National Academies Press: OpenBook

Evaluation of the 13 Controlling Criteria for Geometric Design (2014)

Chapter: Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria

« Previous: Section 1 - Introduction
Page 5
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 5
Page 6
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 6
Page 7
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 7
Page 8
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 8
Page 9
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 9
Page 10
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 10
Page 11
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 11
Page 12
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 12
Page 13
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 13
Page 14
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 14
Page 15
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 15
Page 16
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 16
Page 17
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 17
Page 18
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 18
Page 19
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 19
Page 20
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 20
Page 21
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 21
Page 22
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 22
Page 23
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 23
Page 24
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 24
Page 25
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 25
Page 26
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 26
Page 27
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 27
Page 28
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 28
Page 29
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 29
Page 30
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 30
Page 31
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 31
Page 32
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 32
Page 33
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 33
Page 34
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 34
Page 35
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 35
Page 36
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 36
Page 37
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 37
Page 38
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 38
Page 39
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 39
Page 40
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 40
Page 41
Suggested Citation:"Section 2 - Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria." National Academies of Sciences, Engineering, and Medicine. 2014. Evaluation of the 13 Controlling Criteria for Geometric Design. Washington, DC: The National Academies Press. doi: 10.17226/22291.
×
Page 41

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.

5 S E C T I O N 2 This section presents the results of the review of design criteria, traffic operational and safety effects, and mitiga- tion strategies for the 13 controlling criteria. This informa- tion concerning each of the controlling criteria is presented in Sections 2.1 through 2.13. The information presented in Section 2 is based primarily on published documentation. The primary sources consulted for each of the 13 controlling criteria are as follows: • Design criteria are based primarily on the 2004 and 2011 editions of the AASHTO Green Book (4, 5), unless explicitly stated otherwise. Design criteria for freeways on the Interstate highway system are also presented in AASHTO’s A Policy on Design Standards—Interstate Sys- tem (14). Published FHWA guidance on the scope and interpretation of the 13 controlling criteria is also pre- sented (7). • Traffic operational effects are based primarily on the 2010 TRB Highway Capacity Manual (HCM) (13). • Traffic safety effects are based primarily on the 2010 AASHTO Highway Safety Manual (HSM) (12). • Mitigation strategies are based primarily on the FHWA guidance presented in Mitigation Strategies for Design Exceptions (7) and AASHTO’s A Guide for Achieving Flex- ibility in Highway Design (8). In addition, the discussion of the traffic operational and safety effects of the individual design criteria includes all relevant findings of the research conducted in this project, as reported in Section 4. Separate discussions of design criteria, traffic oper- ational effects, and traffic safety effects are presented, where appropriate, for each of four roadway types: rural two-lane highways; rural multilane highways (nonfreeways); urban and suburban arterials (nonfreeways); and freeways. Through- out this report, the term “freeways” applies to both rural and urban freeways except where the terms “rural freeway” or “urban freeway” are used explicitly. Design Criteria, Traffic Operational and Safety Effects, and Mitigation Strategies for the 13 Controlling Criteria In cases where the primary sources present no informa- tion or only limited information on the traffic operational or safety effects of a particular issue, or where there may be concerns about the completeness of the primary sources, results of additional relevant research are presented. For safety effects, many such sources are cited in the FHWA Crash Modification Factors Clearinghouse (CMF Clearinghouse) website (15), which includes star ratings to assess the qual- ity of the studies cited. The ratings range from one star (the weakest research) to five stars (the strongest research). Only CMFs included in the HSM or rated three stars or better in the FHWA CMF Clearinghouse website are cited in this sec- tion of the report. Table 1 shows with circular bullets which of the 13 control- ling criteria have documented traffic operational and safety effects for each of four roadway types (rural two-lane high- ways, rural multilane highways, urban and suburban arterials, and freeways). These documented traffic operational and safety effects are presented in Sections 2.1 through 2.13. The traffic operational effects of the 13 controlling criteria are summarized in Section 2.14. The traffic safety effects of the 13 controlling criteria are summarized in Section 2.15. The traffic operational and safety effects include findings from published literature and from research conducted as part of NCHRP Project 17-53, which are reported in Section 4. 2.1 Design Speed AASHTO defines design speed as (4): Design speed is a selected speed used to determine the various geometric features of the roadway. The assumed design speed should be a logical one with respect to the topography, antici- pated operating speed, the adjacent land use, and the functional classification of the highway. Design speed is unique among the 13 controlling criteria since it has no direct effect on the design of the roadway, but

6only an indirect effect. Once a design speed for a project is selected, however, that design speed influences the values (or value ranges) of other controlling criteria, including hori- zontal alignment, vertical alignment, stopping sight distance, and lane width. Thus, design speed actually serves as a design control rather than a design criterion. Design speeds should reflect the speeds that drivers expect to travel, which is determined by the physical limitations of the roadway and surrounding traffic rather than by the functional class of the roadway. Specific recommendations for design speeds are provided in several exhibits in the Green Book and are based on roadway classification, type of terrain, and volume. Ranges are as follows: • For local rural roads, design speeds range from 20 mph for low-volume roads in mountainous terrain to 50 mph on high-volume roads in level terrain. • For rural arterials, the recommended design speed ranges from 40 to 75 mph based on terrain, driver expectancy, and alignment. • For urban arterials, the design speed should fall between 30 and 60 mph. In more developed areas, such as central busi- ness districts, the lower end of that range should be used, while in suburban or developing areas, the higher end of the range may be appropriate. • For urban freeways, a design speed in the range of 50 to 70 mph should be used with higher speeds being more desirable when alignment and interchange spacing permit. Where lower design speeds are used, speed enforcement may also be needed. For rural freeways, a 70 mph design speed is recommended. Lower design speeds that are con- sistent with driver expectations are appropriate in moun- tainous terrain. Table 2 summarizes the Green Book guidance on design speed. Another aspect of design speed also serves as part of the controlling criteria. Green Book Exhibit 10-56 provides guide values for selection of ramp design speeds as a function of the highway design speed. According to the Green Book, ramp Traffic operational effects Traffic safety effects Design criteria Rural two- lane highways Rural multilane highways Urban and suburban arterials Freeways Rural two-lane highways Rural multilane highways Urban and suburban arterials Freeways Design speed a a a a a a a a Lane width b Shoulder width Bridge width b Structural capacity N/A N/A N/A N/A N/A N/A N/A N/A Horizontal alignment b b b Vertical alignment (sag vertical curve length) Grade Stopping sight distance b Cross slope Superelevation Vertical clearance N/A N/A N/A N/A N/A N/A N/A N/A Horizontal clearance (lateral offset) c c c d d d a There are no direct operational or safety effects of design speed; however, design speed may influence operations and safety indirectly through the criteria for lane width, horizontal alignment, vertical alignment, and stopping sight distance. b New relationships were developed in this research. c No effect anticipated when full shoulders are present. d There are no known direct effects of lateral offset on safety; however, the influence of lateral offset on safety is known indirectly through the influence of shoulder width. • • • • • • • • • • • • • • • • • • • • • • Table 1. Summary table for operational and safety effects of the controlling criteria. Table 2. Ranges for design speed by roadway functional class (4, 7). Roadway functional classification Terrain Design speed (mph) Rural Urban Freeway Level 70 50 min Rolling 70 50 min Mountainous 50 to 60 50 min Arterial Level 60 to 75 30 to 60 Rolling 50 to 60 30 to 60 Mountainous 40 to 50 30 to 60 Collector Level 40 to 60 30+ Rolling 30 to 50 30+ Mountainous 20 to 40 30+ Local Level 30 to 50 20 to 30 Rolling 20 to 40 20 to 30 Mountainous 20 to 30 20 to 30

7 design speeds should not be less than the low range presented in Exhibit 10-56, with other specific guidance offered for particular types of ramps (loops as well as direct and semi- direct connections). Some states have adopted design policies requiring the use of middle or higher range values for certain cases, such as system interchanges. Designers are occasionally confronted with situations in which the appropriate ramp design speed shown in Green Book Exhibit 10-56 may not be achievable. Such cases are almost always associated with the inability to achieve minimum radius for the controlling curvature of the exit or entrance ramp. Not meeting the lower (50 percent) range shown in Green Book Exhibit 10-56 requires a design exception per FHWA policy. Where the design issue involves curvature, a design excep- tion should be prepared for the non-standard horizontal curve rather than for the use of a lower design speed for the ramp (7). There are no explicit traffic operational effects of design speed. Any traffic operational and safety effects of design speed result from the other design elements that are influenced by design speed. Experience shows that vehicle speeds cannot be reduced merely by reducing the posted speed limit or the design speed. Adjustment of a broad range of design and roadway environment factors is needed to influence vehicle speeds. In accordance with Manual on Uniform Traffic Control Devices for Streets and Highways (MUTCD) criteria (16), posted speed limits are typically set to approximate the 85th percentile speed of traffic, on the assumption that most driv- ers select speeds that are reasonable for conditions. Design speed, posted speed, and the roadway environment should all send a clear and consistent message to drivers about the appropriate speed for the roadway. A 2009 paper by Hauer (17) documents the current state of knowledge about the relationship between highway travel speed and safety. Hauer indicates that vehicle travel speeds are affected by the roadway design, speed limits and enforce- ment, traffic controls, and many other factors. The travel speeds that are chosen by drivers affect the safety perfor- mance of the roadway. Although higher speeds will tend to increase the severity of crashes, Hauer states that there is little evidence to support the notion that faster travel speeds necessarily result in a greater likelihood of a crash. However, since higher speeds increase crash severity, higher speeds may increase the likelihood of a reported crash. Hauer also indi- cates that travel speeds on roadways tend to change over time, and, although this fact is well documented, little is known about why these changes occur. As indicated by the design speed ranges shown in Table 2, the AASHTO Green Book provides substantial flexibility in the choice of an appropriate design speed. As written, AASHTO policy presents little need for design exceptions, because the choice of a design speed is left to the discretion of the designer. FHWA’s report, Mitigation Strategies for Design Exceptions (7), states that the selected design speed should be high enough that an appropriate regulatory speed limit will be less than or equal to it, but this is not a formal FHWA policy. Mitigation strategies for design speed would typically involve revision of both design elements and the roadway environment to encourage lower vehicle speeds. The FHWA Interactive Highway Safety Design Model (IHSDM) includes a design consistency tool that can be used to evaluate mitiga- tion strategies for design speed (10). However, the IHSDM design consistency tool is currently applicable only to rural two-lane highways. In actual practice, as documented in Section 3 of this report, design exceptions for design speed appear to be seldom requested or approved by highway agencies. Highway agen- cies generally seek design exceptions for specific design ele- ments that do not meet the criteria for the selected design speed rather than seeking a blanket exception to reduce the design speed. The rare exception is where a highway agency may deem it appropriate to utilize a lower design speed for an entire corridor (or a substantial segment of a corridor) due to topographic or environmental constraints. 2.2 Lane Width Lane width determines the area where a vehicle can maneu- ver laterally without encroaching into the path of another vehicle or onto the shoulder. Table 3 summarizes the lane width design criteria in the AASHTO Green Book. Separate criteria have also been established for auxiliary lanes, includ- ing turn lanes at intersections and center two-way left-turn lanes. Formal design exceptions for lane width are required by FHWA policy for all travel lanes including auxiliary lanes and ramps that do not meet Green Book criteria. Some high- way agencies have lane width policies that provide less flex- ibility than the Green Book (e.g., specifying the use of 12-ft lanes in nearly all cases). This approach is not required by FHWA policy and may result in more design exceptions than FHWA policy would require. The AASHTO Green Book also includes criteria for lane widening on horizontal curves to Functional class Lane width (ft) Rural Urban Freeway 12 12 Ramps (one-lane) 12 to 30a 12 to 30a Arterial 11 to 12 10 to 12 Collector 10 to 12 10 to 12 Local 9 to 12 9 to 12 a For wider ramps, some of the specified width may be provided by shoulders. Table 3. Ranges for lane width by roadway functional class (4, 5, 7).

