National Academies Press: OpenBook

Development of a Posted Speed Limit Setting Procedure and Tool (2021)

Chapter: APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS

« Previous: CHAPTER 4. CONCLUSIONS, OPTIONS, AND SUGGESTED RESEARCH
Page 31
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 31
Page 32
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 32
Page 33
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 33
Page 34
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 34
Page 35
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 35
Page 36
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 36
Page 37
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 37
Page 38
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 38
Page 39
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 39
Page 40
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 40
Page 41
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 41
Page 42
Suggested Citation:"APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 42

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.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 19 APPENDIX A. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR URBAN/SUBURBAN STREETS Several factors are known or are suspected to affect driver speed selection or crashes on a city street. The following sections discuss the findings reported in the research literature on the relationships among traffic crashes, operating speed, and roadway factors, including posted speed limit. Note that the goal of this literature review was to identify variables that influence driver speed choice and crash potential. These variables would then be considered for inclusion in the SLS-Procedure that was the key objective for this research project (NCHRP Project 17-76). TRAFFIC—VEHICLES Annual Average Daily Traffic Relationship with Crashes Annual average daily traffic (AADT) is positively correlated with crash frequencies (19). Greibe (20) developed generalized linear models (GLMs) to predict crashes based on data from 88 mi of roads located in urban areas in Denmark. The author divided these roads into 314 homogeneous segments averaging 450 m in length. AADT was the most powerful predictor of crashes. Das and Abdel-Aty (21) found higher average daily traffic (ADT) was associated with increased crashes, and the absence of on-street parking was associated with reduced injury severity in crashes. Traffic flow (often referred to as general AADT) is considered the most determinant variable for the occurrence of crashes and, as such, is typically used as the only variable in the model (22, 23, 24). Other studies have also found that AADT contributes to speed selection (25, 26, 27, 28, 29). The findings of a study conducted by Dong et al. (30) suggested that major road traffic volume contributes positively to the frequency of all crash types, and the same results were observed with respect to minor road traffic volume. Congestion Relationship with Crashes Kononov et al. (31) found a direct relationship between congestion and crashes. Additionally, according to Hanbali and Fornal (32), the traffic flow fluctuation in congestion can also cause increased collisions. Stipancic et al. (33) also showed that congestion index is positively correlated with crash counts. Percent Trucks Relationship with Crashes Results suggest that the involvement of car crashes decreases when truck percentages increase. One possible reason is that the frequency of lane changing and overtaking movements by cars decreases with a higher presence of trucks. On the other hand, the frequencies of truck- involved crashes and car-truck-involved crashes increase as the truck percentage increases. Additionally, one study showed that more trucks are involved in truck-car crashes than in truck- truck and single-truck crashes (34). Operating Speed Relationship with Crashes Gargoum and El-Basyouny (10) used data for 353 two-lane urban roads in the city of Edmonton, Canada, to explore the association between speed and safety. The posted speed limits

