Design Guidance for Freeway Mainline Ramp Terminals (2012)

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.

21 Passenger cars are currently the primary design vehicle for freeway mainline ramp terminals. However, many freeway mainline ramp terminals serve a considerable amount of truck traffic and, given the difference in performance capabilities between passenger cars and trucks, a primary question to be answered is whether freeway mainline ramp terminal design should consider the performance characteristics of heavy vehicles. For example, several studies (Reilly et al., 1989; Ervin et al., 1986; Firestine et al., 1989; and Gattis et al., 2008) recom- mend longer acceleration/deceleration lane lengths to better accommodate heavy vehicles. No safety study has been conducted to indicate whether trucks have difficulty using freeway ramps designed according to the current criteria. Therefore, a crash analysis was conducted to determine if there are specific truck-related safety prob- lems near freeway mainline ramp terminals. If specific truck- related safety problems were identified near freeway mainline ramp terminals, then an argument could be made that a truck rather than a passenger car should be the design vehicle for an entrance or exit ramp (at least in some instances). On the other hand, if crash data were to indicate that truck crashes are not over-represented near freeway mainline ramp terminals, then continuing to design freeway mainline ramp terminals based upon the performance characteristics of passenger cars may be justified. 4.1 Overview of Crash Analysis An exploratory crash analysis was conducted to investigate the vehicle types involved in crashes near freeway mainline ramp terminals and other related issues. The crash analysis was conducted using the Highway Safety Information System (HSIS) database for the state of Washington. Supplemental interchange characteristic data were also obtained by review- ing interchange diagrams. The Washington database was used because it has a relatively complete set of ramp inventory data, including information on acceleration and deceleration lanes. In addition, information was available in the database to assign the location of a crash occurring near a freeway mainline ramp terminal to a specific area, such as the acceleration lane, decel- eration lane, ramp proper, or freeway mainline adjacent to the acceleration/deceleration lane. Other databases were also considered but lacked the level of detail necessary for such an analysis. The scope of the investigation was limited to interchange areas including the ramp proper, acceleration lanes, decel- eration lanes, and the adjoining mainline roadway segments. Because the initial inventory data contained all ramps for inter- changes occurring on state routes, study sites were screened to exclude ramps connecting to auxiliary lanes (e.g., collector/ distributor roads and parallel frontage roads), ramps with left exits/entrances, and ramps that become additional lanes on the freeway. Ramps containing two-way sections and weaving areas were also excluded. Three years (2002 to 2004) of traffic volume and crash his- tory data were obtained. The geometric and traffic volume variables of interest for each interchange area included: • Annual average daily traffic (AADT), • Annual truck volume percent, • Length, • Interchange configuration, and • Type of ramp (entrance/exit). Criteria were developed to assign crashes to the study loca- tions. Crashes that occurred within the termini locations of the sites that were identified as “non-intersection and non- intersection related” were included in the analysis. Addition- ally, truck crashes were identified if Vehicle 1 or Vehicle 2 was classified as one of the following: • Flatbed truck, • Truck over 10K, • Truck tractor, • Truck tractor and semitrailer, and • Other truck combination. S e c t i o n 4 Analysis of Truck-Related Crashes Near Freeway Mainline Ramp Terminals

22 In the analysis, ramps were grouped into seven configu- rations: • Diamond, • Parclo loop, • Free-flow loop, • Outer connection, • Direct or semi-direct connection, • Button hook, scissor, and slip ramps, and • Others. Figure 4 illustrates the typical ramp configurations included in the seven groupings used in this analysis. Table 5 presents the number of ramps included in the analysis by configuration type and by entrance or exit ramp. Data for 525 entrance ramps and 550 exit ramps were included in the database. The majority of Figure 4. Typical ramp configurations (Bauer and Harwood, 1998). Diamond Parclo loop Free-flow loop Outer connection Direct or semi- direct connection Button hook, scissor, and slip ramps Other Total Entrance ramps 316 24 21 29 34 20 81 525 Exit ramps 329 18 17 47 41 14 84 550 Table 5. Number of ramps by configuration.

