner of the bumper came into contact with the W-beam guard- rail. The blockouts supported the W-beam and began loading the posts laterally. The W-beam flattened out, forming a rib- bon that engaged the vehicle. As the W-beam deflected lat- erally it developed tension that forced the vehicle to yaw counterclockwise. Maximum dynamic deformation of the guardrail was 0.5 m, and the permanent deformation was 0.3 m. In the initial impact, the right front wheel forced posts 15 and 16 to deform and the vehicle engaged two W-beam rails. The vehicle slid to a stop 22 m downstream of its posi- tion at impact and came to rest against the downstream sec- tion of guardrail. The pickup sustained minor dents in the bumper and right front fender, a major dent in the bed on the driver’s side, major damage to the front right wheel and sus- pension, and the frame was bent. There was no windshield contact or damage, and negligible deformation of the vehicle interior. E-TECH Test 52-2556-007 is summarized in Figure 43. In this test, the vehicle bumper contacted the guardrail 2.5 m upstream of the connection splice at Post 15. The right cor- ner of the bumper came into contact with the W-beam guard- rail. The blockouts supported the W-beam and began loading the posts laterally. The W-beam flattened out, forming a rib- bon that engaged the vehicle. As the W-beam deflected later- ally it developed tension that forced the vehicle to yaw coun- terclockwise. Maximum dynamic deformation of the guardrail was 0.7 m, and the permanent deformation was 0.4 m. In the initial impact, the right front wheel forced posts 15 and 16 to deform and the vehicle engaged two W-beam rails. The vehi- cle slid to a stop 30 m downstream of its position at impact and came to rest against the downstream section of guardrail. The pickup sustained minor dents in the bumper and right front fender, a major dent in the bed on the driver’s side, and major damage to the front right wheel and suspension; the frame was bent. There was no windshield contact or damage, and negligible deformation of the vehicle interior. In all the tests, the damage to the guardrail was catego- rized as substantial since one or more replacement posts and W-beam sections would be needed for repair. Most other components of the installations were judged reusable. Summary of Crash Test Results The results of the seven full-scale crash tests were evaluated using the structural adequacy, occupant risk, and vehicle tra- jectory evaluation criteria for longitudinal barrier Test 11 from NCHRP Report 350, as shown in Table 39. Note that the eval- uations of the test results were based on the nominal impact speeds. The relevant evaluation criteria were as follows: • Test article should contain and redirect the vehicle; the vehicle should not penetrate, underride, or override the

80 t = 0.144 sec t = 0.288 sec t = 0.432 sec t = 0.576 sec General Information Test Agency ............................................................... E-TECH Testing Services, Inc. Test Designation ...................................................... NCHRP 350 Test 3-11 (modified) Test No. ..................................................................... 52-2556-004 Date ............................................................................ 8/14/2003 Test Article Curb Type ................................................................. New York T100 Barrier Length ....................................................... 53.34 m (overall) Height (mm - relative to approach) ...................... 550 Setback (m - relative to curb) ............................... 4.5 Material and key elements ...................................... AASHTO SGR04a Guardrail with ............................................................................ SEW02a End Terminal equipped ............................................................................ Re-Block recycled plastic blockouts ............................................................................ of 50% HDPE / 50% PP Foundation Type and Condition ..................................... NCHRP 350 Strong Soil, dry Test Vehicle Type ........................................................................... Production Model Designation ............................................................... 2000P Model ......................................................................... 1989 GMC ............................................................................ 3/4 Ton Pickup Mass (kg) Curb .................................................................. 1947 Test inertial ....................................................... 2014 Impact Conditions Speed (km/h) ............................................................ 81.3 Angle (deg) ................................................................ 23 Impact Severity (kJ) .............................................. 78.4 Exit conditions Speed (km/h) ............................................................ N/A Angle (deg - veh. c.g.) .............................................. N/A Occupant Risk Values Impact Velocity (m/s) x-direction ........................................................ 4.7 y-direction ........................................................ -3.6 Ridedown Acceleration (g's) x-direction ......................................................... -4.1 y-direction ......................................................... -4.9 European Committee for Normalization (CEN) Values THIV (km/h) ............................................................. 20.8 PHD (g's) ................................................................... 5.2 ASI ............................................................................ 0.6 Post-Impact Vehicular Behavior (deg - rate gyro) Maximum Roll Angle .............................................. 31.9 Maximum Pitch Angle ............................................ 11.4 Maximum Yaw Angle .............................................. -12.9 Test Article Deflections (m) Dynamic .................................................................... 0.4 Permanent ................................................................. 0.3 Vehicle Damage (Primary Impact) Exterior VDS .................................................................... RFQ-3 CDC ................................................................... 01RFEW3 Interior VCDI ................................................................. AS0000000 Maximum Deformation (mm) ......................... Negligible New York Type T100 Curb t = final SGR04a Guardrail Figure 40. Summary of curb–guardrail crash test 52-2556-004, NY curb offset 4.5 m from guardrail, guardrail height 550 mm.