8accommodate truck offtracking; a formal design exception is not required where lane widening is not provided on a hori- zontal curve (7). 2.2.1 Rural Two-Lane Highways Design Criteria Chapter 7 (Arterials) of the Green Book provides the fol- lowing guidance for the design of lane widths on rural arteri- als. The Green Book recommends the lane widths shown in Table 4 on rural arterials as a function of design speed and design volume (expressed as an average daily traffic volume, or ADT). Where lane widths narrower than those shown in Table 4 are used, a design exception is required by FHWA pol- icy. In the case that is described in Note a of Table 4, a design exception is not required, although the justification for use of 11-ft lanes should be documented in the project files (7). Traffic Operational Effects Chapter 15 (Two-Lane Highways) of the HCM provides the estimates shown in Table 5 for reduction in free-flow speed on two-lane highways with lane widths less than 12 ft or shoulder widths less than 6 ft. The values in Table 5 are used to estimate the actual free- flow speed of traffic on a two-lane highway from the free-flow speed for base conditions, as follows: FFS BFFS f f (1)LS A= − − where FFS = free-flow speed (mph) BFFS = base free-flow speed (mph) fLS = adjustment for lane shoulder width (mph) from Table 5 fA = adjustment for access-point density (mph) from HCM Exhibit 15-8 FFS may also be estimated directly from field data. FFS is used in estimating the average travel speed (ATSd), one of the ser- vice measures used to determine level of service (LOS) for two-lane highways. The shoulder-width effects included in fLS are discussed in Section 2.3.1 of this report. Traffic Safety Effects Chapter 10 (Rural Two-Lane Highways) of the HSM pro- vides CMFs for lane widths on rural two-lane highways. The Minimum width of traveled way (ft)a for specified design volume Design speed (mph) Under 400 (veh/day) 400 to 1,500 (veh/day) 1,500 to 2,000 (veh/day) Over 2,000 (veh/day) 40 22 22 22 24 45 22 22 22 24 50 22 22 24 24 55 22 22 24 24 60 24 24 24 24 65 24 24 24 24 70 24 24 24 24 75 24 24 24 24 a On roadways to be reconstructed, an existing 22-ft traveled way may be retained where alignment is satisfactory and there is no crash pattern suggesting the need for widening. SOURCE: Based on Green Book Table 7-3 (abridged). Table 4. Minimum width of traveled way for rural arterials (4, 5). Lane width (ft) Reduction in free-flow speed (mph) Shoulder width (ft) ≤ 0 < 2 ≤ 2 < 4 ≤ 4 < 6 ≥ 6 ≥ 9 < 10 6.4 4.8 3.5 2.2 ≥ 10 < 11 5.3 3.7 2.4 1.1 ≥ 11 < 12 4.7 3.0 1.7 0.4 ≥ 12 4.2 2.6 1.3 0.0 NOTE: The values in Table 5 are used as fLS in Equation 1. SOURCE: Based on HCM Exhibit 15-7. Table 5. HCM adjustment to free-flow speed for lane and shoulder width on two-lane highways (13).

9 CMF is calculated using the equations shown in Table 6 based on the lane width and the average annual daily traffic (AADT). A 12-ft lane is considered to be the base condition (CMF = 1.0). The lane-width CMF is illustrated graphically in Figure 1. The lane-width CMF illustrated in Table 6 and Figure 1 applies only to single-vehicle run-off-the-road crashes and multiple-vehicle head-on, opposite-direction sideswipe, and same-direction sideswipe crashes. The following equation can be used to adjust the lane-width CMF in Table 6 and Figure 1 to CMFs applicable to total crashes: CMF CMF 1.0 p 1.0 (2)ra ra( )= − × + where CMFra = CMF for the effect of lane width on related crashes (i.e., single-vehicle run-off-the-road crashes and multiple-vehicle head-on, opposite-direction side- swipe, and same-direction sideswipe crashes), such as the CMF for lane width shown in Table 6 pra = proportion of total crashes constituted by crash types related to lane and shoulder width The proportion of related crashes, pra, (i.e., single-vehicle run-off-the-road crashes and multiple-vehicle head-on, opposite-direction sideswipe, and same-direction sideswipe crashes) is estimated as 0.574 (i.e., 57.4 percent) based on Table 6. CMF for lane width on rural two-lane roadway segments (12, 18, 19). Lane width Average annual daily traffic (AADT) (veh/day) < 400 400 to 2000 > 2000 9 ft or less 1.05 1.05 + 2.81 x 10-4(AADT − 400) 1.50 10 ft 1.02 1.02 + 1.75 x 10-4(AADT − 400) 1.30 11 ft 1.01 1.01 + 2.5 x 10-5(AADT − 400) 1.05 12 ft or more 1.00 1.00 1.00 NOTE: The collision types related to lane width to which these CMFs apply are single- vehicle run-off-the-road crashes and multiple-vehicle head-on, opposite-direction sideswipe, and same-direction sideswipe crashes. Standard error of the CMF is unknown. To determine the CMF for changing lane width and/or AADT, divide the “new” condition CMF by the “existing” condition CMF. SOURCE: Based on HSM Table 10-8. SOURCE: Based on HSM Figure 10-7. Figure 1. CMF for lane width on rural two-lane roadway segments (12).

10 the default distribution of crash types presented in HSM Table 10-4. This default crash type distribution and, there- fore, the value of pra may be updated from local data as part of the calibration process. It should be noted that the CMFs for 11- and 12-ft lanes are not very different, which is consistent with both 11- and 12-ft lanes being shown as appropriate over broad ranges of conditions in Table 4. 2.2.2 Rural Multilane Highways Design Criteria Table 4 applies to rural multilane arterials as well as to rural two-lane arterials. Where lane widths narrower than those shown in Table 4 are used, a design exception is required by FHWA policy. In the case that is described in Note a of Table 4, a design exception is not required, although the jus- tification for use of 11-ft lanes should be documented in the project files (7). Traffic Operational Effects Chapter 14 (Multilane Highways) of the HCM provides the estimated reduction in free-flow speed for rural and sub- urban multilane highways based on lane width as shown in Table 7. The values in Table 7 are used to estimate the actual FFS of traffic on a multilane highway from the BFFS, as follows: FFS BFFS f f f f (3)LW LC M A= − − − − where fLW = adjustment for lane width (mph) from Table 7 fLC = adjustment for total lateral clearance (mph) from HCM Exhibit 14-9 fM = adjustment for median type (mph) from HCM Exhibit 14-10 fA = adjustment for access-point density (mph) from HCM Exhibit 14-11 FFS may also be estimated directly from field base. FFS is used to determine the mean speed of traffic(s) using the rela- tionships show in HCM Exhibits 14-2 and 14-3 and the traffic density (D) using HCM Equation 14-5. Density is the service measure used to determine LOS for multilane highways. Traffic Safety Effects Chapter 11 (Rural Multilane Highways) of the HSM pre- sents CMFs for lane widths on rural multilane roadways. The CMFs are calculated differently for undivided sections and divided sections, as shown in Tables 8 and 9. The calculation in either case is based on lane width and AADT. These CMFs are illustrated in Figures 2 and 3, respectively. The CMFs shown in Tables 8 and 9 and Figures 2 and 3 are applicable to single-vehicle run-off-the-road crashes, multiple-vehicle head-on crashes, opposite-direction side- swipe crashes, and same-direction sideswipe crashes. Equa- tion 2 can be used to convert these CMFs to CMFs for total crashes. The default value of pra in Equation 2 is 0.27 for rural multilane undivided highways and 0.50 for rural multilane divided highways. Lane width (ft) Reduction in free-flow speed (mph) ≥ 12 0.0 ≥ 11 1.9 ≥ 10 6.6 NOTE: The values in Table 7 are used as fLW in Equation 3. SOURCE: Based on HCM Exhibit 14-8. Table 7. HCM adjustment to free- flow speed for average lane width on rural and suburban multilane highways (13). Table 8. CMF for lane width on undivided rural multilane roadway segments (12, 20). Lane width Average annual daily traffic (AADT) (veh/day) < 400 400 to 2000 > 2000 9 ft or less 1.04 1.04 + 2.13 x 10-4(AADT − 400) 1.38 10 ft 1.02 1.02 + 1.31 x 10-4(AADT − 400) 1.23 11 ft 1.01 1.01 + 1.88 x 10-5(AADT − 400) 1.04 12 ft or more 1.00 1.00 1.00 NOTE: The collision types related to lane width to which these CMFs apply are single- vehicle run-off-the-road crashes and multiple-vehicle head-on, opposite-direction sideswipe, and same-direction sideswipe crashes. Standard error of the CMF is unknown. To determine the CMF for changing lane width and/or AADT, divide the “new” condition CMF by the “existing” condition CMF. SOURCE: Based on HSM Table 11-11.

11 2.2.3 Urban and Suburban Arterials Design Criteria AASHTO policy provides substantial flexibility in the use of 10- to 12-ft lanes on urban arterials. In particu- lar, Chapter 7 of the Green Book includes the following guidance: • Lane widths of 12 ft are most desirable and should be used, where practical, on higher speed, free-flowing, principal arterials. • Lane widths of 11 ft are used quite extensively for urban arterial street designs. Under interrupted-flow operating conditions at low speeds (45 mph or less), narrower lane widths are normally adequate and have some advantages. For example, narrower lane widths allow more lanes to be provided in some areas with restricted right-of-way and allow shorter pedestrian crossing times because of reduced crossing distances. Arterials with 11-ft lane widths are also more economical to construct. An 11-ft lane width is adequate for through lanes, continuous two-way left-turn lanes, and lanes adjacent to a painted median. Table 9. CMF for lane width on divided rural multilane roadway segment (12, 20). Lane width Average annual daily traffic (AADT) (vehicles/day) < 400 400 to 2000 > 2000 9 ft or less 1.03 1.03 + 1.381 x 10-4(AADT − 400) 1.25 10 ft 1.01 1.01 + 8.75 x 10-4(AADT − 400) 1.15 11 ft 1.01 1.01 + 1.25 x 10-5(AADT − 400) 1.03 12 ft or more 1.00 1.00 1.00 NOTE: The collision types related to lane width to which these CMFs apply are single- vehicle run-off-the-road crashes and multiple-vehicle head-on, opposite-direction sideswipe, and same-direction sideswipe crashes. Standard error of the CMF is unknown. To determine the CMF for changing lane width and/or AADT, divide the “new” condition CMF by the “existing” condition CMF. SOURCE: Based on HSM Table 11-16. SOURCE: Based on HSM Figure 11-8. Figure 2. CMF for lane width on undivided segments on rural multilane highways (12, 20).

12 • Lane widths of 10 ft may be used in highly restricted areas having little or no truck traffic. Left-turn and combination lanes used for parking during off-peak hours and for traffic during peak hours may be 10 ft in width. The Green Book also makes reference to the AASHTO bicycle guide (21) because use of narrow lane widths may be critical at many locations in reconstruction of existing arterials to provide space for bicycle facilities. Traffic Operational Effects Chapter 17 (Urban Street Segments) of the HCM includes a procedure to determine the effect of the features of an urban street segment on free-flow speed. However, lane width is not one of the factors that influences free-flow speed. This suggests that lane width either has no effect on the free-flow speed of an urban street segment or an effect that is very small in comparison to the factors that are in the procedure (see HCM Exhibit 17-5). This zero or negligible effect for lane width in the current HCM contrasts with the HCM 2000 (22), which speculated that lane width influenced free-flow speed for urban streets, but did not quantify that effect. The HCM adjustment for lane width presented in Table 6 is applicable to suburban multilane highways, but not to urban streets. Recent research by Potts et al. (23, 24) investigated the effect of lane width on midblock vehicle speeds on urban and suburban arterials based on spot speed measurements at pairs of sites upstream and downstream of lane width transi- tions. The research of Potts et al. (23, 24) found that mean speeds at sites with wider lanes (ranging from 11.9 to 13.3 ft) were approximately 4 mph higher than mean speeds at sites with narrower lanes (ranging from 9.4 to 10.3 ft in width). This finding suggested that lane width has an effect on traffic operations. However, the sample size in the study was rela- tively small (five pairs of wide- and narrow-lane sites) and was not sufficient to develop a formal relationship between lane width and traffic speed. A similar evaluation in the NCHRP Project 17-53 research considered a total of 23 additional sites on urban and sub- urban arterials in the Eastern, Midwest, and Western regions of the United States (see Section 4.1). This evaluation found that lane width had no effect on traffic speeds on urban and suburban arterials. Based on this finding, it appears that the HCM is correct in assuming that lane width has no effect on traffic speeds on urban and suburban arterials. Chapter 18 (signalized intersections) of the HCM includes an adjustment factor for the effect of lane width on saturation flow rate at signalized intersections (see HCM Exhibit 18-3). However, given that this adjustment is applicable only to SOURCE: Based on HSM Figure 11-10. Figure 3. CMF for lane width on divided roadway segments on rural multilane highways (12, 20).