23 ramps were of the diamond type. Table 6 shows the overall mileage of the ramps and roadway segments included the analy- sis. The mileage is divided by freeway mainline ramp terminal location (i.e., acceleration/deceleration lane, ramp proper, and freeway mainline adjacent to acceleration/deceleration lane). Figure 5 illustrates the three locations that accidents were assigned near the freeway mainline ramp terminals. 4.2 Analysis Results The crash analysis results focus on comparing crash frequencies/rates for all vehicle types combined (i.e., total crashes) to crash frequencies/rates for heavy vehicles (i.e., trucks). Table 7 shows total crash frequencies and truck-related crash frequencies by ramp configuration and crash location (i.e., acceleration or deceleration lane, freeway mainline adja- cent to acceleration/deceleration lane, or ramp proper). Table 7 shows that 38 percent of the total crashes (4,049) occurred at or near diamond ramps and 35 percent (3,907) occurred near “other” ramps. Similar percentages hold for truck-related crashes; 38 percent of truck-related crashes (407) occurred near diamond ramps, while 35 percent (379) occurred near “other” ramps. These frequencies and percentages could be somewhat misleading because 645 of the 1,075 ramps included in the analysis were diamond ramps and 165 ramps were classi- fied as “other” (see Table 5). The next most common ramp type was outer connections, with 75 included in the database, so it is expected that the greatest number of total and truck-related crashes occurred on the two most common type of ramp con- figurations. Exposure levels are not considered in Table 6. It can also be reasoned that by definition the “other” ramp type is not a typical ramp configuration or design so these ramps could go against driver expectation. Without looking at exposure levels for Table 7, but based upon the number of ramps in the database, it appears that operations near entrance ramps experience a greater number of crashes than operations near exit ramps. The following crash fre- quencies were calculated per ramp for the 3-year analysis period: • 14.33 total crashes per entrance ramp (4.77 total crashes per site per year), • 1.39 truck crashes per entrance ramp (0.46 truck crashes per site per year), • 6.46 total crashes per exit ramp (2.15 total crashes per site per year), and • 0.64 truck crashes per exit ramp (0.21 truck crashes per site per year). Table 7 also indicates the location in the vicinity of the free- way mainline ramp terminal where the crashes occurred. For entrance ramps, 87 percent of the total 7,522 crashes were assigned to the mainline, and 93 percent of the 731 truck- related crashes were assigned to the mainline. The smallest proportion of entrance ramp crashes was assigned to the acceleration lane (94 total crashes and 6 truck-related crashes). For any given ramp configuration, the highest proportion of crashes assigned to the acceleration lane was 3 to 4 percent Diamond Parclo loop Free- flow loop Outer connection Direct or semi-direct connection Button hook, scissor, and slip ramps Other Total Entrance ramps Total length (mi) Acceleration lane 76.07 5.25 3.12 6.65 5.74 3.88 14.86 115.57 Freeway mainline 76.07 5.25 3.12 6.65 5.57 3.88 14.86 115.40 Ramp proper 55.61 1.93 3.85 5.38 3.11 2.21 13.16 85.25 Total 207.75 12.43 10.09 18.68 14.42 9.97 42.88 316.22 Exit ramps Total length (mi) Deceleration lane 34.26 1.90 1.43 4.65 4.20 1.48 9.87 57.79 Freeway mainline 34.26 1.90 1.43 4.65 4.25 1.48 9.87 57.84 Ramp proper 60.10 1.99 3.14 9.76 7.73 2.22 12.84 97.78 Total 128.62 5.79 6.00 19.06 16.18 5.18 32.58 213.41 Table 6. Ramp mileage by ramp configuration. Accel. Lane Mainline Decel. Lane Mainline Entrance-Ramp Ramp Proper Ramp Prop er Exit-Ramp Figure 5. Locations of crashes assigned to entrance and exit ramps.

Location Ramp configuration Grand total Diamond Parclo loop Free-flow loop Outer connection Direct or semi- direct connection Button hook, scissor, and slip ramps Others Total Truck Total Truck Total Truck Total Truck Total Truck Total Truck Total Truck Total Truck Truck percent Entrance ramps Accel. lane 28 1 3 1 3 0 7 0 16 1 2 0 35 3 94 6 6.38 Freeway mainline 2,808 297 445 38 196 11 308 42 355 36 541 50 1,895 208 6,548 682 10.42 Ramp proper 189 6 13 1 55 7 23 1 45 3 28 2 527 23 880 43 4.89 Total 3,025 304 461 40 254 18 338 43 416 40 571 52 2,457 234 7,522 731 9.72 Exit ramps Decel. lane 59 2 2 0 8 0 22 3 41 6 1 0 119 10 252 21 8.33 Freeway mainline 696 77 30 6 116 14 222 16 223 22 75 9 1,169 123 2,531 267 10.55 Ramp proper 269 24 11 2 67 4 108 10 130 10 25 3 162 12 772 65 8.42 Total 1,024 103 43 8 191 18 352 29 394 38 101 12 1,450 145 3,555 353 9.93 Grand total 4,049 407 504 48 445 36 690 72 810 78 672 64 3,907 379 11,077 1,084 9.79 Note: Accel. = acceleration, Decel. = deceleration. Table 7. Total crashes and truck crashes by ramp configuration and location.