81 t = 0.192 sec t = 0.384 sec t = 0.576 sec t = 0.768 sec t = final General Information Test Agency ............................................................... E-TECH Testing Services, Inc. Test Designation ...................................................... NCHRP 350 Test 3-11 (modified) Test No. ..................................................................... 52-2556-005 Date ............................................................................ 9/5/2003 Test Article Curb Type ................................................................. New York T100 Barrier Length ....................................................... 53.34 m (overall) Height (mm - relative to approach) ...................... 550 Setback (m - relative to curb) ............................... 4.5 Material and key elements ...................................... AASHTO SGR04a Guardrail with ............................................................................ SEW02a End Terminal equipped ............................................................................ Re-Block recycled plastic blockouts ............................................................................ of 50% HDPE / 50% PP Foundation Type and Condition ..................................... NCHRP 350 Strong Soil, dry Test Vehicle Type ........................................................................... Production Model Designation ............................................................... 2000P Model ......................................................................... 1994 Chevrolet ............................................................................ C2500 Pickup Mass (kg) Curb .................................................................. 1904 Test inertial ....................................................... 1999 Impact Conditions Speed (km/h) ............................................................ 80.5 Angle (deg) ................................................................ 24 Impact Severity (kJ) .............................................. 82.7 Exit conditions Speed (km/h) ............................................................ 43.3 Angle (deg - veh. c.g.) .............................................. 12 Occupant Risk Values Impact Velocity (m/s) x-direction ........................................................ 4.2 y-direction ........................................................ -3.8 Ridedown Acceleration (g's) x-direction ......................................................... -7.6 y-direction ......................................................... -5.8 European Committee for Normalization (CEN) Values THIV (km/h) ............................................................. 19.4 PHD (g's) ................................................................... 9.1 ASI ............................................................................ 0.5 Post-Impact Vehicular Behavior (deg - rate gyro) Maximum Roll Angle .............................................. -8.9 Maximum Pitch Angle ............................................ -4.9 Maximum Yaw Angle .............................................. -37.2 Test Article Deflections (m) Dynamic .................................................................... 0.6 Permanent ................................................................. 0.5 Vehicle Damage (Primary Impact) Exterior VDS .................................................................... RFQ-3 CDC ................................................................... 01RFEW3 Interior VCDI ................................................................. AS0000000 Maximum Deformation (mm) ......................... Negligible New York Type T100 Curb SGR04a Guardrail t = 0.000 sec Figure 41. Summary of curb–guardrail crash test 52-2556-005, NY curb offset 4.5 m from guardrail, guardrail height 650 mm.