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 20 for these roads ranged between 25 and 56 mph (U.S. standard unit equivalents for speed limits posted with metric values). The average speeds were obtained from on-pavement sensors installed between a couple of days to a week at a site. The speed data were collected over a 5-year period. Only vehicles with a headway of 6 seconds or more (i.e., free flow) were used to determine the average speed. The speed data were linked to the crash frequency at each location during the same time frame, along with other factors such as road, traffic, and climate. The results showed the following as having statistically significant effects on crashes: average speed, traffic volume, segment length, median presence, and horizontal curve. Average speed was found to be positively correlated, indicating that a higher crash frequency is anticipated at road segments with higher average speeds. Wang et al. (35) integrated spatio-temporal speed fluctuation of a single vehicle with speed differences between vehicles using taxi-based high-frequency global positioning system (GPS) data collected from 234 one-way urban arterial segments in Shanghai. The results showed that a 1 percent increase in mean speed was found to be related to a 0.7 percent higher crash frequency. Percentage of Trucks Relationship with Operating Speed Islam and El-Basyouny (36) showed that the proportion of vans, buses, and trucks was found to have a positive correlation with the free-flow speed. Vehicle Characteristics Relationship with Operating Speed Vehicle characteristics, such as vehicle class and vehicle age, also impact speed choice. Wasielewski (37) found that drivers of heavy vehicles chose a higher speed than drivers of passenger cars. Giles (38) also reached similar conclusions by observing that increases in vehicle length were associated with increased speeds. Studies have also found that newer vehicles seem to be driven at higher speeds than older ones (37, 39). TRAFFIC CONTROL DEVICE Posted Speed Limit Relationship with Crashes The earliest study to examine the relationship between speed and crash rate was completed by Solomon (14). A subsequent study by Cirillo (40) in 1968 confirmed Solomon’s findings by using both rural and urban roadways. To identify the relationship between safety and speed variance, Garber and Gadiraju (41) examined all types of roads in Virginia, including highways, arterials, and major rural collectors. The developed relationship did not follow Solomon’s U-shaped curve. Fildes et al. (39) conducted a survey study by conducting interviews with drivers after observing operative speed relative to the speed limit on two urban (35 mph) roads in Australia. Numerous studies since Fildes et al. have sought to identify association between speed and safety. Elvik performed a meta-analysis (42) by incorporating 115 studies with 526 effect estimates. Greibe (20) found speed limit as a significant factor. Dixit et al. (43) found higher speeds were positively correlated with both rear-end and angle crash rates. Tay (44) found that crashes on roadways with lower posted speeds (35 mph to 45 mph) were more likely to occur in urban areas. Since higher speed requires longer stopping distance, studies identified a positive relationship between speed and crash counts (45, 46) because higher speed requires longer stopping distance. The negative relationship found in Shi et al. (47) is mostly due to the