25 for direct or semi-direct connections. It is interesting that the entrance ramp configuration with the greatest propor- tion of crashes occurring on the ramp proper is the free- flow loop ramp (22 percent of total crashes and 39 percent of truck crashes). This can be explained possibly by the need to negotiate changes in horizontal alignment, while also possibly accelerating along the ramp. For all entrance ramp configura- tions combined, 12 percent of total crashes and 6 percent of truck crashes occurred on the ramp proper. Of the 682 truck crashes that occurred on the freeway mainline, it could not be distinguished from the database whether the truck involved in the crash was merging into freeway traffic from the accelera- tion lane or whether the truck was operating on the freeway mainline and another vehicle was merging onto the freeway. For the distribution of crashes assigned to the three loca- tions along exit ramps, the majority of crashes were assigned to the freeway mainline (71 percent of total crashes and 76 percent of truck crashes); however, for both total and truck-related crashes, the percentages were lower compared to percentages of crashes assigned to the freeway mainline for entrance ramps. Very few crashes occurred on the deceleration lanes of exit ramps (7 percent of total crashes and 6 percent of truck crashes). Finally, a greater percentage of crashes occurred on the ramp proper of exit ramps (22 percent of total crashes and 18 percent of truck crashes) compared to the ramp proper of entrance ramps. This distribution is logical considering that a significant portion of crashes would be expected along the mainline as vehicles change lanes in an effort to diverge. Also, once on the deceleration lane, few crashes would be expected unless a leading vehicle decelerates at a greater rate than expected or a queue backs up into the deceleration lane (in which case this is not necessarily a deceleration lane problem but more of a storage problem). Finally, the high number of crashes on the ramp proper of exit ramps could be the result of vehicles traveling too fast to negotiate the ramp alignment or the interaction between leading and following vehicles. Table 7 also indicates that trucks were involved in approxi- mately 10 percent of the total crashes at both freeway main- line ramp terminal entrance and exit ramps. Interestingly, Golob and Regan (2003) conducted a safety evaluation of truck operations along six major urban freeways in Orange County, California. The analysis included a total of approximately 131 centerline miles. Of the 19,202 mainline crashes of these routes in 2000 through 2001, 1,952—or 10.2 percent—involved trucks or tractor-trailers larger than two-axle, four-tire pickups and vans. Thus, based on comparing the percentage of truck crashes to total crashes along different facility types, trucks are involved in approximately the same percentage of crashes along urban freeway mainlines as they are in crashes near freeway mainline ramp terminal entrance and exit ramps (i.e., 10 percent). Table 8 provides a breakdown of the crash frequency for total crashes and truck crashes based on the area type (urban versus rural) for both entrance and exit ramps. Approximately 38 percent (411) of the ramps included in the analysis were in rural areas, while approximately 62 percent (664) were in urban areas. The total number of crashes that occurred in the vicinity of rural entrance ramps was 886 in the 3-year analysis period, of which 112 involved trucks. For exit ramps, 348 total crashes occurred in the vicinity of exit ramps, of which 46 involved trucks. For urban entrance ramps, 6,636 total crashes occurred in the vicinity of the ramps, of which 619 involved trucks, and for urban exit ramps, 3,207 total crashes occurred in the vicinity of the exit ramps, of which 307 involved trucks. Not taking into consideration exposure, this translates to the following crash rates per year: • 1.41 total crashes and 0.18 truck crashes per rural entrance ramp, • 0.58 total crashes and 0.08 truck crashes per rural exit ramp, • 7.02 total crashes and 0.66 truck crashes per urban entrance ramp, and • 3.06 total crashes and 0.29 truck crashes per urban exit ramp. Interestingly, rural exit ramps had the highest percentage of truck crashes (13.22 percent), followed next by rural entrance Location Rural Urban Grand Total Total Truck Truck percent Total Truck Truck percent Total Truck Truck percent Entrance ramps Acceleration lane 10 0 0.00 84 6 7.14 94 6 6.38 Freeway mainline 816 106 12.99 5,732 576 10.05 9,548 682 10.42 Ramp proper 60 6 10.00 820 37 4.51 880 43 4.89 Total 886 112 12.64 6,636 619 9.33 7,522 731 9.72 Total number of entrance ramps 210 315 525 Exit ramps Deceleration lane 23 1 4.35 229 20 8.73 252 21 8.33 Freew ay mainline 193 32 16.58 2,338 235 10.05 2,531 267 10.55 Ramp proper 132 13 9.85 640 52 8.13 772 65 8.42 Total 348 46 13.22 3,207 307 9.57 3,555 353 9.93 Total number of exit ramps 201 349 550 Grand total 1,234 158 12.80 9,843 926 9.41 11,077 1,084 9.79 Table 8. Total crashes and truck crashes by area type.