13 signalized intersection approaches and not to midblock sec- tions of arterials, it is not presented in this report, since inter- section design criteria are outside the scope of the research. Traffic Safety Effects Chapter 12 (Urban and Suburban Arterials) of the HSM does not include a CMF for lane width on urban and sub- urban arterials. Recent research by Potts et al. (23, 24) under NCHRP Project 03-72 found no difference in safety perfor- mance for urban and suburban arterials in lane widths rang- ing from 10 to 12 ft, with only limited exceptions that could represent random effects. Lanes narrower than 12 ft may be a design concern on streets with substantial volumes of bi- cycles, trucks, and buses. 2.2.4 Freeways Design Criteria According to the Green Book, freeway lanes should be 12 ft wide. Lane widths of 12 ft are also called for in the AASHTO design standards for the Interstate highway system. Traffic Operational Effects Chapter 11 (Basic Freeway Segments) of the HCM pre- sents the estimated reduction in free-flow speed for freeways with lane widths less than 12 ft as shown in Table 10. The values in Table 10 are used to estimate the actual free- flow speed of traffic on a freeway from the estimated free- flow speed for base conditions, 75.4 mph. This adjustment is made as follows: FFS 75.4 f f 3.22 TRD (4)LW LC 0.84= − − − where fLW = adjustment for lane width (mph) from Table 10 fLC = adjustment for right-side lateral clearance (mph) from HCM Exhibit 11-9 TRD = total ramp density (ramps/mi) FFS may also be estimated directly from field data. FFS is used to determine the mean speed of traffic (S) using the relationships shown in HCM Exhibits 11-2 and 11-3 and the traffic density (D) using HCM Equation 11-4. Density is the service measure used to determine LOS for freeways. Traffic Safety Effects Results from NCHRP Project 17-45, which developed a proposed HSM safety prediction methodology for freeways, include the following CMF for lane width on freeways where We = average lane width for all through lanes (ft) (25): CMF exp 0.0376 W 12 , if W 13 ft (5)e e( )( )= − − < CMF 0.963, if W 13 ft (6)e= ≥ The base condition for this CMF is a 12-ft lane width, (CMF = 1.0). We represents the average lane width for all through lanes on a freeway segment in both directions of travel excluding managed lanes and auxiliary lanes associ- ated with a weaving section. The CMF is applicable to lane widths in the range of 10 to 14 ft. The CMF is intended for application to both multiple- and single-vehicle crashes on rural freeways with four to eight lanes and urban freeways with four to ten lanes. 2.2.5 Mitigation Strategies Mitigation strategies for lane width are most important on higher speed roadways (speeds above 45 mph). On roadways with speeds of 45 mph or less, there are often good reasons for using narrow lanes as a flexibility measure to obtain other benefits: shorter pedestrian crossing distances, inclusion of turn lanes, medians, bicycle lanes, etc. These other benefits for road users, in and of themselves, constitute mitigation for the use of narrower lanes. The best use of available cross- section width should be determined on a case-by-case basis. The mitigation strategies where narrower lanes are used on higher speed facilities include (7): • Provide warning of lane width reduction • Improve ability of drivers to stay within their travel lane through use of enhanced pavement markings, delineations, lighting, shoulder rumble strips, painted edge line rumble strips, and/or centerline rumble strips • Improve ability to recover if driver leaves the lane (paved or partially paved shoulders, safety edge treatment) • Reduce crash severity if the driver leaves the roadway (clear recovery area, traversable slopes, breakaway safety hard- ware, and barriers where appropriate) • Provide pull-off areas where shoulder width is limited Lane width (ft) Reduction in free-flow speed (mph) 12 0.0 11 1.9 10 6.6 NOTE: The values in this table are used as fLW in Equation 4. SOURCE: Based on HCM Exhibit 11-8. Table 10. HCM adjustment to free- flow speed for lane width on freeways (13).

14 2.3 Shoulder Width Shoulder width affects both capacity and safety on road- ways. A wide shoulder increases capacity by reducing lateral friction between traffic and roadside objects and thereby increasing driver comfort. Shoulders can reduce the likeli- hood of crashes in several ways, including providing a loca- tion for emergency stops and broken-down vehicles outside the traveled way, providing a space for drivers of errant vehi- cles to make steering corrections before leaving the road- way, and providing space for evasive maneuvers. Shoulders also provide space for enforcement activities, maintenance activities, and bicycle accommodations. Table 11 summa- rizes the range of minimum shoulder widths for travel lanes and ramps presented in the Green Book. 2.3.1 Rural Two-Lane Highways Design Criteria The shoulder widths presented in Table 12 are recom- mended in the Green Book, as a function of AADT. The usable shoulder-width values in Table 12 require a design exception if they are not met. Usable shoulder width is mea- sured from the edge of the traveled way to the point of inter- section of the shoulder slope and mild slope (for example, 1V:4H or flatter) or to the beginning of rounding to slopes steeper than 1V:4H (7). Traffic Operational Effects Chapter 15 (Two-Lane Highways) of the HCM presents the estimated reductions in free-flow speed for two-lane highways with lane widths less than 12 ft or shoulder widths less than 6 ft, as shown in Table 5. The values shown in Table 5 are used as fLS in Equation 1 to estimate the free-flow speed on two-lane highways (see Section 2.2.1). Traffic Safety Effects Chapter 10 (Rural Two-Lane Highways) of the HSM pro- vides CMFs for paved shoulders on rural two-lane roadways for specific crash types related to lane encroachment. The value of CMFwra for shoulder width is calculated using the equations shown in Table 13 based on the shoulder width and the traffic volume (AADT). A 6-ft shoulder is consid- ered to be the base condition (CMF = 1.0). Wider shoulders have CMFs less than 1.0, and narrower shoulders have CMFs Functional class Shoulder width (ft) Rural Urban Freeway 4 to 12 4 to 12 Ramps (one-lane) 1 to 10 1 to 10 Arterial 2 to 8 2 to 8 Collector 2 to 8 2 to 8 Local 2 to 8 — NOTE: Ranges shown include both right and left shoulder widths for ramps and divided highways. Table 11. Ranges for minimum shoulder width by roadway functional class (4, 5, 7). Minimum width of usable shoulder (ft) for specified design volume Under 400 veh/day 400 to 1,500 veh/day 1,500 to 2,000 veh/day Over 2,000 veh/day 4 6 6 8 NOTE: Usable shoulders on arterials should be paved; however, where volumes are low or a narrow section is needed to reduce construction impacts, the paved shoulder may be reduced to 2 ft. SOURCE: Based on Green Book Table 7-3 (abridged). Table 12. Minimum width of usable shoulder for rural arterials (4, 5). Shoulder width Average annual daily traffic (AADT) (veh/day) < 400 400 to 2000 > 2000 0 ft 1.10 1.10 + 2.5 x 10-4(AADT − 400) 1.50 2 ft 1.07 1.07 + 1.43 x 10-4(AADT − 400) 1.30 4 ft 1.02 1.02 + 8.125 x 10-5(AADT − 400) 1.15 6 ft 1.00 1.00 1.00 8 ft or more 0.98 0.98 - 6.875 x 10-5(AADT − 400) 0.87 NOTE: The collision types related to lane width to which these CMFs apply include single- vehicle run-off-the-road crashes and multiple-vehicle head-on, opposite-direction sideswipe, and same-direction sideswipe crashes. Standard error of the CMF is unknown. To determine the CMF for changing paved shoulder width and/or AADT, divide the “new” condition CMF by the “existing” condition CMF. SOURCE: Based on HSM Table 10-9. The values from Table 13 are used as CMFwra in Equation 7. Table 13. CMFs for shoulder width on rural two-lane roadway segments (CMFwra) (12, 18).

15 SOURCE: Based on HSM Figure 10-8. Figure 4. CMF for shoulder width on roadway segments for two-lane highway (12, 18). Table 14. CMFs for shoulder types and shoulder width on roadway segments (CMFtra) (12, 18). Shoulder type Shoulder width (ft) 0 1 2 3 4 6 8 Paved 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Gravel 1.00 1.00 1.01 1.01 1.01 1.02 1.02 Composite 1.00 1.01 1.02 1.02 1.03 1.04 1.06 Turf 1.00 1.01 1.03 1.04 1.05 1.08 1.11 NOTE: The values for composite shoulders in this table represent a shoulder for which 50 percent of the shoulder width is paved and 50 percent of the shoulder width is turf. SOURCE: Based on HSM Table 10-10. greater than 1.0. The shoulder-width CMF for rural two-lane highways is illustrated in Figure 4. The base condition for shoulder type is paved (CMF = 1.0). Table 14 presents values for CMFtra, which adjusts for the safety effects of gravel, turf, and composite shoulders as a function of shoulder width. A combined CMF for shoulder width and type is com- puted as CMF CMF CMF 1.0 p 1.0 (7)wra tra ra( )= × − × + where CMFwra = crash modification factor for shoulder width from the equations in Table 13 CMFtra = crash modification factor for shoulder type from Table 14 If the shoulder types and/or widths for the two directions of a roadway segment differ, the CMFs are determined sepa- rately for the shoulder type and width in each direction of travel and the resulting CMFs are then averaged. The CMFs for shoulder width and type shown above apply only to the collision types that are most likely to be affected by shoulder width and type: single-vehicle run-off-the-road crashes, multiple-vehicle head-on crashes, opposite-direction sideswipe crashes, and same-direction sideswipe crashes. The CMFs expressed on this basis are, therefore, adjusted to total crashes using Equation 7. The HSM default value for pra for two-lane highways in Equation 7 is 0.574.

16 2.3.2 Rural Multilane Highways Design Criteria The Green Book states that the design criteria for shoulder width on rural two-lane highways presented in Table 12 are generally applicable to rural undivided multilane arterials, as well. For rural divided multilane arterials, the shoulder widths presented in Table 15 are recommended. Traffic Operational Effects Chapter 14 (Multilane Highways) of the HSM estimates free-flow speed based on the total lateral clearance, defined as the sum of the lateral clearance on the left side of the roadway (maximum of 6 ft) and the right side of the roadway (maxi- mum of 6 ft). Lateral clearance is defined as the distance from the edge of the travel lane to the nearest obstruction. Thus, roadways with wide shoulders inherently have larger lateral clearance values than roadways with narrow shoulders. Total lateral clearance for multilane highways is generally inter- preted as equivalent to the sum of the left (inside) and right (outside) shoulder widths, since some objects (e.g., guard- rail) may be located immediately outside the shoulders. The free-flow speed reduction values are shown in Table 16; these values are used in Equation 3 (see Section 2.2.2). In addition, Chapter 14 of the HCM predicts a free-flow speed reduction of 1.6 mph for an undivided roadway relative to a divided highway or a highway with a two-way left-turn lane. This value is used in fM in Equation 3, where applicable. Traffic Safety Effects Chapter 11 (Rural Multilane Highways) of the HSM pre- sents CMFs for paved shoulders on rural multilane roadways. CMFs are calculated differently for undivided and divided roadways. CMFs for undivided sections of multilane high- ways are calculated using the same equations as two-lane highways, as shown in Table 13 (see also Figure 4). The base condition for this CMF is a 6-ft shoulder (CMF = 1.0). As for rural two-lane highways, this CMF is adjusted to total crashes using Equation 7. The HSM default value for pra for rural multilane undivided highways used in Equation 7 is 0.27. CMFs for divided sections of multilane highways are pre- sented in Table 17. The base condition (CMF = 1.0) is an 8-ft shoulder. This CMF applies to total crashes and is not adjusted using a pra value. 2.3.3 Urban and Suburban Arterials Design Criteria Chapter 7 of the Green Book states that shoulders are desir- able on any highway, but high right-of-way costs in urban areas may often preclude their use. When sufficient right-of- way is available, the design criteria previously presented for rural highways apply. Shoulders are not required by the Green Book for urban areas, and many such roadways are built using curbed cross sections, rather than shoulders. Traffic Operational Effects Chapter 17 (Urban Street Segments) of the HCM includes a procedure to determine the effect of the features of an urban street segment on free-flow speed. However, shoulder width Table 16. Adjustment to free-flow speed for lateral clearance on rural and suburban multilane highways (13). Four-lane highways Six-lane highways Total lateral clearance (ft) Reduction in free- flow speed (mph) Total lateral clearance (ft) Reduction in free- flow speed (mph) 12 0.0 12 0.0 10 0.4 10 0.4 8 0.9 8 0.9 6 1.3 6 1.3 4 1.8 4 1.7 2 3.6 2 2.8 0 5.4 0 3.9 NOTE: The values for reduction in free-flow speed presented in this table are used as fLW in Equation 3. SOURCE: Based on HCM Exhibit 14-9. Table 15. Recommended shoulder widths for rural multilane divided arterials (4, 5). Number of lanes in single direction Recommended right (outside) shoulder width (ft) Recommended left (inside) shoulder width (ft) 2 lanes 8 4 3 or more lanes 8 8 SOURCE: Adapted from Green Book Chapter 7.

17 is not one of the factors that influences free-flow speed. This suggests that shoulder width either has no effect on the free- flow speed on an urban street segment or an effect that is very small in comparison to the factors that are in the proce- dure (see HCM Exhibit 17-5). This contrasts with the HCM 2000 (22) which speculated that shoulder width influenced free-flow speed for urban streets, but did not quantify that effect. Traffic Safety Effects The HSM does not provide a CMF for shoulder width on urban and suburban arterials. 2.3.4 Freeways Design Criteria Chapter 8 of the Green Book recommends the shoulder widths for freeways shown in Table 18. The AASHTO policy on design standards for the Interstate highway system (14) requires a right (outside) shoulder with 10 ft of paved width. Where truck traffic exceeds a directional design hour volume (DDHV) of 250, a paved shoulder width of 12 ft should be considered. On a four-lane section, the paved width of the left (inside) shoulder is required to be at least 4 ft. On sections with six or more lanes, a left (inside) shoulder with a 10-ft width should be provided. Where truck traffic exceeds 250 DDHV, a paved width of 12 ft should be considered for the left (inside) shoulder. On four- to six-lane freeways in mountainous terrain, 8-ft paved right (outside) shoulders and 4-ft paved left (inside) shoulders may be used. On sections with eight or more lanes in mountainous terrain, a minimum paved shoulder width of 8 ft should be used on both sides of the roadway. Traffic Operational Effects Chapter 11 (Basic Freeway Segments) of the HCM esti- mates free-flow speed based on the lateral clearance on the right side of the roadway. Lateral clearance is measured from the edge of the travel lane to the edge of the paved shoulder. If the right-side lateral clearance is greater than or equal to 6 ft, no reduction in free-flow speed is made. The amount of free- flow speed reduction increases as the right-side lateral clear- ance decreases. Left-side lateral clearance is assumed to be greater than or equal to 2 ft for all cases. The free-flow speed reductions for right shoulder lateral clearance (generally interpreted as equivalent to right [outside] shoulder width) are shown in Table 19. The values in Table 19 are used as fLC in Equation 4 to determine free-flow speed (see Section 2.2.4). Traffic Safety Effects Results from NCHRP Project 17-45, which developed a proposed HSM safety prediction methodology for freeways, include CMFs for both right (outside) shoulder width and left (inside) shoulder width on freeways (25). The CMF for right (outside) shoulder width (where Ws = average right [outside] shoulder width for both directions of travel com- bined [ft]) is the following: • For fatal-and-injury single-vehicle crashes on tangent sections, ))((= − −CMF exp 0.0647 W 10 (8)s • For fatal-and-injury single-vehicle crashes on horizontal curves, CMF exp 0.097 W 10 (9)s( )( )= − − • For property-damage-only single-vehicle crashes on tan- gent sections, CMF 1.0 (10)= Table 17. CMFs for paved right (outside) shoulder width on multilane divided highway segments (12, 26). Average paved shoulder width 0 ft 2 ft 4 ft 6 ft 8 ft or more 1.18 1.13 1.09 1.04 1.00 SOURCE: Based on HSM Table 11-17. Table 18. Recommended shoulder widths for freeways (4, 5). Side of roadway DDHV for truck traffic (veh/h) Total number of freeway lanes Recommended shoulder width (ft) Right shoulder ≤250 All 10 Right shoulder >250 All 12 Left shoulder ≤250 Less than 6 4 Left shoulder ≤250 6 or more 10 Left shoulder >250 All 12 SOURCE: Adapted from Chapter 8 of the AASHTO Green Book.