82 Exit conditions Speed (km/h) ............................................................ 36.5 Angle (deg - veh. c.g.) .............................................. 12 Occupant Risk Values Impact Velocity (m/s) x-direction ........................................................ 4.2 y-direction ........................................................ -4.2 Ridedown Acceleration (g's) x-direction ......................................................... -5.3 y-direction ......................................................... -5.0 European Committee for Normalization (CEN) Values THIV (km/h) ............................................................. 19.6 PHD (g's) ................................................................... 7.3 ASI ............................................................................ 0.5 Post-Impact Vehicular Behavior (deg - rate gyro) Maximum Roll Angle .............................................. 6.9 Maximum Pitch Angle ............................................ -7.2 Maximum Yaw Angle .............................................. -39.2 Test Article Deflections (m) Dynamic .................................................................... 0.5 Permanent ................................................................. 0.3 Vehicle Damage (Primary Impact) Exterior VDS .................................................................... RFQ-3 CDC ................................................................... 01RFEW3 Interior VCDI ................................................................. AS0000000 Maximum Deformation (mm) ......................... Negligible t = 0.2400 sec t = 0.4800 sec t = 0.7200sec t = 0.9600 sec t = final General Information Test Agency ............................................................... E-TECH Testing Services, Inc. Test Designation ...................................................... NCHRP 350 Test 2-11 Test No. ..................................................................... 52-2556-006 Date ............................................................................ 10/7/2003 Test Article Curb Type ................................................................. New York T100 Barrier Length ....................................................... 53.34 m (overall) Height (mm - relative to approach) ...................... 650 Setback (m - relative to curb) ............................... 2.5 Material and key elements ...................................... AASHTO SGR04a Guardrail with ............................................................................ SEW02a End Terminal equipped ............................................................................ Re-Block recycled plastic blockouts ............................................................................ of 50% HDPE / 50% PP Foundation Type and Condition ..................................... NCHRP 350 Strong Soil, dry Test Vehicle Type ........................................................................... Production Model Designation ............................................................... 2000P Model ......................................................................... 1990 Chevrolet ............................................................................ C2500 Pickup Mass (kg) Curb .................................................................. 1862 Test inertial ....................................................... 2007 Impact Conditions Speed (km/h) ............................................................ 69.6 Angle (deg) ................................................................ 25 Impact Severity (kJ) .............................................. 67.0 New York Type T100 Curb SGR04a Guardrail t = 0.0000 sec Figure 42. Summary of curb–guardrail crash test 52-2556-006, nominal speed 70 km/h, NY curb offset 2.5 m from guardrail.

83 Exit conditions Speed (km/h) ............................................................ 41.8 Angle (deg - veh. c.g.) .............................................. 15 Occupant Risk Values Impact Velocity (m/s) x-direction ........................................................ 5.0 y-direction ........................................................ -4.3 Ridedown Acceleration (g's) x-direction ......................................................... -10.0 y-direction ......................................................... -17.8 European Committee for Normalization (CEN) Values THIV (km/h) ............................................................. 20.7 PHD (g's) ................................................................... 17.8 ASI ............................................................................ 0.8 Post-Impact Vehicular Behavior (deg - rate gyro) Maximum Roll Angle .............................................. 17.0 Maximum Pitch Angle ............................................ -17.3 Maximum Yaw Angle .............................................. -40.2 Test Article Deflections (m) Dynamic .................................................................... 0.7 Permanent ................................................................. 0.4 Vehicle Damage (Primary Impact) Exterior VDS .................................................................... RFQ-3 CDC ................................................................... 01RFEW3 Interior VCDI ................................................................. AS0000000 Maximum Deformation (mm) ......................... Negligible t = 0.144 sec t = 0.288sec t = 0.432sec t = 0.576 sec t = final General Information Test Agency ............................................................... E-TECH Testing Services, Inc. Test Designation ...................................................... NCHRP 350 Test 3-11 (modified) Test No. ..................................................................... 52-2556-007 Date ............................................................................ 12/4/03 Test Article Curb Type ................................................................. New York T100 Barrier Length ....................................................... 53.34 m (overall) Height (mm - relative to approach) ...................... 650 Setback (m - relative to curb) ............................... 2.5 Material and key elements ...................................... AASHTO SGR04a Guardrail with ............................................................................ SEW02a End Terminal equipped ............................................................................ Re-Block recycled plastic blockouts ............................................................................ of 50% HDPE / 50% PP Foundation Type and Condition ..................................... NCHRP 350 Strong Soil, drained Test Vehicle Type ........................................................................... Production Model Designation ............................................................... 2000P Model ......................................................................... 1994 GMC ............................................................................ C2500 Pickup Mass (kg) Curb .................................................................. 1870 Test inertial ....................................................... 2001 Impact Conditions Speed (km/h) ............................................................ 85.6 Angle (deg) ................................................................ 25 Impact Severity (kJ) .............................................. 101.0 New York Type T100 Curb SGR04a Guardrail t = 0.0000 sec Figure 43. Summary of curb–guardrail crash test 52-2556-007, nominal speed 85 km/h, NY curb offset 2.5 m from guardrail.