26 ramps (12.64 percent). In urban areas, truck crashes accounted for less than 10 percent of the total crashes on both entrance ramps (9.33 percent) and exit ramps (9.57 percent). Table 9 shows the distributions of severity levels for total crashes and truck crashes for entrance ramps, exit ramps, and combined. For entrance ramps, truck-related fatal crashes are a slightly higher percentage (0.41 percent) of the total truck crashes than the percentage of fatal crashes for total crashes (0.27 percent). However, it should be pointed out that the pro- portion of fatal crashes for both total and truck-related crashes is still relatively low (i.e., less than 0.5 percent) in both cases. The proportion of injury crashes for total crashes (36.24 percent) is about 5 percentage points higher than for truck crashes (30.92 percent). Consequently, the proportion of property- damage-only crashes for total crashes (63.49 percent) is about 5 percentage points lower than for truck crashes (68.67 per- cent). Because of the extra mass associated with trucks, it is rea- sonable to expect that the proportion of fatal-and-injury truck crashes would be greater than the proportion of fatal-and- injury crashes for total crashes, but this was not the case. The proportions of severity levels for exit ramps were very similar to the proportions for entrance ramps. The data are not specifically provided in Table 9, but calcu- lations were performed to compare the severity levels of the crash data from the Washington database to national averages for large truck crashes. In 2004, the National Highway Traffic Safety Administration (NHTSA) (2005) reported that one out of eight traffic fatalities resulted from a collision involving a large truck. In the four fatal truck crashes in the Washington database, five individuals were killed, while in the other 27 fatal crashes, 33 fatalities occurred. In other words, approximately 13 percent of all traffic fatalities at freeway-ramp terminals in Washington involved a large truck, compared to the 12 percent of all traffic fatalities reported in 2004 for all facility types com- bined. NHTSA (2005) also reported that in 2004, large trucks accounted for 8 percent of all vehicles involved in fatal crashes and 4 percent of all vehicles involved in injury and property- damage-only crashes. From the Washington ramp database, large trucks accounted for 11 percent of all vehicles involved in fatal crashes and 7 percent of all vehicles involved in injury and property-damage-only crashes. Table 10 shows the distribution of total crashes by crash type. Of the 7,522 total crashes that occurred near entrance ramps, 2,383 (32 percent) were single-vehicle crashes and 5,139 (68 percent) were multiple-vehicle crashes. Rear-end crashes (3,627) on the mainline accounted for approximately 48 percent of the crashes at freeway-ramp terminal entrance ramps. As would be expected, sideswipe, same-direction crashes were also common along the mainline at entrance ramps (12 percent [919] of total crashes at entrance ramps). What was unexpected was the high number of single-vehicle, fixed-object crashes that occurred on the freeway mainline (14 percent of total crashes at entrance ramps). One explanation may be that vehicles on the ramp have difficulty merging into the traffic lane and strike an object (e.g., guard rail) just beyond the end of the acceleration lane on the right shoulder. Another expla- nation may be that merging traffic from the entrance ramp either crosses the mainline lanes or causes a disturbance along the mainline that forces traffic on the mainline to take evasive maneuvers and cause a run-off-the-road-to-the-left crash or median crash. Considering the exit ramp crashes, 36 percent were single-vehicle crashes and 64 percent were multiple- vehicle crashes. Among the multiple-vehicle crashes, rear-end and same-direction sideswipe crashes were the most common types along the mainline, accounting for approximately 38 and 11 percent of total crashes, respectively. Single-vehicle fixed- object crashes along the freeway mainline and along the ramp proper accounted for approximately 12 and 11 percent of total crashes, respectively. These single-vehicle crashes could be the result of vehicles striking a crash attenuation system at the gore between the freeway mainline and ramp and/or vehicles having difficulty negotiating critical curves along the ramp proper due to high speed. Table 11 shows the distribution of truck crashes by crash type. Of the 731 truck crashes that occurred near entrance ramps, 18 percent were single-vehicle crashes and 82 percent were multiple-vehicle crashes. Rear-end crashes along the mainline accounted for approximately 35 percent of the truck crashes at Data Severity Percent Fatal Injury PDO Fatal Injury PDO Entrance ramps Total crashes 20 2,726 4,776 0.27 36.24 63.49 Truck crashes 3 226 502 0.41 30.92 68.72 Exit ramps Total crashes 11 1,318 2,226 0.31 37.07 62.62 Truck crashes 1 115 237 0.28 32.58 67.14 Total Total crashes 31 4,044 7,002 0.28 36.51 63.21 Truck crashes 4 341 739 0.37 31.46 68.17 Note: PDO = property damage only. Table 9. Total crashes and truck crashes by severity level.