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 21 selection of an expressway of interest that is located in the urban area with a high percentage of commuter traffic. A 2016 study (10) matched average speed with crashes for 353 segments in the city of Edmonton, Canada. The study found that higher speeds were associated with more crashes; however, the direct effects of PSL on crash frequency were statistically insignificant. The authors noted that “it is the higher speeds on roads with higher speed limits that cause collisions and not the speed limits (i.e., PSLs affect collisions through their effects on speeds).” A 2015 study by Islam and El-Basyouny (36) included 27 urban residential collector sites where the PSL was reduced from 31 mph to 25 mph. Another 287 sites where the PSL remained at 31 mph were selected as reference sites. All of the treated and reference sites were two-lane (one lane in each direction) urban residential collector road segments. The authors noted that “a comprehensive effort was made, involving various education and enforcement activities, as well as the placement of the new PSL signs along each of the segments. Three years of before and 3 years of after crash data were included in the analyses. The full Bayesian before-after evaluation found that the PSL reduction was effective in reducing crashes of all severities. Posted Speed Limit Relationship with Operating Speed Several studies have found posted speed limits have a significant effect on free-flow speed on urban streets (48, 49, 50, 51, 52, 53, 54). In 2001, Fitzpatrick et al. (48) considered data from 19 horizontal curve sites and 36 straight section sites on four-lane suburban arterials in Texas. When all variables were considered, posted speed limit was the most significant variable for both curves and straight sections. For typical suburban arterials (e.g., 30 to 45 mph), all speed limit values influenced operating speed. Other significant variables for curve sections were deflection angle and access density class. A later Fitzpatrick et al. study (49) modeled operating speeds at 78 suburban/urban sites in six states and only found posted speed limit as being a statistically significant predictor of the 85th percentile operating speed. Segment access density was included as a predictor of operating speed in a second model; however, it was not statistically significant. A study by Ali et al. (50) examined 35 four-lane urban street segments with posted speed limits between 35 and 45 mph. Median width and segment length ratio (defined as the ratio of study segment length to the maximum signal spacing) were also significant. Figureroa and Tarko (51) included both rural and suburban roadways. In addition to posted speed limit, they also found other variables that are statistically significant. Those most commonly associated with suburban roads would include intersection and driveway density along with the presence of a TWLTL. More recent research has included larger numbers of study locations. A study conducted by Thiessen et al. (53) considered 249 tangent segments in Edmonton with equivalent posted speed limits ranging from 24.8 to 62.1 mph (values provided are converted from metric units). They found that arterials with higher PSLs had higher operating speeds. They offered a caution regarding the finding for the collector roads since most had the same posted speed limit. A study by Eluru et al. (54) considered 49 collectors, with about half having a 24.8-mph posted speed limit and the other half being 31.1 mph. They also studied 71 arterials with a posted speed limit range of 18.6 to 43.5 mph but with almost all being 31.1 mph. Posted speed limit was only significant for the arterial streets. A 2016 study (10) matched average speed with crashes for 353 segments in the city of Edmonton, Canada. The study found that higher average speeds were associated with higher PSLs.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 22 Signalized Intersection Presence Relationship with Crashes As report in 2013, nearly 21 percent of all crashes and 24 percent of all fatal and injury crashes occurred at signalized intersections in the United States (55). In place of examining the presence of a signalized intersection in safety estimation, many studies have used crashes associated with signalized intersections to identify the significant factors (29, 56, 57). ROADWAY GEOMETRY Horizontal Alignment Relationship with Crashes Past studies have found mixed effects of horizontal curves. Hauer (58), Abdel-Aty and Radwan (59), and Bonneson et al. (60) suggested that sharper curvature would increase the likelihood of crash occurrences. Abdel-Aty and Abdalla (61) also found that the presence of horizontal curvature increases the likelihood of a crash occurrence. In contrast, Milton and Mannering (62), Haynes et al. (63), and Ahmed et al. (64) found that sharper horizontal curves were negatively related to crash frequency since drivers tend to drive slowly on those segments. Horizontal or Vertical Alignment Relationship with Operating Speed Using data from curved road segments in Australia, McLean (65) developed a regression model to predict speeds on horizontal curves. The study found that attributes of horizontal curves (e.g., radius of curve and degree of curvature) had major effects on driver speed choice. Fitzpatrick et al. (66) studied 14 horizontal curves and 10 vertical curves on suburban roadways in Texas. Regression analysis indicated that the curve radius for horizontal curves and the inferred design speed for vertical curves can be used to predict the 85th percentile speed on curves for vehicles on the outside lane of a four-lane divided suburban arterial. A study by Poe and Mason (67) collected speed data for 27 urban collectors and found the degree of curve and grade as being significant variables. A study by Eluru et al. (54) for local streets found lower speeds for sites with higher vertical grades. In contrast, Yagar and Van Aerde (68) used data from Ontario, Canada, to develop a multiple linear regression model. The study found that road curvature had no statistically significant effects on speeds. Similarly, Fildes et al. (39) did not find significant differences in speed choice when comparing curved and straight segments. Vertical Alignment Relationship with Crashes Yanmaz-Tuzel and Ozbay (69) conducted a before-after analysis on four urban arterials in New Jersey with speed limits less than 45 mph. This study examined the effects of increases in lane width, installation of median barriers, and improvements to vertical and horizontal road alignments. Using a full Bayesian approach, this study observed reductions in crash rate after adjusting the vertical alignments. Median Relationship with Crashes The Yanmaz-Tuzel and Ozbay (69) study described above showed that the increase of median barrier installations was associated with crash rate reduction for an urban principal arterial site in New Jersey. Two studies showed that installations of a TWLTL on urban four-lane roadways in Louisiana were associated with crash reductions (70, 71). Rahman et al. (70) examined nine sites