18 • For property-damage-only single-vehicle crashes on hori- zontal curves, CMF exp 0.0840 W 10 (11)s( )( )= − − The base condition for this CMF is a 10-ft shoulder width (CMF = 1.0). The CMF is applicable to shoulders in the range of 4 to 14 ft. This CMF applies only to single-vehicle crashes; right (outside) shoulder width does not appear to have any effect on multiple-vehicle crashes. The CMF for left (inside) shoulder width (where Wis = average inside shoulder width for both directions of travel combined [ft]) is the following: • For fatal-and-injury crashes, CMF exp 0.0172 W 6 (12)is( )( )= − − • For property-damage-only crashes, CMF exp 0.0153 W 6 (13)is( )( )= − − The base condition for this CMF is a 6-ft shoulder width. The CMF is applicable to left (inside) shoulders in the range of 2 to 12 ft. The CMF applies to both multiple- and single- vehicle crashes. 2.3.5 Mitigation Strategies All the mitigation strategies for lane width presented in Section 2.2.5 also apply to shoulder width, with the obvious exception that adding paved or partially paved shoulders does not apply because the lack of a full shoulder is the condition to be mitigated. 2.4 Bridge Width Bridge width is the total width of all lanes and shoulders on a bridge, measured between the points on the bridge rail, curb, or other vertical elements that project farthest onto the roadway. A bridge width that meets design crite- ria maintains the minimum acceptable lane and shoulder width for the particular design condition as defined by area, functional class, design speed, and traffic volume. FHWA policy requires a design exception when a bridge is pro- posed to be constructed or retained with narrower lanes, shoulders, or both (7). Chapter 7 (Arterials) of the Green Book includes specific guidance on bridge widths that may remain in place on reconstruction projects (see Sections 2.4.1 and 2.4.2). Potential concerns associated with narrow bridges are two- fold. Narrow bridges that are relatively short represent a dis- continuity that may affect driver behavior. The narrowed cross section can make some drivers uncomfortable and cause them to dramatically reduce speed, increasing the risk of rear-end crashes and degrading operations on high-speed, high-volume facilities. The bridge rail may be close enough to the travel lanes to cause drivers to move toward the cen- terline or into adjacent lanes. In narrow bridges, the bridge railing itself is closer to the edge of pavement and thus rep- resents a roadside hazard. Even when properly designed and delineated, there is an increased risk of a roadside collision with the bridge railing or bridge end being closer to the edge of traveled way. A second set of concerns is evident for narrow bridges that are longer (say, greater than 500 ft in length). The safety and operational concerns at narrow bridges are similar to those on roads with narrow shoulders. There may be inad- equate space for storage of disabled vehicles, enforcement activities, emergency response, and maintenance work. The lack of shoulder width on the bridge may make it impossible to avoid a crash or object on the roadway ahead. In addi- tion, options are limited for non-motorized users such as bicyclists, forcing them onto the traveled lanes or close to the bridge rail. Narrow bridges on horizontal curves can have limited hori- zontal stopping sight distance past the bridge rail. Operations can be degraded, particularly on long bridges on high-speed roadways, because of speed reductions as drivers enter the Table 19. Adjustments for free-flow speed right-side lateral clearance on freeways (13). Right Shoulder Lateral Clearance (ft) Reduction in free-flow speed (mph) Number of lanes in one direction 2 lanes 3 lanes 4 lanes ≥5 lanes ≥ 6 0.0 0.0 0.0 0.0 5 0.6 0.4 0.2 0.1 4 1.2 0.8 0.4 0.2 3 1.8 1.2 0.6 0.3 2 2.4 1.6 0.8 0.4 1 3.0 2.0 1.0 0.5 0 3.6 2.4 1.2 0.6 NOTE: The values in this table are used as fLC in Equation 4. SOURCE: Based on HCM Exhibit 11-9.

19 narrowed cross section as well as decreased driver comfort on the bridge. 2.4.1 Rural Two-Lane Highways Design Criteria The minimum lane widths and shoulder widths shown in Tables 4 and 12, based on Green Book Exhibit 7-3, serve as the recommended minimum bridge widths for rural two- lane arterials. The combined minimum widths (lane width plus shoulder width) range from 30 ft (for a design speed of 40 mph and ADT less than 400 veh/day) to 40 ft (for a design speed of 75 mph and an ADT above 2,000 veh/day). On long bridges, defined as bridges with lengths of more than 200 ft, the offset to the parapet, rail, or barrier should be at least 4 ft from the edge of the traveled way or both sides of the roadway. Chapter 7 of the Green Book indicates that bridges with widths equal to the width of the traveled way plus 2 ft of clearance on each side may remain in place in reconstruction projects on arterials. Traffic Operational Effects Chapter 15 (Two-Lane Highways) of the HCM provides estimates for free-flow speeds on rural two-lane highways based on lane width and shoulder width. Bridges wide enough to accommodate 12-ft lanes and 6-ft shoulders will not reduce the free-flow speed below the base free-flow speed of the roadway; bridges of lesser widths will result in reduced free-flow speeds. Sections 2.2.1 and 2.3.1 of this report present more detailed information. The actual reduction in free-flow speed may be even greater than sug- gested in the HCM, particularly for long bridges, because the lateral obstruction is generally presented for the entire length of the bridge. Traffic Safety Effects The effects of lane and shoulder widths on safety for rural two-lane highways have been documented in Sections 2.2.1 and 2.3.1 of this report. While the design criteria for bridge width are based on the lane and shoulder width design cri- teria, it seems likely that safety might be more sensitive to bridge width than the lane and shoulder width, because every bridge has lateral obstructions (i.e., bridge rail or curb) at the outside edge of the shoulder. Turner (27) conducted research to predict crash rates as a function of bridge width, but the results appear potentially biased because only bridges that had experienced at least one crash were studied. A recent study by Bigelow et al. (28) in the FHWA CMF Clearinghouse provides a CMF for changing bridge width (bridge minus roadway width) from X to Y. The CMF is CMF 100 1 exp 0.116 Y X (14)( )( )( )= − − −p where X = bridge width before improvement (ft) Y = bridge width after improvement (ft) This is applicable to all crash types and severities. However, this CMF applies only to low-volume roads with AADT less than or equal to 400 veh/day and speed limits greater than or equal to 45 mph. Research conducted under NCHRP Project 17-53 (see Sec- tion 4.3) included analysis of the crash history of 624 bridges on rural two-lane highways in California and 337 bridges on rural two-lane highways in Washington and found no statisti- cally significant effect of differences between roadway width on the approach roadway and on the bridge on crash frequency. 2.4.2 Rural Multilane Highways Design Criteria Design criteria for bridge widths on rural multilane high- ways are based on the lane and shoulder-width design criteria presented in Sections 2.2.2 and 2.3.2. Those design criteria in Chapter 7 of the Green Book recommend 12-ft lane widths for rural divided multilane arterials. For long bridges over 200 ft in length, the Green Book states that 4-ft right and left shoulders are acceptable. For shorter bridges, the normal rec- ommendation of an 8-ft right shoulder applies. Chapter 7 of the Green Book indicates that bridges with widths equal to the width of the traveled way plus 2 ft of clearance on each side may remain in place in reconstruction projects. Traffic Operational Effects Chapter 14 (Multilane Highways) of the HCM provides esti- mates for free-flow speeds on multilane highways based on lane width and lateral clearance. Bridges wide enough to accom- modate 12-ft lanes and at least 6 ft of lateral clearance on both the left and right sides of the road will not reduce the free-flow speed below the base level; bridges of lesser widths will result in reduced free-flow speed levels. Sections 2.2.2 and 2.3.2 of this report present more detailed information. The actual reduction in free-flow speed may be even greater than suggested in the HCM, particularly for long bridges, because the lateral obstruc- tion is generally present for the entire length of the bridge. Traffic Safety Effects See discussion in Section 2.4.1 of this report.

20 2.4.3 Urban and Suburban Arterials Design Criteria Chapter 7 of the Green Book states that the minimum clear width for new bridges should be the same as the minimum curb-to-curb distance of the roadway for general conditions. For bridges that exceed 200 ft in length, the offsets to parapets, rails, or barriers may be reduced to 4 ft where shoulders or parking lanes are provided on the arterial. Traffic Operational Effects According to the “Limitations of the Methodology” discus- sion in Chapter 17 (Urban Streets) of the HCM, the HCM urban streets methodology does not directly account for capac- ity constraints such as a narrow bridge between intersections. Traffic Safety Effects See discussion in Section 2.4.1 of this report. 2.4.4 Freeways Design Criteria Minimum widths for lanes and shoulders on freeways are presented in Chapter 8 of the Green Book and have been summarized in Sections 2.2.4 and 2.3.4 of this report. A total bridge width for a freeway would depend on these minimum width values. As a general example, the following widths are recommended for a two-way viaduct freeway with ramps: • Median width: 10 to 22 ft • Lane width: 12 ft • Right shoulder width: 10 ft • Left shoulder width: 4 to 10 ft • Parapet width: 2 ft • Clearance between structure and building line: 15 ft Traffic Operational Effects Chapter 11 (Basic Freeway Segments) of the HCM provides estimates for free-flow speeds on freeways based on lane width and lateral clearance. Bridges wide enough to accommodate 12-ft lanes, at least 6 ft of right-side lateral clearance, and at least 2 ft of left-side lateral clearance will not reduce the free- flow speed below the base value; bridges of lesser widths will result in reduced free-flow speed values. Sections 2.2.4 and 2.3.4 of this report present more detailed information. The actual reduction in free-flow speed may be even greater than suggested in the HCM, particularly for long bridges, because the lateral obstruction is generally presented for the entire length of the bridge. Traffic Safety Effects See discussion in Section 2.4.1 of this report. 2.4.5 Mitigation Strategies Strategies for mitigating narrow bridge widths are directed primarily at improving a driver’s ability to see or to anticipate the narrowed cross section of the bridge, the bridge rail, and the lane lines. Typical mitigation strategies include the fol- lowing (7): • Advance signing • Improved delineation (pavement makings, lane delinea- tion, roadside reflectors, high-visibility bridge rail) • Bridge lighting • Skid-resistant pavement • Anti-icing systems • Crashworthy bridge rail and approach guardrail • Emergency pull-off areas • Surveillance (for long, high-volume bridges) 2.5 Structural Capacity Structural capacity has no effect on traffic operations, and its effect on safety is related only to the probability of a struc- tural failure, not to the likelihood of traffic crashes. For this reason, structural capacity is not reviewed here and will not be addressed in this research. 2.6 Horizontal Alignment Horizontal alignment involves design of the horizontal curves and tangents along a roadway section. In the context of the controlling criteria for design, horizontal alignment addresses only horizontal curves, not tangent sections, and the horizontal alignment criterion addresses only curve radius. Superelevation of horizontal curves is addressed by a separate controlling criterion. While the length of a horizontal curve and the length of tangent preceding a horizontal curve may influence traffic operations and safety and should be consid- ered as part of the design process, they are not part of the con- trolling criteria and do not require design exceptions. Chapter 3 of the Green Book provides guidance for select- ing minimum radii for horizontal curves based on design speed, the maximum superelevation rate (emax), and the max- imum side friction factor (fmax), which sets an upper limit on lateral acceleration based on driver comfort. This methodol- ogy is applicable to each of the road types discussed below, although additional guidance is provided for each road type individually as well. Table 20 presents design criteria for min- imum curve radius for three selected maximum supereleva- tion rates.