Location Single-vehicle crashes Multiple-vehicle crashes Totals C o l l i s i o n w i t h p a r k e d m o t o r v e h i c l e C o l l i s i o n w i t h r a i l r o a d t r a i n C o l l i s i o n w i t h b i c y c l i s t C o l l i s i o n w i t h p e d e s t r i a n C o l l i s i o n w i t h a n i m a l C o l l i s i o n w i t h f i x e d o b j e c t C o l l i s i o n w i t h o t h e r o b j e c t O t h e r s i n g l e - v e h i c l e c o l l i s i o n O v e r t u r n F i r e o r e x p l o s i o n O t h e r s i n g l e - v e h i c l e n o n - c o l l i s i o n R e a r - e n d H e a d - o n R e a r - t o - r e a r A n g l e S i d e s w i p e , s a m e d i r e c t i o n S i d e s w i p e , o p p o s i t e d i r e c t i o n O t h e r m u l t i p l e - v e h i c l e c o l l i s i o n Entrance ramps Accel. lane 3 0 0 0 0 39 1 0 6 2 4 24 0 0 0 10 0 5 94 Freeway mainline 41 0 0 5 84 1,079 85 16 241 68 62 3,627 3 0 0 919 3 315 6,548 Ramp proper 12 0 1 2 4 514 2 1 101 6 4 153 2 0 0 47 0 31 880 Total 56 0 1 7 88 1,632 88 17 348 76 70 3,804 5 0 0 976 3 351 7,522 Exit ramps Decel. lane 2 0 0 0 2 78 0 0 18 3 2 107 1 0 1 32 0 6 252 Freeway mainline 23 0 0 6 25 425 20 7 72 22 23 1,382 4 0 1 401 0 120 2,531 Ramp proper 13 0 0 1 3 382 1 2 118 29 11 166 2 0 0 35 1 8 772 Total 38 0 0 7 30 885 21 9 208 54 36 1,655 7 0 2 468 1 134 3,555 Grand total 94 0 1 14 118 2,517 109 26 556 130 106 5,459 12 0 2 1,444 4 485 11,077 Table 10. Total crashes by crash type.