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 23 where the four-lane undivided (4U) cross-section was converted to a cross-section with four lanes and a TWLTL (5T). A reduction of crashes was experienced at all the sites, with an estimated overall crash modification factor (CMF) of 0.482. The relationship of CMF with driveway density showed that 4U to 5T conversion performed better in driveway densities up to 55 driveways per mile. The Sun et al. (71) study focused on two corridors in Louisiana where the four-lane undivided roadway was converted to a 5T. They found a reduction in crashes, especially in rear-end crashes. A study by Schultz et al. (72) summarized research performed to identify relationships between access management and the safety characteristics of arterial road segments to better quantify the effectiveness of access management principles and techniques. The CMF Clearinghouse (73) reported that its study found a CMF of 0.29 for all crash types on urban principal arterials with the installation of a raised median. Median Relationship with Operating Speed In the 2001 Fitzpatrick et al. study (48), when the posted speed was not included in the model on curve sections, median presence and roadside development were significant variables with respect to operating speed. Speeds were lower when either the median with a TWLTL or no median was present as compared to when a raised median was present. This finding is similar to the Bonneson and McCoy study (74). Median Width Relationship with Crashes Previous studies suggested that wider median widths are associated with higher car and car-truck crash frequencies. Wider median widths allow greater degrees of spatial freedom for turning vehicles. Moreover, wider medians increase the likelihood of wrong-way driving. Medians with openings increase U-turn movements that have a negative effect on traffic safety. The potential for such problems is limited where crossroad and U-turn volumes are low but may increase at higher volumes (30). In the same context, increasing the median width also reduces rear-end, sideswipe, total, and fatal/injury crashes. Another study showed that a reduction in median width could increase rear-end, sideswipe, total, and fatal/injury crashes (75). Median Width Relationship with Operating Speed The study by Eluru et al. (54) found an increase in operating speeds for an increase in median width. Lu et al.’s (76) study also found that speed increases with the increase of median width. Number of Lanes Relationship with Crashes To develop safety prediction models for different geometric characteristics, EI-Basyouny and Sayed (27) conducted a study on 392 urban road segments in Vancouver. They found that the number of lanes is positively associated with crash frequency. Gomes (28) developed a crash prediction model for urban roads located in Lisbon, Portugal. He used a negative binomial (NB) model to examine crashes in relation to vehicle and pedestrian traffic flow and highway geometric design features. The crash prediction model identified that roadways with four lanes or more are associated with crash frequencies on urban roads.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 24 Number of Lanes Relationship with Operating Speed A study by Wang et al. (77) that considered 35 tangent corridors in Atlanta, Georgia, and a study by Eluru et al. (54) that considered 49 local streets and 71 arterials found that the number of lanes was positively associated with operating speed. Similar results were found in another study (78). Aarts et al. (79) developed the SaCredSpeed algorithm by using input data of design, image, and traffic characteristics to assess a safe speed and found the number of lanes as a significant variable, as opposed to the Lu et al. (76) study that did not find any association between number of lanes and speed. Lane Width Relationship with Crashes Yanmaz-Tuzel and Ozbay (69) observed that an increase in lane width is associated with crash rate reduction. Other studies indicated that drivers show improved lane keeping, more accurate steering behavior, and reduced driving speed with decreased lane width (80). The results in Dong et al.’s (30) study show that decreased lane width has positive effects on intersection safety for passenger cars. The safety effects of the roadways with narrow lane width can be higher than the roadways with wide lane width for specific roadway conditions (81, 82). Lane Width Relationship with Operating Speed In the 2001 Fitzpatrick et al. (48) study, when the posted speed was not included in the model, only lane width was a significant variable for straight sections, with higher speeds associated with wider lane widths. The study by Poe and Mason (67) found lane width as being one of the significant variables. Speeds increase on wider lanes for the tangent segment and decrease for wider lanes at the point of curvature and the midpoint of the curve, attributed to wider lanes within the curve. Yagar and Van Aerde (68) used data from Ontario, Canada, to develop a multiple linear regression model. Lane width was found to be statistically significant. The findings of Islam and El-Basyouny (36) contradicted the common finding that wider roadways encourage speeding. Shoulder Width Relationship with Crashes Haleem et al. (75) found that increasing the outside shoulder could reduce the total, rear- end, and fatal/injury crashes. Curb or Shoulder Relationship with Operating Speed A study by Wang et al. (77) considered 35 tangent corridors in Atlanta, Georgia, and found the presence of curb as being positively associated with operating speed. Bicycle Lanes Relationship with Crashes Park et al. (83) estimated the safety effectiveness of bike lanes using the cross-sectional method. This study showed that installation of bike lanes has positive safety effects on reducing four different crash types: total crashes, injury crashes, bike crashes, and bike injury crashes. Chen et al. (84) evaluated the safety effects of the installation of on-street bicycle lanes in New York. The results showed that the installation of bicycle lanes did not lead to an increase in crashes. One possible explanation is that speeding nature and number of collisions between vehicles and bicyclists decreased due to the presence of bike lanes. Sadek et al. (85) performed a survey analysis on bike lane installations. The results showed that bike lanes increased awareness