21 2.6.1 Rural Two-Lane Highways Design Criteria The design criteria for minimum curve radius presented in Table 20 apply to rural two-lane highways. Traffic Operational Effects Chapter 15 (Two-lane Highways) of the HCM uses free-flow speed in the determination of LOS. The chapter states that the base free-flow speed is the speed that would be expected on the basis of the facility’s horizontal and vertical alignment, if standard lane and shoulder widths were present and there were no roadside access points. However, the HSM provides no methodology to determine the effect of horizontal curva- ture on base free-flow speed. The IHSDM design consistency module (10, 29) includes a series of models for predicting the reduction in vehicle speed on horizontal curves from the design speed or tangent speed. These models are presented in Table 21. It should be noted Design speed (mph) Maximum e (%) Maximum f Total (e/100 + f ) Calculated minimum radius (ft) Rounded minimum radius (ft) 10 6.0 0.38 0.44 15.2 15 15 6.0 0.32 0.38 39.5 39 20 6.0 0.27 0.33 80.8 81 25 6.0 0.23 0.29 143.7 144 30 6.0 0.20 0.26 230.8 231 35 6.0 0.18 0.24 340.3 340 40 6.0 0.16 0.22 484.8 485 45 6.0 0.15 0.21 642.9 643 50 6.0 0.14 0.20 833.3 833 55 6.0 0.13 0.19 1,061.4 1,060 60 6.0 0.12 0.18 1,333.3 1,330 65 6.0 0.11 0.17 1,656.6 1,660 70 6.0 1.10 0.16 2,041.7 2,040 75 6.0 0.09 0.15 2,500.0 2,500 80 6.0 0.08 0.14 3,047.6 3,050 10 8.0 0.38 0.46 14.5 14 15 8.0 0.32 0.40 37.5 38 20 8.0 0.27 0.35 76.2 76 25 8.0 0.23 0.31 134.4 134 30 8.0 0.20 0.28 214.3 214 35 8.0 0.18 0.26 314.1 314 40 8.0 0.16 0.24 444.4 444 45 8.0 0.15 0.23 587.0 587 50 8.0 0.14 0.22 757.6 758 55 8.0 0.13 0.21 960.3 960 60 8.0 0.12 0.20 1,200.0 1,200 65 8.0 0.11 1.09 1,482.5 1,480 70 8.0 1.10 0.18 1,847.8 1,810 75 8.0 0.09 0.7 2,205.9 2,210 80 8.0 0.08 1.16 2,666.7 2,670 10 12.0 0.38 0.50 13.3 13 15 12.0 0.32 0.44 34.1 34 20 12.0 0.27 0.39 68.4 68 25 12.0 0.23 3.35 119.0 119 30 12.0 0.20 0.32 187.5 188 35 12.0 0.18 0.30 272.2 272 40 12.0 0.16 0.28 381.0 381 45 12.0 0.15 0.27 500.0 500 50 12.0 0.14 0.26 641.0 641 55 12.0 0.13 0.25 806.7 807 60 12.0 0.12 0.24 1,000.0 1,000 65 12.0 0.11 0.23 1,224.6 1,220 70 12.0 0.10 0.22 1,484.8 1,480 75 12.0 0.099 0.24 1,785.7 1,790 80 12.0 0.08 0.20 2,133.3 2,130 SOURCE: Based on Green Book Table 3-7 (abridged). Table 20. Design criteria for minimum curve radius for three selected maximum superelevation rates (4, 5).

22 that Table 21, as it appears in the original research, uses met- ric units for speed and curve radius. Traffic Safety Effects Chapter 10 (Rural Two-Lane Highways) of the HSM pro- vides a CMF for horizontal curves on rural two-lane roads which is computed as shown in Equation 15: CMF 1.55 L 80.2 R 0.012 S 1.55 L (15) c c ( )( ) ( ) ( )= × − × × where Lc = Length of horizontal curve including length of spiral transitions, if present (mi) R = Radius of curvature (ft) S = 1 if spiral transition curve is present: 0 if spiral transition curve is not present The base condition (CMF = 1.0) is a tangent segment with no curvature. This CMF applies to total crashes and is based on research by Zegeer et al. (30). An alternative CMF that incorporates the effects of both horizontal curvature and grade on straight grades (i.e., grades with constant percent grade) has been developed by Bauer and Harwood (31) in an FHWA study for consideration for a future edition of the HSM: • For fatal-and injury-crashes, CMF G R SG FI, . . . = + ×   + exp ln0 044 0 19 2 5730 4 52 1 1 R L for horizontal C           curves G for tangents on nonlevel gradexp 0 044.[ ] es for level tangents base condition1 0. ( )      ( )16 Table 21. IHSDM speed prediction equations for passenger vehiclesa (10, 29). AC EQ#b Alignment condition Equationc # of sites R2 MSE 1. Horizontal curve on grade: −9% ≤ G < −4% R 13.307710.102V85 21 0.58 51.95 2. Horizontal curve on grade: −4% ≤ G < 0% R 90.370798.105V85 25 0.76 28.46 3. Horizontal curve on grade: −0% ≤ G < 4% R 51.357482.104V85 25 0.76 24.34 4. Horizontal curve on grade: −4% ≤ G < 9% R 19.2752V85 23 0.53 52.54 5. Horizontal curve combined with sag vertical curve R 19.343832.105V85 25 0.92 10.47 6. Horizontal curve combined with non-limited sight distance crest vertical curve d 13 n/a n/a 7. Horizontal curve combined with limited- sight-distance crest vertical curve (i.e., K ≤ 43 m/%) R 51.357624.103V85 22 0.74 20.06 8. Sag vertical curve on horizontal tangent 7 n/a n/a 9. Vertical crest with non-limited-sight- distance (i.e., K > 43 m/%) on horizontal tangent V85 = assumed desired speed V85 = assumed desired speed 6 n/a n/a 10. Vertical crest with limited sight distance (i.e., K ≤ 43 m/%) on horizontal tangent K 69.14908.105V85 9 0.60 31.10 a Check the speeds predicted from Equations 1 or 2 in this table (for the downgrade) and Equations 3 or 4 in this table (for the upgrade) and use the lowest speed. This will ensure that the speed predicted along the combined curve will not be better than if just the horizontal curve was present (i.e., that the inclusion of a limited-sight-distance crest vertical curve will result in a higher speed). b AC EQ# = Alignment condition equation number; MSE = mean squared error. c Where: V85 = 85th percentile speed of passenger cars (km/h) R K = rate of vertical curvature G = grade (%)= radius of curvature (m) d Use lowest speed of the speeds predicted from Equations 1 or 2 in this table (for the downgrade) and Equations 3 or 4 in this table (for the upgrade). 61.96

23 • For property-damage-only crashes, exp 0.040 0.13 ln 2 5730 3.80 1 1 exp 0.040 1.0 (17) ,CMF G R R L for horizontal curves G for tangents on nonlevel grades for level tangents base condition SG PDO C ( ) ( ) [ ] ( ) = + × +              where G = absolute value of percent grade 2.6.2 Rural Multilane Highways Design Criteria The design criteria for minimum curve radius presented in Table 20 apply to rural multilane highways. Traffic Operational Effects Chapter 14 (Multilane Highways) of the HCM uses free- flow speed in the determination of LOS. The chapter states that the base free-flow speed is the speed that would be expected on the basis of the facility’s horizontal and vertical alignment, if standard lane and shoulder widths were present and there were no roadside access points. However, the HCM provides no methodology to determine the effect of horizon- tal curvature on base free-flow speed. Research conducted under NCHRP Project 17-53 (see Sec- tion 4.4) quantified the effect of horizontal curve radius on traffic speed for rural multilane highways as follows: Speed Speed 3136 R (18)curve approach= − where Speedcurve = Speed of traffic on horizontal curve (mph) Speedapproach = Speed of traffic on tangent approaching curve (mph) R = Radius of curvature (ft) Traffic Safety Effects Chapter 11 (Rural Multilane Highways) of the HSM does not include any CMFs for horizontal curves on rural multi- lane highways. Thus, the safety effect of horizontal curves on rural multilane highways has not been documented. There are several CMFs for horizontal curve radius in the FHWA CMF Clearinghouse, but none of these is specifically appli- cable to rural multilane highways. Research conducted under NCHRP Project 17-53 (see Sec- tion 4.4) developed the following CMFs for the effect of hori- zontal curvature on rural four-lane divided highways: • For fatal-and-injury crashes, CMF exp 0.87L 0.22 ln 2 5730 R (19)c ( )= − + ×  • For property-damage-only crashes, CMF exp 0.95L 0.26 ln 2 5730 R (20)c ( )= − + ×  No comparable CMFs are available for rural four-lane undivided highways. 2.6.3 Urban and Suburban Arterials Design Criteria The design criteria for minimum curve radius presented in Table 20 apply to urban and suburban arterials. On low- speed urban streets, with design speeds of 45 mph or less, minimum radii sharper than those shown in Table 20 can be used (see Green Book Exhibit 3-16). Traffic Operational Effects Chapter 17 (Urban Street Segments) of the HSM includes a method for estimating the free-flow speed for an urban street section. The factors considered include speed limit, median type, curb presence, and access-point density. There is no effect of horizontal alignment in the procedure. In essence, the procedure assumes that the effect of curvature on speed is minimal. Research conducted under NCHRP Project 17-53 (see Sec- tion 4.4) quantified the effect of horizontal curve radius on traffic speed urban and suburban arterials as follows: Speed Speed 2203 R (21)curve approach= − Traffic Safety Effects Chapter 12 (Urban and Suburban Arterials) of the HSM does not include any CMFs for the effect of horizontal curves on urban and suburban arterials. Recent research by Hauer et al. (32) observed on-road crash frequencies for horizon- tal curves on urban four-lane undivided arterials to be lower than tangent sections in the same corridors; the opposite was found to be the case for run-off-road crashes. Since on- road crashes are predominant on urban arterials, Hauer et al.

24 concluded that the role of horizontal curvature in safety for this type of road may need reconsideration. There are several CMFs for horizontal curve radius in the FHWA CMF Clear- inghouse, but none of these is specifically applicable to urban and suburban arterials. 2.6.4 Freeways Design Criteria The design criteria for minimum curve radius presented in Table 20 apply to freeways. Traffic Operational Effects Chapter 11 (Multilane Highways) of the HCM uses free- flow speed in the determination of LOS. The chapter states that the base free-flow speed is the speed that would be expected on the basis of the facility’s horizontal and vertical alignment, if standard lane and shoulder widths were pres- ent and there were no roadside access points. However, no methodology to determine the effect of horizontal curvature on base free-flow speed is provided in the HCM. Traffic Safety Effects Results from NCHRP Project 17-45, which developed a proposed HSM safety prediction methodology for freeways, includes a CMF for the safety effect of horizontal curves on safety (25). The CMFs for horizontal curves (where R = radius of curvature [ft])are the following: • For fatal-and-injury multiple-vehicle crashes, CMF 1.0 0.0172 5730 R (22) 2( )= + × • For property-damage-only multiple-vehicle crashes, CMF 1.0 0.0340 5730 R (23) 2( )= + × • For fatal-and-injury single-vehicle crashes, CMF 1.0 0.0719 5730 R (24) 2( )= + × • For property-damage-only single-vehicle crashes, CMF 1.0 0.0626 5730 R (25) 2( )= + × 2.6.5 Mitigation Strategies Mitigation strategies for horizontal curves with sharper radii than established design criteria include the following (7): • Advance warning with signing and pavement markings • Dynamic message signs • Delineation (chevrons, post-mounted delineators, reflec- tors on barriers) • Roadway widening • Skid-resistant pavement • Lighting • Shoulder, painted edgeline, or centerline rumble strips • Paved or partially paved shoulders • Safety edge treatment • Roadside improvements (clear recovery area, traversable slopes, breakaway safety hardware, barrier where appropriate) 2.7 Vertical Alignment Vertical alignment generally consists of two elements: grades and vertical curves. Both of these elements are considered in the controlling criteria. Grade is treated as a separate con- trolling criterion (see Section 2.8). Two types of vertical curves are considered in vertical alignment design: crest ver- tical curves and sag vertical curves. Both crest and sag verti- cal curves have two types, known as Type 1 and Type 2, as illustrated in Figure 5. The Green Book design criteria for crest vertical curve lengths are illustrated in Figure 6. Crest verti- cal curve length is selected primarily to achieve minimum stopping sight distance on the vertical curve. Stopping sight distance is treated as a separate controlling criterion (see Sec- tion 2.9). Thus, the only element of vertical alignment not dealt with by a separate controlling criterion is sag vertical curve length. Sag vertical curve length is normally selected so that the curve does not restrict the length of roadway illuminated by vehicle headlights, which would reduce stop- ping sight distance at night. Figure 7 presents the Green Book design criteria for sag vertical curve length. The parameter, K, in Figures 6 and 7 is the ratio of the algebraic difference in grade, A, to the length of the vertical curve. Recent research on sag vertical curves is documented in NCHRP Web-Only Document 198: Sag Vertical Curve Design Criteria for Head- light Sight Distance. 2.7.1 Rural Two-Lane Highways Design Criteria The design criteria for crest and sag vertical curves, pre- sented in Figures 6 and 7, respectively, are applicable to rural two-lane highways.

25 SOURCE: Based on Green Book Figure 3-41. Figure 5. Types of vertical curves (4, 5). SOURCE: Based on Green Book Figure 3-43. Figure 6. Design controls for crest vertical curves—open road conditions (4, 5).