Location Single-vehicle crashes Multiple-vehicle crashes Totals C o l l i s i o n w i t h p a r k e d m o t o r v e h i c l e C o l l i s i o n w i t h r a i l r o a d t r a i n C o l l i s i o n w i t h b i c y c l i s t C o l l i s i o n w i t h p e d e s t r i a n C o l l i s i o n w i t h a n i m a l C o l l i s i o n w i t h f i x e d o b j e c t C o l l i s i o n w i t h o t h e r o b j e c t O t h e r s i n g l e - v e h i c l e c o l l i s i o n O v e r t u r n F i r e o r e x p l o s i o n O t h e r s i n g l e - v e h i c l e n o n - c o l l i s i o n R e a r - e n d H e a d - o n R e a r - t o - r e a r A n g l e S i d e s w i p e , s a m e d i r e c t i o n S i d e s w i p e , o p p o s i t e d i r e c t i o n O t h e r m u l t i p l e - v e h i c l e c o l l i s i o n Entrance ramps Accel. lane 1 0 0 0 0 2 0 0 0 1 0 1 0 0 0 1 0 0 6 Freeway mainline 6 0 0 0 3 48 12 2 8 3 25 260 0 0 0 248 0 67 682 Ramp proper 2 0 0 0 0 9 0 0 11 0 2 8 0 0 0 5 0 6 43 Total 9 0 0 0 3 59 12 2 19 4 27 269 0 0 0 254 0 73 731 Exit ramps Decel. lane 0 0 0 0 0 2 0 0 2 0 1 9 0 0 0 6 0 1 21 Freeway mainline 4 0 0 1 0 13 6 1 3 3 3 112 0 0 1 100 0 20 267 Ramp proper 5 0 0 0 0 14 0 0 14 3 4 12 0 0 0 10 0 3 65 Total 9 0 0 1 0 29 6 1 19 6 8 133 0 0 1 116 0 24 353 Grand total 18 0 0 1 3 88 18 3 38 10 35 402 0 0 1 370 0 97 1,084 Table 11. Truck crashes by crash type.

29 entrance ramps, and sideswipe, same-direction crashes along the mainline were also common at freeway-ramp terminal entrance ramps (34 percent of truck crashes at entrance ramps). As with total crashes, fixed-object crashes were the most com- mon type of truck-related single-vehicle crash at entrance ramps. Other multiple-vehicle collisions were another com- mon type of truck crash at entrance ramps. Considering the truck-related exit ramp crashes, 22 percent were single-vehicle crashes and 78 percent were multiple-vehicle crashes. Among the multiple-vehicle crashes, rear-end and sideswipe, same- direction crashes along the mainline were the most common. Single-vehicle fixed-object crashes along the freeway mainline and along the ramp proper each accounted for approximately 4 percent of truck crashes. Table 12 categorizes the distribution of truck crashes by crash type based upon three ramp truck average daily traffic (RTADT) levels: • RTADT < 500, • 500 ≤ RTADT < 1000, and • RTADT ≥ 1000. This stratification of RTADT was selected somewhat arbi- trarily, but it illustrates that the majority of truck crashes occurred at ramps with RTADT levels above 1,000 trucks/ day. In part, this could have to do with exposure level, and it could also be related to the operational conditions of the traffic stream. Ramps with higher RTADT likely experience poorer operating conditions throughout the day because of greater traffic volumes in general, thus reducing the free- dom to maneuver within the traffic stream. It could also be a function of the number of ramps included in each categorization. Table 13 shows the crash rates for total crashes and truck crashes by ramp configuration and ramp location. For entrance ramps, the overall crash rates were approximately the same for total crashes (0.81 crashes/million vehicle miles traveled [MVMT]) and truck crashes (0.80 crashes/MVMT). Simi- larly for exit ramps, the overall crash rates were approximately equal for total crashes (0.67 crashes/MVMT) and truck crashes (0.70 crashes/MVMT). For the various ramp configurations, there was no consistent trend suggesting that trucks were more or less likely to be involved in a crash along a certain location of the ramp for either entrance or exit ramps. The four ramp configurations where the truck crash rate exceeded the overall total crash rate by the greatest amount included: • Outer connection entrance ramp (difference between crash rates—0.18 crashes/MVMT), • Parclo loop exit ramp (difference between crash rates— 0.22 crashes/MVMT), • Free-flow loop exit ramp (difference between crash rates— 0.22 crashes/MVMT), and • Other exit ramp (difference between crash rates—0.19 crashes/MVMT). Related to Table 13, NHTSA (2005) reported that trucks were involved in fatal crashes at a rate of 2.19 crashes/100 MVMT taking consideration of truck operations along the entire road- way network. Based upon the Washington ramp database, trucks were involved in fatal crashes at a rate of 0.28 crashes/100 MVMT in the vicinity of entrance and exit ramps. 