installation although controlled lateral deflection of the test article is acceptable. • Detached elements, fragments, or other debris from the test article should not penetrate or show potential for penetrating the occupant compartment or present an undue hazard to other traffic, pedestrians, or personnel in a work zone. Deformations of, or intrusions into, the occupant compartment that could cause serious injuries should not be permitted. • The vehicle should remain upright during and after col- lision, although moderate rolling, pitching, and yawing are acceptable. • After collision, it is preferable that the vehicle’s trajec- tory not intrude into adjacent traffic lanes. • The OIV in the longitudinal direction should not exceed 12 m/s and the ORAs in the longitudinal direction should not exceed 20 Gs. • The exit angle from the test article preferably should be less than 60% of the test impact angle, measured at time of vehicle loss of contact with test device. In the tests shown in Table 39 that passed the NCHRP Report 350 criteria, the vehicles were contained and redi- rected. The vehicles in the failed tests either completely or nearly completely vaulted the installation. In the successful tests, the vehicle’s post-collision trajectory was acceptable and the recommended maximum longitudinal OIV and max- imum ridedown acceleration were not exceeded. The vehicle trajectory in the failed tests was completely or very nearly 84 completely behind the guardrail, which is unacceptable. In Test 52-2556-003, which is noted as “marginal pass,” the vehicle came to rest on the rail with the cab on the backside of the rail. In none of the tests was there any debris larger than a 6.4-kg plastic blockout to present a potential hazard to other traffic, pedestrians, or personnel in a work zone. The vehicle occupant compartment deformation evident after each pass- ing test was also negligible. There was no windshield contact and no intrusion into the occupant compartment for the pass- ing tests, and the test vehicles remained upright during and after the collision with moderate rolling, pitching, and yawing. In tests 52-2556-003 and 004, the occupant risk criteria were satisfied in a similar fashion. In failed test 52-2556-002, however, the vehicle rolled over and a downstream post likely penetrated the windshield; the maximum cab deformation was 330 mm. SUMMARY Several types of analyses were used in developing the guidelines discussed in Chapter 6: review of prior studies, analyses of actual crash and geometric data, FEA simula- tions, and full-scale curb traversal and crash tests. The results of these analyses provided insights into the nature and sever- ity of crashes with curbs and curb–guardrail combinations, the behavior of pickup trucks in such crashes, and the effects of various curb and impact conditions. E-TECH test no. Nominal speed (km/h) Curb type Curb offset (m) Guardrail height (mm) relative to approach NCHRP 350 Test 11 evaluation 52-2556-001 85 B 0.0 550 Pass 52-2556-002 85 B 2.5 550 Fail 52-2556-003 80 NY 2.5 550 Marginal Pass 52-2556-004 80 NY 4.5 550 Fail 52-2556-005 80 NY 4.5 650 Pass 52-2556-006 70 NY 2.5 650 Pass 52-2556-007 85 NY 2.5 650 Pass TABLE 39 Summary of full-scale crash test results

Next: Chapter 6 - Design Guidelines for the Use of Curbs with Guardrails »
Recommended Guidelines for Curb and Curb-Barrier Installations Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Report 537: Recommended Guidelines for Curb and Curb–Barrier Installations presents the findings of a research project to develop guidelines for the use of curbs and curb–guardrail combinations on high-speed roadways. The report includes recommendations concerning the location of curbs with respect to the guardrail for various operating speeds.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!