26 Traffic Operational Effects Chapter 15 (Two-Lane Highways) of the HCM provides a methodology for adjusting the LOS boundaries on rural two-lane highways to account for vertical alignment, consid- ering general terrain classes or specific grades, as well as the percentages in the traffic flow of two types of heavy vehicles (trucks and recreational vehicles). Since these vertical align- ment effects are primarily a function of grade, they are dis- cussed in Section 2.8 of this report. Crest vertical curve effects are addressed in Section 2.9 of this report. There are no known quantifiable operational effects of sag vertical curve length; it is likely that any such effects are minimal as long as the ride comfort criteria in Green Book Equation 3-51 are met. Traffic Safety Effects Chapter 10 (Rural Two-Lane Highways) of the HSM includes a factor for the effect of grade on safety; this effect is discussed in Section 2.8 of this report. Chapter 10 (HSM) does not include any effect of crest or sag vertical curves on safety. The effect of crest vertical curves on safety is likely related to stopping sight distance and is discussed in Section 2.9 of this report. There is no known effect of sag vertical curve length on safety. Sag vertical curve length is essentially irrelevant to safety under daytime conditions, because the driver can see beyond the sag vertical curve unless a horizontal curve is present. At night, drivers at speeds of 50 mph or more gen- erally outdrive their headlights. This is generally true what- ever the vertical alignment, so there is no special risk on sag vertical curves. Furthermore, as discussed in Section 2.9, the object most likely to be struck by a driver in a limited-sight- distance situation is another vehicle on the roadway ahead. The taillights of such vehicles and the dispersion of light from their headlights should make such vehicles clearly visible at night, even beyond the limits of the sag vertical curve unless a horizontal curve is also present. Thus, it seems unlikely that sag vertical curve length would have much effect on safety. An important exception occurs when an overpass that might block the driver’s view of the road ahead is located on a sag vertical curve. This situation is addressed explicitly in Green Book Chapter 3. It should also be noted that overpass struc- tures on rural two-lane highways are not common. Recent research for FHWA by Bauer and Harwood (31) completed since the publication of the first edition of the HSM, developed the following CMFs for Type 1 crest vertical curves (LVC = length of vertical curve): • For fatal-and injury-crashes, exp 0.0088 5730 1.0 1 1.0 (26) 1,CMF R L K for horizontal curves for tangents at Type crests for level tangents base condition C FI VC( ) ( ) =          SOURCE: Based on Green Book Figure 3-44. Figure 7. Design controls for sag vertical curves—open road conditions (4, 5).

27 • For property-damage-only crashes, exp 0.0046 5730 1.0 1 1.0 (27) 1,CMF R L K for horizontal curves for tangents at Type crests for level tangents base condition C PDO VC( ) ( ) =          The equivalent CMFs for Type 2 crest vertical curves are the following: • For fatal-and injury-crashes, exp 0.20 ln 2 5730 1.0 2 1.0 (28) 2,CMF R for horizontal curves for tangents at Type crests for level tangents base condition C FI ( ) ( ) = ×        • For property-damage-only crashes, exp 0.10 ln 2 5730 1.0 2 1.0 (29) 2,CMF R for horizontal curves for tangents at Type crests for level tangents base condition C PDO ( ) ( ) = ×        Bauer and Harwood (31) also developed the following CMFs for Type 1 sag vertical curves: • For fatal-and injury-crashes, exp 10.51 1 0.011 5730 exp 10.51 1 1 1.0 (30) 1,CMF K R L K for horizontal curves K for tangents at Type sags for level tangents base condition S FI VC( ) ( ) = +               • For property-damage-only crashes, exp 8.62 1 0.010 5730 exp 8.62 1 1 1.0 (31) 1,CMF K R L K for horizontal curves K for tangents at Type sags for level tangents base condition S PDO VC( ) ( ) = +            The equivalent CMFs for Type 2 sag vertical curves are the following: • For fatal-and injury-crashes, exp 0.188 ln 2 5730 1.0 2 1.0 (32) 2,CMF R for horizontal curves for tangents at Type sags for level tangents base condition S FI ( ) ( ) = ×        • For property-damage-only crashes, exp 0.022 5730 1.0 2 1.0 (33) 2,CMF R A for horizontal curves for tangents at Type sags for level tangents base condition S PDO ( ) ( ) =          2.7.2 Rural Multilane Highways Design Criteria The design criteria for crest and sag vertical curves, pre- sented in Figures 6 and 7, respectively, are applicable to rural multilane highways. Traffic Operational Effects Chapter 14 (Multilane Highways) of the HCM provides a methodology for adjusting the LOS boundaries on a multilane highway to account for vertical alignment considering gen- eral terrain classes or specific grades, as well as the percentages in the traffic flow of two types of heavy vehicles (trucks

28 and recreational vehicles). Since these vertical alignment effects are primarily a function of grade, they are discussed in Section 2.8. Crest vertical curve effects are addressed in Sec- tion 2.9 of this report. There are no known quantifiable opera- tional effects of sag vertical curve length; it is likely that such effects are minimal, as long as the ride comfort criteria in Green Book Equation 3-51 are met. Traffic Safety Effects Chapter 11 (Rural Multilane Highways) of the HSM does not include any factors to account for the effects of grade, crest vertical curve length, or sag vertical curve length on safety. Based on the reasoning presented in Section 2.7.1, sag vertical curve length in particular seems unlikely to have much influence on safety except where an overpass is located on a sag vertical curve. 2.7.3 Urban and Suburban Arterials Design Criteria The design criteria for crest and sag vertical curves, pre- sented in Figures 6 and 7, respectively, are applicable to urban and suburban arterials. Traffic Operational Effects Chapter 17 (Urban Street Segments) of the HCM recom- mends that free-flow speeds for urban street segments be measured in the field or estimated based on the street’s func- tional and design categories. No specific quantitative proce- dures are provided. Traffic Safety Effects Chapter 12 (Urban and Suburban Arterials) of the HSM does not include any factors to account for the effects of grade, crest vertical curve length, or sag vertical curve length on safety. Crest vertical curve effects are addressed in Section 2.9 of this report. There are no known quantifiable operational effects of sag vertical curve length; it is likely that such effects are minimal, as long as the ride comfort criteria in Green Book Equation 3-51 are met. 2.7.4 Freeways Design Criteria The design criteria for crest and sag vertical curve length, presented in Figures 6 and 7, respectively, are applicable to freeways. Traffic Operational Effects Chapter 11 (Basic Freeway Segments) of the HCM pro- vides a methodology for adjusting the LOS boundaries on a freeway to account for vertical alignment considering general terrain classes or specific grades, as well as the percentages in the traffic flow of two types of heavy vehicles (trucks and recreational vehicles). Since these vertical alignment effects are primarily a function of grade, they are discussed in Sec- tion 2.8. Crest vertical curve effects are addressed in Section 2.9 of this report. There are no known quantifiable opera- tional effects of sag vertical curve length; it is likely that such effects are minimal. Traffic Safety Effects The HSM safety prediction methodology for freeways developed in NCHRP Project 17-45 does not include any safety effects for grades, crest vertical curve length, or sag vertical curve length (25). 2.7.5 Mitigation Strategies Most design exceptions for vertical alignment are related to grades and crest vertical curves. Appropriate mitigation strategies for grades and crest vertical curves are discussed in Sections 2.8 and 2.9, respectively. Sag vertical curve lengths that do not meet established criteria do not often need design exceptions (7). Mitigation of sag vertical curve lengths that do not meet established criteria is unlikely to be needed unless there is a specific crash pattern of rear-end crashes or an overpass is present on the sag vertical curve. If mitigation is needed, the provision of lighting is an obvious strategy. 2.8 Grade Grade is the rate of change of vertical elevation along a roadway. The controlling criterion for grade includes both maximum and minimum grades. Maximum grades are estab- lished for specific roadway types and functional classes (see below). A design exception is needed where steeper grades are to be provided or retained. Chapter 3 of the Green Book provides general guidance for selecting acceptable grades for roadways. Generally, a maxi- mum grade of 5 percent is appropriate for a design speed of 70 mph, while maximum grades of 7 to 12 percent are appro- priate for design speeds of 30 to 50 mph. Green Book Exhibits 3-55 and 3-56 (not shown here) esti- mate running speeds of typical heavy trucks based on the percent grade and the length of the roadway section at that grade. These exhibits or the Truck Speed Performance Model (TSPM) developed by Harwood et al. (33) can be used to

29 establish critical lengths of grade that would produce a dif- ferential of 15 mph or more between the minimum speed of trucks and the average speed of traffic. Depending on traf- fic and truck volumes, locations with critical length of grade may warrant the addition of truck climbing lanes. However, the truck climbing lane criteria are not part of the controlling criterion for grade and do not require design exceptions. In fact, quite the opposite is true—the critical length of grade criteria merely suggest locations where truck climbing lanes might be considered. 2.8.1 Rural Two-Lane Highways Design Criteria Chapter 7 of the Green Book provides additional guidance for maximum grade selection for rural arterials, including rural two-lane highways. Table 22 shows the recommended maximum grades for rural arterials based on terrain type and design speed. Traffic Operational Effects Chapter 15 (Two-Lane Highways) of the HCM provides a methodology for adjusting demand flow rates for two-lane highways based on grade. Two adjustment factors in Chapter 15 (HCM) are affected by grade: the grade adjustment factor (fg) and the heavy vehicle adjustment factor (fHV). Separate adjust- ments are made in the computations for the two service mea- sures for two-lane highways: average travel speed and percent time spent following. Average Travel Speeds. The grade adjustment factor, fg, accounts for vehicles traveling more slowly on grades than they would on a level roadway. A smaller value of fg will result in a higher demand flow rate. Table 23 presents values of fg for various flow rates for level or rolling terrain. For segments with mountainous terrain, or on any segment with a grade steeper than 3 percent over a distance of 0.6 mi or more, the procedure for calculating fg relies on more extensive criteria partially illustrated in Table 24. Type of terrain Maximum grade (%) for specified design speed 40 mph 45 mph 50 mph 55 mph 60 mph 65 mph 70 mph 75 mph 80 mph Level 5 5 4 4 3 3 3 3 3 Rolling 6 6 5 5 4 4 4 4 4 Mountainous 8 7 7 6 6 5 5 5 5 SOURCE: Based on AASHTO Green Book Table 7-2. Table 22. Maximum grade for rural arterials (4, 5). One-direction demand flow rate (veh/h) Type of terrain Level terrain and specific downgrades Rolling terrain ≤ 100 1.00 0.67 200 1.00 0.75 300 1.00 0.83 400 1.00 0.90 500 1.00 0.95 600 1.00 0.97 700 1.00 0.98 800 1.00 0.99 ≥ 900 1.00 1.00 SOURCE: Based on HCM Exhibit 15-9. Table 23. Grade adjustment factor (fg) to determine speeds on two-way and directional segments for two-lane highways (13). Grade (%) Grade length (mi) Grade adjustment factor, fg Directional demand flow rate vvph (veh/h) ≤ 100 200 300 400 500 600 700 800 ≥ 900 ≥ 3 < 3.5 0.25 0.78 0.84 0.87 0.91 1.00 1.00 1.00 1.00 1.00 0.50 0.75 0.83 0.86 0.90 1.00 1.00 1.00 1.00 1.00 0.75 0.73 0.81 0.85 0.89 1.00 1.00 1.00 1.00 1.00 1.00 0.73 0.79 0.83 0.88 1.00 1.00 1.00 1.00 1.00 1.50 0.73 0.79 0.83 0.87 0.99 0.99 1.00 1.00 1.00 2.00 0.73 0.79 0.82 0.86 0.98 0.98 0.99 1.00 1.00 3.00 0.73 0.78 0.82 0.85 0.95 0.96 0.96 0.97 0.98 ≥ 4.00 0.73 0.78 0.81 0.85 0.94 0.94 0.95 0.95 0.96 SOURCE: Based on HCM Exhibit 15-10. Table 24. Grade adjustment factor for estimating travel speed on specific upgrades for two-lane highways (13).