4.3 Key Findings This section highlights several key findings from the crash analysis, focusing on comparing crash frequencies/rates for all vehicle types combined (i.e., total crashes) to crash frequencies/ rates for heavy vehicles (i.e., trucks). The findings from this crash analysis combined with the findings from the obser- vational study described in Section 5, provide the basis for recommending a particular type of design vehicle for freeway mainline ramp terminals. The key findings are as follows: 1. The crash rates for all vehicles compared to the crash rates of trucks are for practical purposes the same at both entrance and exit ramps. At entrance ramps, the crash rate for all vehicles is 0.81 crashes/MVMT compared to 0.80 crashes/MVMT for truck-related crashes. At exit ramps, the crash rate for all vehicles is 0.67 crashes/MVMT com- pared to 0.70 crashes/MVMT for truck-related crashes. 2. Trucks are involved in approximately 10 percent of all crashes in the vicinity of freeway mainline ramp terminals. As a comparison, trucks are involved in approximately 10 percent of all crashes along the mainline of urban freeways (Golob and Regan, 2003). Thus, truck crashes are neither over-represented nor under-represented near free- way mainline ramp terminals. 3. Rural exit ramps had the highest percentage of truck crashes (13.2 percent), followed by rural entrance ramps (12.6 per- cent). In urban areas, truck crashes accounted for less than 10 percent of the total crashes on both entrance ramps (9.3 percent) and exit ramps (9.6 percent). Several explanations for having a slightly higher percentage of truck crashes at rural freeway mainline ramp terminals as compared to urban terminals include potentially greater truck exposure levels in rural areas, driver adaptation to the high-speed rural environment, and greater driver attention levels in urban areas. 4. When comparing the severity distributions at the person level, approximately 13 percent of all traffic fatalities at freeway mainline ramp terminals in Washington involved

Area Single-vehicle crashes Multiple-vehicle crashes T otal C o l l i s i o n w i t h p a r k e d m o t o r v e h i c l e C o l l i s i o n w i t h r a i l r o a d t r a i n C o l l i s i o n w i t h b i c y c l i s t C o l l i s i o n w i t h p e d e s t r i a n C o l l i s i o n w i t h a n i m a l C o l l i s i o n w i t h f i x e d o b j e c t C o l l i s i o n w i t h o t h e r o b j e c t O t h e r s i n g l e - v e h i c l e c o l l i s i o n O v e r t u r n F i r e o r e x p l o s i o n O t h e r s i n g l e - v e h i c l e n o n - c o l l i s i o n R e a r - e n d H e a d - o n R e a r - t o - r e a r A n g l e S i d e s w i p e , s a m e d i r e c t i o n S i d e s w i p e , o p p o s i t e d i r e c t i o n O t h e r m u l t i p l e - v e h i c l e c o l l i s i o n RT AD T < 500 Entrance ramp s Acceleration lane 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 Free wa y mainline 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ramp proper 1 0 0 0 0 3 0 0 1 0 1 0 0 0 0 0 0 2 8 To tal 1 0 0 0 0 3 0 0 1 1 1 0 0 0 0 0 0 2 9 Exit ramps Deceleration lane 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 0 0 3 Free wa y mainline 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ramp proper 1 0 0 0 0 7 0 0 3 2 1 3 0 0 0 1 0 1 1 9 To tal 1 0 0 0 0 7 0 0 4 2 1 4 0 0 0 2 0 1 2 2 500 < R TA DT < 1,000 Entrance ramp s Acceleration lane 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 2 Free wa y mainline 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ramp proper 1 0 0 0 0 3 0 0 2 0 0 4 0 0 0 2 0 1 13 To tal 1 0 0 0 0 4 0 0 2 0 0 4 0 0 0 3 0 1 15 Exit ramps Deceleration lane 0 0 0 0 0 1 0 0 1 0 0 1 0 0 0 0 0 1 4 Free wa y mainline 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ramp proper 1 0 0 0 0 4 0 0 8 1 2 2 0 0 0 2 0 0 20 To tal 1 0 0 0 0 5 0 0 9 1 2 3 0 0 0 2 0 1 24 RT AD T > 1,000 Entrance ramp s Acceleration lane 1 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 3 Free wa y mainline 6 0 0 0 3 48 12 2 8 3 2 5 260 0 0 0 248 0 6 7 682 Ramp proper 0 0 0 0 0 3 0 0 8 0 1 4 0 0 0 3 0 3 2 2 To tal 7 0 0 0 3 5 2 12 2 16 3 2 6 265 0 0 0 251 0 7 0 707 Exit ramps Deceleration lane 0 0 0 1 0 0 0 0 1 7 0 0 0 5 0 0 1 4 Free wa y mainline 4 0 0 1 0 13 6 1 3 3 3 112 0 0 1 100 0 2 0 267 Ramp proper 3 0 0 0 0 3 0 0 3 0 1 7 0 0 0 7 0 2 2 6 To tal 7 0 0 1 0 1 7 6 1 6 3 5 126 0 0 1 112 0 2 2 307 Table 12. Truck crashes by crash type and RTADT.