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 25 of drivers and bicyclists. On the other hand, Rodegerdts et al. (86) suggested that adding a bike lane reduces bike-related crashes (KABCO). Nosal and Miranda-Moreno (87) estimated the injury risk of bicycle facilities (cycle tracks, bicycle lanes) and explored the differences in injury risk between different types of bicycle facilities in Montreal, Canada. The study compared injury risk between the treated sites and control streets to assess the impact of bicycle facilities. The results showed that the safety effects of cycle tracks and bicycle lanes of treated streets were higher than the corresponding control streets. Similar to this study, Lusk et al. (88) also found that relative risk of riding bicycles on the cycle tracks versus on regular streets was lower in injury rates. Reynolds et al. (89) reviewed 23 studies that assessed the effect of transportation infrastructure on bicyclist safety. Based on the previous studies that examined impacts of infrastructures at straightaways (e.g., bike lanes or paths) and intersections (e.g., roundabouts, traffic lights), they found that bicycle-specific facilities generally reduced crashes and injuries. On the other hand, Jensen (90) concluded that adding a bike lane increased frequencies of all crashes (KABCO, KABC) and bike crashes (KABCO) for roadways in Copenhagen, Denmark. Bicycle Lane Presence Relationship with Operating Speed The study by Eluru et al. (54) that considered 49 local streets of Montreal along with the study by Thiessen et al. (53) of 249 tangent segments in Edmonton found the presence of bicycle lanes being associated with higher vehicle speeds. They commented that the variable was probably a reflection that bicycle lanes are typically installed on wider roads with high vehicular flow. Segment Length Relationship with Crashes EI-Basyouny and Sayed (27) collected data from 392 arterials in Vancouver, British Columbia, and found segment length as a significant factor. Wang et al. (91) extracted GPS data from taxis operating on Shanghai’s urban roadways. They observed that segment length is positively related to crash frequency. This finding is in line with Greibe’s (20) findings that longer segment lengths are associated with higher crash frequencies. Segment Limit Relationship with Operating Speed The roadway features that define the end limits of a segment along with the length of the segment can obviously influence operating speed. For example, a signalized intersection can cause a vehicle to need to either accelerate from a stop or decelerate to a stop. Selecting speed measurement points away from such influences would allow a study to focus on other roadway features; however, in an urban environment, sufficient distances between influential end features may not be common. In the Thiessen et al. (53) study of 249 tangent segments in Edmonton, an increase in speed was found with an increase in segment length. Intersection Angle Relationship with Crashes Dong et al. (30) found that intersection angle is associated with car, car-truck, and truck crash frequencies.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 26 Intersection Lighting Relationship with Crashes Intersection lighting has a statistically significant impact on the crash count for all crash types (30). As expected, intersection lighting appears to significantly decrease car, truck, and car-truck crash frequencies. On-Street Parking Relationship with Operating Speed A study by Wang et al. (77) that considered 35 tangent corridors in Atlanta, Georgia, and a study by Eluru et al. (54) that considered 49 local streets found the presence of on-street parking to be negatively associated with operating speed. Nissan et al.’s (92) study concluded that the frequency of parking maneuvers along the roads significantly reduces moving vehicles’ speed. These results are in line with other studies (93). SURROUNDINGS/ROADSIDE Access Density (Driveways and Intersections) Relationship with Crashes Ferreira and Couto (29) found that the density of minor intersections is associated with crash counts. Driveway density is positively correlated with crash frequencies (95). EI-Basyouny and Sayed (27) found that unsignalized intersection density is positively correlated with crash frequency. This result is in line with the findings from Wang et al. (91). Urban interchange ramp density has a positive influence on the crash frequency (94). Access Density (Driveways and Intersections) Relationship with Operating Speed Several measures have been used to represent the presence or quantity of interruptions within a suburban or urban street, including the presence of an intersection or signalized intersection within the segment or the number of driveways and/or intersections, sometimes expressed as density (number of access points within a mile). A study by Wang et al. (77) found driveway density and T-intersection density as being negatively associated with operating speed. The study by Eluru et al. (54) found operating speed decreased as access increased. By using data from Shanghai’s 50,000+ taxis equipped with GPSs, Wang et al. (95) found that speed variation would increase as access points increased in the urban arterials. School/School Zone Relationship with Crashes The presence of a school zone requires a driver’s careful attention in speed reduction and cautious driving. Many studies examined different factors associated with school-zone-related crashes. The dominant factors were types of school zones (school zone compared to playground) (44, 96), types of schools (97), number of lanes (44, 96), and length of speed zone (96). School/School Zone Relationship with Operating Speed School zones are used to reduce operating speeds during select times of the day when students are moving between home and school. Several studies have investigated the relationship between the school zone speed limit and operating speed (98, 99, 100, 101, 102). While speeds are typically lower during an active school zone, drivers are still not in full compliance with the speed limit.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 27 Parking Relationship with Crashes Greibe (20) found parking as a statistically significant factor. Islam and El-Basyouny (36) observed that the presence of street parking was associated with an increase in property damage only (PDO) crashes but demonstrated a statistically insignificant association with severe crashes. Since street parking leaves fewer driving spaces on the road, there is a high likelihood of crash involvement. Few studies (103, 104) found association between parking and crashes associated with older pedestrians and bicyclists. Liquor Store Relationship with Crashes Islam and El-Basyouny (36) found that the presence of licensed liquor stores was associated with increases in severe and PDO crashes. This result is intuitive since the percentage of impaired driving is expected to be higher near licensed premises. Sidewalk Presence Relationship with Operating Speed Wang et al. (77) found that the presence of sidewalks was negatively associated with operating speed. The study by Eluru et al. (54) that considered 49 Montreal local streets found that the number of sidewalks (coded as 0, 1, or 2) tended to increase vehicle speed; however, they cautioned that the variable had a large standard deviation. In the Thiessen et al. (53) study, sidewalks were categorized by whether they were connected to the curb and gutter (called monolithic walk) or if there was a space between the curb and sidewalk (called boulevard walk). They found that roads with sidewalks that were farther away from the road (boulevard) were associated with higher operating speed, while locations with monolithic walks on both sides of the road had lower operating speeds. Other studies have shown that vulnerable road users along the road or crossing the road significantly influence vehicle speed and road capacity (105, 106). Roadside Hazard Rating or Pole/Tree Density Relationship with Operating Speed The study by Poe and Mason (67) found roadside hazard rating to be one of the significant variables for tangents and within horizontal curves. A study by Wang et al. (77) found that roadside density was negatively associated with operating speed. Roadside density was defined as the density of trees and utility poles (number/mile) divided by their average offset from the roadway. The study by Eluru et al. (54) found that roads with low object density and/or tree density were associated with higher operating speed. Development (Surrounding Land Use) Relationship with Operating Speed In the 2001 Fitzpatrick et al. (48) study, when posted speed was not included in the model, median presence and roadside development were significant variables for curve sections. The categories used within roadside development included residential, commercial, park, or school. A study by Wang et al. (77) found that commercial and residential lane use was positively associated with operating speed. The study by Eluru et al. (54) found the type of development to be significant (categorized as being either downtown commercial, mixed high to medium density, mixed low density, and open urban). Mixed low-density areas had the highest operating speed, while the mixed high to medium density had the lowest operating speeds. Wilmot and Khanal (107) mentioned that land use plays a role in driver perception of a safe speed limit at a certain location. On arterial roads, locations of direct control land use recorded the highest compliance rates with a relative drop in probability in both commercial and