30 The heavy vehicle adjustment factor, fHV, accounts for heavy vehicles traveling more slowly on grades than passen- ger cars. A larger value of the passenger-car equivalence fac- tors for heavy vehicles, ET or ER, results in a higher demand flow rate. Table 25 presents passenger-car equivalence factors for trucks (ET) and recreational vehicles (ER). For segments with mountainous terrain, or on any segment with a grade steeper than 3 percent over a distance of 0.6 mi or more, the procedures for calculating fHV rely on the more extensive criteria in Tables 26 and 27. The demand flow rate in the analysis direction of travel for use in the average travel speed determination is computed as: v V PHF f f (34)d d g HV = × × where vd = demand flow rate for analysis direction (pc/L) PHF = peak hour factor Vehicle type Directional demand flow rate, Vvph (veh/h) Passenger-car equivalents for level terrain and specific downgrades Passenger-car equivalents for rolling terrain Trucks, ET ≤ 100 1.9 2.7 200 1.5 2.3 300 1.4 2.1 400 1.3 2.0 500 1.2 1.8 600 1.1 1.7 700 1.1 1.6 800 1.1 1.4 ≥900 1.0 1.3 RVs, ER All flows 1.0 1.1 SOURCE: Based on HCM Exhibit 15-11. Table 25. Passenger-car equivalents for trucks (ET) and recreational vehicles (RVs) (ER) to determine speeds on directional segments for two-lane highways (13). Grade (%) Grade length (mi) Passenger-car equivalent for trucks, ET Directional demand flow rate vvph (veh/h) ≤ 100 200 300 400 500 600 700 800 ≥ 900 ≥ 3 < 3.5 0.25 2.6 2.4 2.3 2.2 1.8 1.8 1.7 1.3 1.1 0.50 3.7 3.4 3.3 3.2 2.7 2.6 2.6 2.3 2.0 0.75 4.6 4.4 4.3 4.2 3.7 3.6 3.4 2.4 1.9 1.00 5.2 5.0 4.9 4.9 4.4 4.2 4.1 3.0 1.6 1.50 6.2 6.0 5.9 5.8 5.3 5.0 4.8 3.6 2.9 2.00 7.3 6.9 6.7 6.5 5.7 5.5 5.3 4.1 3.5 3.00 8.4 8.0 7.7 7.5 6.5 6.2 6.0 4.6 3.9 ≥ 4.00 9.4 8.8 8.6 8.3 7.2 6.9 6.6 4.8 3.7 SOURCE: Based on HCM Exhibit 15-12. Table 26. Passenger-car equivalents for trucks for estimating travel speed on specific upgrades for two-lane highways (13). Grade (%) Grade length (mi) Passenger-car equivalent for RVs, ER Directional demand flow rate vvph (veh/h) ≤ 100 200 300 400 500 600 700 800 ≥ 900 ≥ 3 < 3.5 ≤ 0.25 1.1 1.1 1.1 1.0 1.0 1.0 1.0 1.0 1.0 > 0.25 ≤ 0.75 1.2 1.2 1.1 1.1 1.0 1.0 1.0 1.0 1.0 > 0.75 ≤ 1.25 1.3 1.2 1.2 1.1 1.0 1.0 1.0 1.0 1.0 > 1.25 ≤ 2.25 1.4 1.3 1.2 1.1 1.0 1.0 1.0 1.0 1.0 > 2.25 1.5 1.4 1.3 1.2 1.0 1.0 1.0 1.0 1.0 SOURCE: Based on HCM Exhibit 15-13. Table 27. Passenger-car equivalents for RVs for estimating travel speed on specific upgrades for two-lane highways (13).

31 fg = grade adjustment factor from Table 23 or 24 fHV = heavy vehicle adjustment factor from HCM Equa- tions 15-4 or 15-5, which utilize data from Tables 25 through 27 The demand flow rate in the opposing direction is determined in a manner entirely analogous to Equation 34. The service measure average travel speed, which is one of two mea- sures used to determine LOS, is then determined with HCM Equation 15-6. Percent Time Spent Following. The demand flow rates are determined slightly differently when used for percent time spent following rather than average travel speed as the service measure. Similar to the methodology for speed cal- culations, two adjustment factors are affected by grade: the grade adjustment factor (fg), and the heavy vehicle adjust- ment factor (fHV). Demand flow rate for the analysis and opposing directions is determined using Equation 34. How- ever, for these calculations, Tables 28 through 31 are used instead of Tables 23 through 27 to determine the values of fg and fHV. Traffic Safety Effects Chapter 10 (Rural Two-Lane Highways) of the HSM presents the CMF for grade on two-lane highways as shown in Table 32. Table 32 presents the CMF by terrain categories. Directional demand flow rate (veh/h) Level terrain and specific downgrades Rolling terrain ≤ 100 1.00 0.73 200 1.00 0.80 300 1.00 0.85 400 1.00 0.90 500 1.00 0.96 600 1.00 0.97 700 1.00 0.99 800 1.00 1.00 ≥900 1.00 1.00 SOURCE: Based on HCM Exhibit 15-16. Table 28. Grade adjustment factor (fg) to determine percent time spent following on directional segments for two-lane highways (13). Grade (%) Grade length (mi) Grade adjustment factor, fg Directional demand flow rate vvph (veh/h) ≤ 100 200 300 400 500 600 700 800 ≥ 900 ≥ 3 < 3.5 0.25 1.00 0.99 0.97 2.2 1.8 1.8 1.7 1.3 1.1 0.50 1.00 0.99 0.98 3.2 2.7 2.6 2.6 2.3 2.0 0.75 1.00 0.99 0.98 4.2 3.7 3.6 3.4 2.4 1.9 1.00 1.00 0.99 0.98 4.9 4.4 4.2 4.1 3.0 1.6 1.50 1.00 0.99 0.98 5.8 5.3 5.0 4.8 3.6 2.9 2.00 1.00 0.99 0.98 6.5 5.7 5.5 5.3 4.1 3.5 3.00 1.00 1.00 0.99 7.5 6.5 6.2 6.0 4.6 3.9 ≥ 4.00 1.00 1.00 1.00 8.3 7.2 6.9 6.6 4.8 3.7 SOURCE: Based on HCM Exhibit 15-17. Table 29. Grade adjustment factor (fg) for estimating percent time spent following on specific upgrades for two-lane highways (13). Vehicle type Directional demand flow rate (veh/h) Passenger-car equivalents for level and specific downgrades Passenger-car equivalents for rolling terrain Trucks, ET ≤ 100 1.1 1.9 200 1.1 1.8 300 1.1 1.7 400 1.1 1.6 500 1.0 1.4 600 1.0 1.2 700 1.0 1.0 800 1.0 1.0 ≥ 900 1.0 1.0 RVs, ER All 1.0 1.0 SOURCE: Based on HCM Exhibit 15-18. Table 30. Passenger-car equivalents for trucks (ET) and RVs (ER) for estimating percent time spent following on directional segments for two-lane highways (13).

32 The underlying research (34, 35) presents the CMF as a con- tinuous function rather than a step function, as follows: CMF (1.0 0.016 G) (35)= + where G = absolute value of percent grade. In other words, the CMF increases by 0.016 for each percent grade. 2.8.2 Rural Multilane Highways Design Criteria The maximum grade criteria presented in Table 22 also apply to rural multilane highways. Traffic Operational Effects Chapter 14 (Multilane Highways) of the HCM presents a methodology for determining the effect of grades on opera- tions of multilane highways. The procedure is similar to the procedure described above for two-lane highways. The multilane highway methodology is much simpler—the only factor that is used in determining the LOS boundaries is the fHV factor. The heavy vehicles adjustment factor, fHV, adjusts the demand flow rate to account for the fact that heavy vehicles generally travel more slowly on grades than passenger cars. A larger value of ET (or ER) results in a higher demand flow rate. Table 33 presents passenger equivalence factors for trucks and buses (ET) and RVs (ER). For segments with a grade between 2 and 3 percent for more than 0.5 mi or with a grade steeper than 3 percent for more than 0.25 mi, the procedures for calculating ET and ER rely on the more exten- sive Tables 34, 35, and 36. The value of fHV is determined with HCM Equation 14-4, the demand flow rate is deter- mined with HCM Equation 14-3, and density, the service measure for multilane highways, is determined with HCM Equation 14-5. Traffic Safety Effects Chapter 11 of the HSM does not include a CMF for grade on rural multilane highways. 2.8.3 Urban and Suburban Arterials Design Criteria Table 37 presents recommended maximum grades for urban arterials. The Green Book states that when these can- not be attained, climbing lanes should be considered; in this Grade (%) Grade length (mi) Passenger-car equivalent for trucks ET Directional demand flow rate vvph (veh/h) ≤ 100 200 300 400 500 600 700 800 ≥ 900 ≥ 3 < 3.5 ≤ 2.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 3.00 1.5 1.3 1.3 1.2 1.0 1.0 1.0 1.0 1.0 ≥ 4.00 1.6 1.4 1.3 1.3 1.0 1.0 1.0 1.0 1.0 ≥ 3 < 4.5 ≤ 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.50 1.1 1.1 1.0 1.0 1.0 1.0 1.0 1.0 1.0 2.00 1.6 1.3 1.0 1.0 1.0 1.0 1.0 1.0 1.0 3.00 1.8 1.4 1.1 1.2 1.2 1.2 1.2 1.2 1.2 ≥ 4.00 2.1 1.9 1.8 1.7 1.4 1.4 1.4 1.4 1.4 SOURCE: Based on HCM Exhibit 15-19. Table 31. Passenger-car equivalents for trucks for estimating percent time spent following on specific upgrades for two-lane highways (13). Level grade Moderate terrain Steep terrain (≤ 3%) (3% < grade ≤ 6%) (> 6%) 1.00 1.10 1.16 SOURCE: Based on HSM Table 10-11. Table 32. CMF for grade of roadway segments (12). Passenger-car equivalent Type of terrain Level Rolling Mountainous ET (trucks and buses) 1.5 2.5 4.5 ER (RVs) 1.2 2.0 4.0 SOURCE: Based on HCM Exhibit 14-12. Table 33. Passenger-car equivalents for heavy vehicles in general terrain segments on multilane highways (13).

33 case, the use of a climbing lane would be considered a miti- gation strategy and not part of the controlling criterion. Traffic Operational Effects According to Chapter 17 (Urban Street Segments) of the HCM, one of the first steps in determining the LOS for an urban street is determining the free-flow speed of traffic on the road segment. The steeper the upgrade of a roadway segment, the slower the free-flow speed will be. Chapter 17 (HCM) recom- mends that the free-flow speed be measured if possible; other- wise it must be estimated based on the street’s functional and design categories. No methodology is provided for estimating the effect of grade on free-flow speed for an urban street. Table 34. Passenger-car equivalents for trucks and buses on upgrades on multilane highways (13). Upgrade (%) Length (mi) ET Percentage of trucks and buses 2% 4% 5% 6% 8% 10% 15% 20% 25% ≤ 2 All 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 > 2 to 3 0.00 to 0.25 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 > 0.25 to 0.50 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 > 0.50 to 0.75 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 > 0.75 to 1.00 2.0 2.0 2.0 2.0 1.5 1.5 1.5 1.5 1.5 > 1.00 to 1.50 2.5 2.5 2.5 2.5 2.0 2.0 2.0 2.0 2.0 > 1.50 3.0 3.0 2.5 2.5 2.0 2.0 2.0 2.0 2.0 SOURCE: Based on HCM Exhibit 14-13 (abridged). Table 35. Passenger-car equivalents for RVs on upgrades on multilane highways (13). Upgrade (%) Length (mi) ER Percentage of RVs 2% 4% 5% 6% 8% 10% 15% 20% 25% ≤ 2 All 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 > 2 to 3 0.00 to 0.50 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 > 0.50 3.0 1.5 1.5 1.5 1.5 1.5 1.2 1.2 1.2 > 3 to 4 0.00 to 0.25 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 > 0.25 to 0.50 2.5 2.5 2.0 2.0 2.0 2.0 1.5 1.5 1.5 > 0.50 3.0 2.5 2.5 2.5 2.0 2.0 2.0 1.5 1.5 SOURCE: Based on HCM Exhibit 14-14 (abridged). Table 36. Passenger-car equivalents for trucks (ET) on specific downgrades on rural and suburban multilane highways (13). Percent downgrade Length of grade (mi) Proportion of trucks and buses 5% 10% 15% 20% < 4 All 1.5 1.5 1.5 1.5 4 to 5 ≤ 4 1.5 1.5 1.5 1.5 > 4 2.0 2.0 2.0 1.5 > 5 to 6 ≤ 4 1.5 1.5 1.5 1.5 > 4 5.5 4.0 4.0 3.0 > 6 ≤ 4 1.5 1.5 1.5 1.5 > 4 7.5 6.0 5.5 4.5 SOURCE: Based on HCM Exhibit 14-15. Table 37. Maximum grades for urban arterials (13). Type of terrain Maximum grade (%) for specified design speed 30 mph 35 mph 40 mph 45 mph 50 mph 55 mph 60 mph Level 8 7 7 6 6 5 5 Rolling 9 8 8 7 7 6 6 Mountainous 11 10 10 9 9 8 8 SOURCE: Based on Green Book Table 7-4.

34 Traffic Safety Effects Chapter 12 (Urban and Suburban Arterials) of the HSM does not include a CMF for grade on urban and suburban arterials. 2.8.4 Freeways Design Criteria Chapter 8 of the Green Book provides the following specific guidance for urban freeways. Grades on urban freeways should generally be comparable to those in rural areas. Steeper grades can be tolerated in urban areas, but because interchanges may be closely spaced in urban areas, flatter grades are desirable when practical. Table 38 provides recommended maximum grades for rural and urban freeways. Traffic Operational Effects Chapter 11 (Basic Freeway Segments) of the HCM pro- vides a methodology for determining the effect of grades on operations of freeways. The procedure is very similar to the procedure described above for multilane highways. The heavy vehicles adjustment factor, fHV, adjusts the demand volume to account for the tendency of heavy vehicles to travel more slowly on grades than passenger cars. Table 39 provides passenger-car equivalence factors for trucks and buses (ET) and RVs (ER). For any segment with a grade between 2 and 3 percent for more than 0.5 mi or with a grade steeper than 3 percent for more than 0.25 mi, the procedures for calculat- ing ET and ER rely on the more extensive Tables 40, 41, and 42. A larger value of ET or ER results in a larger demand flow rate. The value of fHV is determined with HCM Equation 11-3, the demand flow rate is determined with HCM Equation 11-2, and the service measure for multilane highways is determined with HCM Equation 11-4. Traffic Safety Effects The HSM safety prediction methodology for freeways devel- oped in NCHRP Project 17-45 does not include any safety effects for grades on freeways (25). 2.8.5 Mitigation Strategies The strategies for mitigating steep grades include the fol- lowing (7): • Providing drivers with advance warning signs for steep grades • Providing climbing lanes and downgrade lanes • Providing emergency escape ramps for trucks Table 38. Maximum grades for rural and urban freeways (4, 5). Type of terrain Maximum grade (%) for specified design speed 50 mph 55 mph 60 mph 65 mph 70 mph 75 mph 80 mph Level 4 4 3 3 3 3 3 Rolling 5 5 4 4 4 4 4 Mountainous 6 6 6 5 5 - - SOURCE: Based on Green Book Table 8-1. Table 39. Passenger-car equivalents on extended freeway segments (13). Passenger-car equivalent Type of terrain Level Rolling Mountainous ET (trucks and buses) 1.5 2.5 4.5 ER (RVs) 1.2 2.0 4.0 SOURCE: Based on HCM Exhibit 11-10. Table 40. Passenger-car equivalents for trucks and buses on upgrades for specific grades on freeways (13). Upgrade (%) Length (mi) ET Percentage of trucks and buses 2% 4% 5% 6% 8% 10% 15% 20% 25% < 2 All 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 ≥ 2 to 3 0.00 to 0.25 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 > 0.25 to 0.50 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 > 0.50 to 0.75 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 > 0.75 to 1.00 2.0 2.0 2.0 2.0 1.5 1.5 1.5 1.5 1.5 SOURCE: Based on HCM Exhibit 11-11 (abridged).