Area Ramp configuration Total Diamond Parclo loop Free-flow loop Outer connection Direct or semi-direct connection Button hook, scissor, and slip ramps Others Total Truck Total Truck Total Truck Total Truck Total Truck Total Truck Total Truck Total Truck Entrance ramps Acceleration lane 0 .09 0 .03 0 .13 0 .40 0 .16 0 .00 0 .21 0 .00 0 .37 0 .26 0 .08 0 .00 0 .29 0 .26 0 .16 0 .10 Freeway mainline 0 .60 0 .63 1 .27 1 .11 1 .14 0 .59 0 .69 0 .96 0 .82 0 .95 1 .43 1 .45 1 .09 1 .36 0 .80 0 .86 Ramp proper 0.85 0 .22 1 .41 0 .94 1 .96 2 .08 0 .93 0 .34 1 .54 1 .32 1 .80 1 .49 3 .94 1 .83 1 .90 0 .85 Total 0.58 0 .57 1 .21 1 .06 1 .16 0 .76 0 .67 0 .85 0 .82 0 .91 1 .35 1 .36 1 .23 1 .32 0 .81 0 .80 Exit ramps Deceleration lane 0 .37 0 .12 0 .35 0 .00 0 .55 0 .00 0 .74 0 .87 0 .96 1 .50 0 .14 0 .00 1 .03 1 .00 0 .68 0 .57 Freeway mainline 0 .32 0 .36 0 .44 0 .68 1 .25 1 .80 0 .62 0 .47 0 .73 0 .91 0 .65 0 .75 0 .96 1 .22 0 .59 0 .67 Ramp proper 1.00 0 .79 1 .73 1 .95 2 .72 2 .09 1 .68 1 .23 1 .44 1 .22 1 .87 1 .56 1 .28 1 .04 1 .30 1 .03 Total 0.40 0 .40 0 .53 0 .75 1 .45 1 .67 0 .78 0 .63 0 .90 1 .04 0 .74 0 .81 1 .00 1 .19 0 .67 0 .70 Table 13. Total crash rate (crashes/MVMT) and truck crash rate (crashes/MVMT) by ramp configuration and area.

32 a large truck. This is consistent with the national figures reported by NHTSA indicating that approximately 12 per- cent of all traffic fatalities involve a large truck. Thus, when comparing severity levels at the person level, trucks are not over- or under-represented in the fatal crashes that occur near freeway mainline ramp terminals. 5. When comparing the severity distributions at the vehicle level, trucks are involved in a slightly greater percentage of the fatal crashes (11 percent) and injury and property- damage-only crashes (7 percent) near freeway mainline ramp terminals compared to the national statistics reported by NHTSA for fatal crashes (8 percent) and injury and property-damage-only crashes (4 percent). 6. The entrance ramp configurations where the truck crash rates exceeded the overall total crash rates by the greatest amounts included (a) outer connection, (b) direct or semi- direct connection, and (c) “other” ramps. With the excep- tion of the “other” ramp configuration, where drivers’ expectations may be violated, there is no reasonable expla- nation as to why the truck crash rates exceeded the overall total crash rates by the greatest amounts on these particular entrance ramp configurations. 7. The exit ramp configurations where the truck crash rates exceeded the overall total crash rates by the greatest amounts included (a) parclo loop, (b) free-flow loop, and (c) “other” ramps. In particular, since both parclo and free-flow loop exit ramps have curved alignments, there is concern that trucks may rollover at these types of ramps if trucks are traveling substantially faster than the design speed of the first critical curve. 8. When considering the distribution of crash types at entrance ramps, 48 percent of total crashes are rear-end crashes and 12 percent are sideswipe, same-direction crashes compared to 35 percent for rear-end and 34 percent for sideswipe, same-direction crashes for truck crashes. Combined rear- end and sideswipe, same-direction crashes account for approximately 60 percent of total crashes compared to approximately 69 percent of truck crashes. Thus, trucks are slightly over-represented in these two crash types as might be expected with vehicles operating at lower speeds and acceleration rates. 9. When the RTADT exceeds 1,000 trucks/day, it appears that truck crashes occur more frequently compared to freeway mainline ramp terminals with lower RTADT.