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 28 agricultural areas by 39 percent and 94 percent, respectively, in the study of Gargoum et al. (108). It is worth noting that Marshall et al. (93) and Lu et al. (76) also observed significance of land use in operating speed. OTHER VARIABLES One-Way or Two-Way Relationship with Operating Speed In the Thiessen et al. (53) study of 249 tangent segments in Edmonton, one-way roads had lower operating speeds when compared to two-way roads. This finding agrees with a study that used 49 local street segments in Montreal by Eluru et al. (54). Functional Classification Relationship with Operating Speed Thiessen et al. (53) subdivided their data into collectors and arterials. They identified different variables as being significant between the two functional classes. For example, the presence of a pedestrian crossing was significant in the collector model but not in the arterial model. They also found that some of the variables had an opposite effect on arterials compared to collectors. For example, the presence of bus stops was associated with higher speeds on arterials and lower speeds on collectors. Weather Relationship with Operating Speed Wilmot and Khanal (107) found that weather conditions play a role in the driver’s perception of a safe speed limit at a certain location. Giles (38) observed the effects of weather condition and found that regardless of the posted speed limit, inclement weather does have an effect on driver speed choice. Gargoum et al. (108) found that dry and fine weather encouraged significantly higher speeds compared to wet and cloudy conditions. Although visibility is weather related, Gargoum et al. (108), similar to the findings of Goldenbeld and Van Schagen (109), concluded that a driver’s perceived safe speed limit is higher on good visibility roads where vegetation or trees are less or absent. Romancyshyn et al. (110) showed snow as a hindrance to travel speed on urban arterials. Rising temperature was found to slightly increase travel speeds as it increased. Time of the Day Relationship with Operating Speed One study by Giles (38) observed the effects of the time of day and determined that regardless of the PSL, this factor does have an effect on driver speed choice. The effect of time of day on speed limit violations was also assessed by Nouvier (111) using data from France. The study found that non-compliance was more common in the early morning compared to midday. Similarly, ONISR (112) also found that more violations were observed at night compared to daytime. For local roads, Eluru et al. (54) showed that the only temporal variable that was significant in the analysis for local roads was the late-night hours (12 a.m.–6 a.m.) indicator variable. The variable indicates that vehicle speed during these time periods is usually lower than other time periods. Drivers were most likely careful during these times on local roads. Islam and El-Basyouny (36) showed that the parameter for the time-of-day indicator was found to be negative and therefore indicated that nighttime hour was associated with a higher free-flow speed than daytime hour. Bassani et al. (113) showed that average speeds and deviations from the

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 29 average speed are significantly affected by changes in lighting parameters for the different weather condition and time of the day. OVERVIEW OF VARIABLE RELATIONSHIP WITH SPEED AND CRASHES Based upon information in the literature, several roadway segment characteristics are known or suspected to affect a driver’s speed choice and the likelihood of crashes. Table 1 summarizes the factors that affect crashes on urban/suburban city streets. Table 2 summarizes the characteristics that affect operating speed on urban/suburban city streets. In general, the following relationships were found:  Operating speeds decrease as the access density increases.  Operating speeds decrease as the roadside becomes fuller with objects.  Operating speeds are higher with higher posted speed limits.  Operating speeds are lower on horizontal curves with small radii or larger deflection angle. Table 1. Factors of urban/suburban city streets that affect crashes. Category Factor Key Findings Source Traffic control device Posted speed limit Mixed effect 14, 20, 36, 39, 40, 41, 42, 43, 44, 45, 46, 47, 69 Traffic control device Presence of signalized intersection Positively associated 29, 55, 56, 57, 36 Traffic AADT Positively associated 10, 19, 20, 21, 22, 23, 24, 25, 26, 30, 27, 28, 29 Traffic Congestion Positively associated 31, 32 Traffic Operating speed Mixed effect 10 Traffic Percent trucks Negatively associated with crash frequencies, positively associated with car-truck crash frequencies 34 Surroundings Access density Positively associated 63, 27, 91, 29, 94, 89 Surroundings Liquor store Positively associated 36 Surroundings Parking Positively associated 20, 36 Surroundings Schools Negatively associated 96, 97, 44 Geometry Bike lanes Negatively associated 83, 84, 85, 87, 88, 89, 90 Geometry Horizontal alignment Mixed effect 10, 58, 59, 60, 61, 62, 63, 64 Geometry Intersection angle Positively associated 30 Geometry Intersection lighting Negatively associated 30 Geometry Lane width Mixed effect 30, 69, 80, 81, 82 Geometry Median See references 10, 69 Geometry Median width Positively associated 30, 75 Geometry Number of lanes Positively associated 27, 28 Geometry Segment length Positively associated 10, 20, 27, 91 Geometry Shoulder width Negatively associated 62 Geometry Vertical alignment See reference 69

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 30 Table 2. Factors of urban/suburban streets that affect operating speed. Category Factor Key Findings Source Traffic control device Posted speed limit Positively associated 10, 48, 50, 51 52, 53, 54 Traffic Percent trucks Positively associated 36 Surroundings Access density Negatively associated 54, 77 Surroundings Development Negatively associated 48, 54, 77, 76, 93, 107, 108 Surroundings Hazard ratings Negatively associated 54, 67, 77 Surroundings Presence of sidewalks Negatively associated 53, 54, 77, 105, 106 Surroundings Schools/school zones Negatively associated 98, 100, 101, 102 Geometry Horizontal or vertical alignment Mixed 54, 65, 66, 67, 68, 39 Geometry Lane width Positively associated 48, 67, 68, 36 Geometry Median 48, 74 Geometry Median width Positively associated 54, 76 Geometry Number of lanes Positively associated 77, 78, 79, 76 Geometry Parking Negatively associated 54, 77, 92, 93 Geometry Presence of bike routes See references 53, 54 Geometry Segment limits Positively associated 53 Geometry Shoulder or curb See reference 77 Other factors Functional classification See reference 53 Other factors One way or two way One-way roads had lower operating speeds compared to two-way roads 53, 54 Other factors Time of the day Positively associated in nighttime 54, 36, 38, 111, 112 Other factors Vehicle characteristics Positively associated (new, heavy and long vehicles) with operating speed 39, 37, 38 Other factors Weather Negatively associated 107, 108, 38, 109

Next: APPENDIX B. RELATIONSHIP AMONG ROADWAY CHARACTERISTICS, SPEED, AND SAFETY FOR HIGH-SPEED HIGHWAYS »
Development of a Posted Speed Limit Setting Procedure and Tool Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Several types of speed limits exist, including statutory speed limit, posted speed limit, school zone speed limit, work zone speed limit, variable speed limit, and advisory speed.

The TRB National Cooperative Highway Research Program's NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool documents the research efforts and findings from an NCHRP Project 17-76 to identify factors that influence a driver’s operating speed and the development of a Speed Limit Setting Procedure and Tool.

The document is supplemental to NCHRP Research Report 966: Posted Speed Limit Setting Procedure and Tool: User Guide.

  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!