35 • Reducing the frequency or severity of lane-departure crashes (enhanced pavement markings; delineation; shoulder, painted edgeline, or centerline rumble strips; paved or par- tially paved shoulders; safety edge treatment; clear recov- ery area; traversable slopes; breakaway safety hardware; and barrier where appropriate). The strategies for mitigating flat grades include the follow- ing (7): • Adjusting the gutter profile • Providing special drainage systems 2.9 Stopping Sight Distance Stopping sight distance is the distance required for a driver to perceive or recognize a need to stop, react to that percep- tion, and then decelerate to a stop. Horizontal and vertical curves limit available sight distance for drivers, requiring a careful analysis of stopping sight distance during the design process. Sight distance needs are based on the design speed of the roadway and the grade of the roadway, since cars trav- eling downhill require a greater distance to stop than cars traveling uphill or on the level. The minimum stopping sight distance is calculated using equations provided in the Green Book based on design speed and grade and assumed values of perception-reaction time and deceleration rate. Table 43 pro- vides minimum stopping sight distances for various roadway design speeds and grades. The stopping sight distance criteria shown in Table 43 apply to all roadway types, including ramps and turning roadways. A design exception is required where stopping sight distances less than those shown in Table 43 are provided or retained. Stopping sight distance generally provides drivers with enough distance to make a hurried stop, but these distances may not be adequate for a driver to interpret complex informa- tion or make a complex decision. In some cases, a maneuver other than a quick stop would be preferable, but would require more time for the driver to make that decision. For these rea- sons, the Green Book also provides decision sight distance guidelines for several different avoidance maneuver conditions that each assumes a different perception and reaction time. The decision sight distance criteria are presented in Green Book Table 3-3 (not shown here). Decision sight distance is not part Table 41. Passenger-car equivalents for RVs on upgrades for specific grade segments on freeways (13). Upgrade (%) Length (mi) ER Percentage of RVs 2% 4% 5% 6% 10% 15% 20% 25% ≤ 2 All 1.2 1.2 > 2 to 3 0.00 to 0.50 1.2 1.2 > 0.50 3.0 1.2 > 3 to 4 0.00 to 0.25 1.2 1.2 > 0.25 to 0.50 2.5 1.2 1.2 1.5 1.2 2.5 1.2 1.2 1.5 1.2 2.0 1.2 1.2 1.5 1.2 2.0 8% 1.2 1.2 1.5 1.2 2.0 1.2 1.2 1.5 1.2 2.0 1.5 1.2 1.2 1.2 1.2 1.5 1.2 1.2 1.2 1.2 1.5 SOURCE: Based on HCM Exhibit 11-12 (abridged). Table 42. Passenger-car equivalents for trucks and buses on downgrades on specific grade segments on freeways (13). Downgrade (%) Length (mi) ET Percentage of trucks 5% 10% 15% 20% < 4 All 1.5 1.5 1.5 1.5 4 to 5 ≤ 4 1.5 1.5 1.5 1.5 4 to 5 > 4 2.0 2.0 2.0 1.5 > 5 to 6 ≤ 4 1.5 1.5 1.5 1.5 SOURCE: Based on HCM Exhibit 11-13 (abridged). Table 43. Design criteria for stopping sight distance (4, 5). Design speed (mph) Stopping sight distance (ft) Level Downgrade Upgrade 0% 3% 6% 9% 3% 6% 9% 15 80 80 82 85 75 74 73 20 115 116 120 126 109 107 104 25 155 158 165 173 147 143 140 30 200 205 215 227 200 184 179 35 250 257 271 287 237 229 222 40 305 315 333 354 289 278 269 45 360 378 400 427 344 331 320 50 425 446 474 507 405 388 375 55 495 520 553 593 469 450 433 60 570 598 638 686 538 515 495 65 645 682 728 785 612 584 561 70 730 771 825 891 690 658 631 75 820 866 927 1003 772 736 704 80 910 965 1035 1121 859 817 782 SOURCE: Based on AASHTO Green Book Tables 3-1 and 3-2.

36 of the controlling criteria; no design exceptions are required for decision sight distances less than the Green Book criteria. The HCM does not include any effect of stopping sight dis- tance on LOS for any roadway type. Green Book criteria for stopping sight distance assume that vehicles on a crest vertical curve, or in a region of restricted horizontal sight distance, are traveling at the design speed. There does not appear to be any basis on which to presume that limited stopping sight distance, especially marginal limitations, affects vehicle speeds or other traffic operational performance measures. Research by Fambro et al. (36) found very few collisions on highways with objects smaller than another vehicle, even in areas of limited stopping sight distance. This led to the change in stopping sight distance from a 6-in. object to a 2-ft object (equivalent to the height of vehicle taillights) that was made in the 2001 edition of the Green Book (3). Thus, available research suggests that at most places on the highway with lim- ited stopping sight distance there is unlikely to be anything in the roadway that a driver might strike. Safety is unlikely to be affected by limited stopping sight distance in such cases. How- ever, when the limited sight distance restricts the driver’s view of a location where other vehicles may be slowing or stopping (e.g., intersections, driveways, horizontal curves, entrance or exit ramps, or locations with daily congestion), improving limited sight distance may be very important to safety. Neither the HSM nor the FHWA CMF Clearinghouse includes any CMFs indicating an effect of stopping sight dis- tance on safety. Research conducted under NCHRP Project 17-53 (see Section 4.7) investigated the relationship between stopping sight distance and crash frequency. The research team compared the crash frequencies for crest vertical curves on rural two-lane highways with stopping sight distance less than AASHTO stopping sight distance criteria to crest ver- tical curves with stopping sight distance equal to or more than AASHTO stopping sight distance criteria. A statistical analysis found no differences in crash frequency (either for total crashes or fatal-and-injury crashes) between the crest vertical curves with differing stopping sight distance values, but there was a statistically significant difference in crash fre- quency (for both total crashes and fatal-and-injury crashes) between sites with and without horizontal curves, intersec- tions, or driveways hidden by the presence of the crest vertical curve. The observed effect on crash frequency of the presence of a hidden horizontal curve, intersection, or driveway was 0.36 crashes per mi per year for total crashes and 0.48 crashes per mi per year for fatal-and-injury crashes. Mitigation strategies for limited stopping sight distance include the following (7): • Signing for crest vertical curves • Lighting for intersections, sag vertical curves, or merge/ diverge areas • Lower height barriers to reduce sight distance limitations due to presence of the barrier • Adjustment of lane placement within the roadway cross section on horizontal curves • Selection of cross-sectional elements to manage speed • Wider shoulders and wider clear zones • Static or dynamic warning of intersections or entering traffic • Repositioning, adding, or enhancing intersection signs 2.10 Cross Slope The controlling criterion for cross slope addresses the tra- verse slope of the pavement surface on tangent sections or on horizontal curves where superelevation is not used. Super- elevation on horizontal curves is addressed in Section 2.11. The cross-slope design criterion is important because cross slope facilitates runoff of water from rain, snow, or ice from the pavement surface. In general, the steeper the cross slope, the more efficiently water flows to the edge of the lanes and off the roadway. Flat cross slopes can lead to water ponding on the lanes, especially where a curb is used. At the same time, a steep cross slope can affect steering and can make vehicles more susceptible to cross winds; drivers may tend toward the lower edge of the traveled way, and lateral skidding can become more likely when braking on wet or icy pavement. On road- ways with a center crown, vehicles making passing maneuvers experience double the change in cross slope as they move over the crown, reversing the direction of lateral acceleration, and potentially causing trucks to sway from side to side. For these reasons, a balance must be struck between a steeper cross slope that efficiently moves water to the edge of the roadway and a shallow cross slope that is imperceptible to drivers during lane changes. The Green Book recommends a normal cross slope of 1.5 to 2 percent, although when two or more lanes are inclined in the same direction, each successive lane may be given a greater cross slope by 0.5 to 1.0 percent, not to exceed 4 percent in the outermost lanes. In areas of intense rainfall, a slope of 2.5 percent may be used. The National Transportation Safety Board (NTSB) has asked FHWA and AASHTO to inves- tigate the appropriateness of design criteria for cross-slope breaks at the outside edge of the traveled way on horizontal curves for current passenger cars and trucks, especially trucks with high centers of gravity (37). The research underlying the current 8-percent design criterion for cross-slope breaks was completed in 1982 using an older vehicle dynamics simulation model (HVOSM) that simulated cross-slope break traversals by a 1971 Dodge Coronet passenger car (38). Research for a current passenger car and larger trucks, including trucks with high centers of gravity, would clearly be desirable. Neither the HCM nor the HSM shows any qualitative effect of cross-slope or cross-slope breaks on traffic operations or

37 safety. There are also no safety effects found in the FHWA CMF Clearinghouse. The primary concern for locations with insufficient cross slope is inadequate drainage and ponding of water on the travel lanes. Mitigation strategies for inadequate cross slope include the following (7): • SLIPPERY WHEN WET signing • Grooved, textured, or open-graded pavements to improve surface friction • Slope inside lanes toward the median and outside lanes toward the outside of the roadway (on multilane divided facilities) Mitigation strategies for large pavement/shoulder cross slope breaks include the following: • Adjustment of the high-side shoulder cross slope, including sloping the shoulder toward the traveled way • Rounding of the cross-slope break (feasible for hot-mix asphalt pavements) 2.11 Superelevation The Green Book provides equations and tables for deter- mining the appropriate superelevation rate for specific horizontal curves based on the design speed, curve radius, and assumed maximum values of superelevation rate and friction demand. Maximum superelevation rates (emax) are selected by highway agency policies; Green Book Chapter 3 permits highway agencies to choose emax in the range of 4 to 12 percent. Where snow and ice are factors, the Green Book recommends that superelevation should not exceed 8 percent. For lower speed urban arterials, the Green Book recommends that little or no superelevation be used. Green Book Chapter 8 recommends that superelevation should not exceed 6 percent on freeways with viaducts where snow and ice are factors. Neither the HCM nor any other available source indicates that superelevation has a quantifiable effect on traffic opera- tions. It seems unlikely that minor variations in supereleva- tion from the AASHTO design values would have much effect on traffic operations. HSM Chapter 10 (Rural Two-Lane Highways) presents a CMF for superelevation on rural two-lane highways that is shown in the following equations: CMF 1.00 for SV 0.01 (36)= < CMF 1.00 6 SV 0.01 for 0.01 SV 0.02 (37)( )= + × − ≤ < CMF 1.06 3 SV 0.02 for SV 0.02 (38)( )= + × − ≥ where CMF = crash modification factor for the effect of super- elevation variance on total crashes SV = superelevation variance (ft/ft), which represents the superelevation rate contained in the Green Book minus the actual superelevation of the curve The CMF applies to total roadway segment crashes for road- way segments located on horizontal curves. No CMFs are available and no trends are known for the safety effects of superelevation on roadway types other than rural two-lane highways. The mitigation strategies for superelevation lower than Green Book criteria are the same as those described for hori- zontal alignment in Section 2.6.5 of this report. 2.12 Vertical Clearance In general, vertical clearance does not affect operations on the roadway other than for those vehicles that are taller than the available vertical clearance allows for. When overpasses or other structures do not allow for taller vehicles to pass underneath, these vehicles use an alternate route, potentially increasing travel time. Guidance for vertical clearance is pro- vided in the Green Book as follows: • For rural arterials, the recommended minimum vertical clearance is 16 ft • The preferred vertical clearance on urban arterials is 16 ft; however, when existing structures offer at least 14 ft of clearance, these structures may be retained as long as an alternate route with 16 ft of clearance is provided • The recommended minimum vertical clearance on free- ways is 16 ft; however, in highly developed areas, where replacement of structures would be costly, a minimum clearance of 14 ft is permitted, provided an alternate route with 16 ft of clearance is available. Sign trusses and pedes- trian overpasses should be built with a minimum clearance of 17 ft. There are no operational or safety effects of insufficient vertical clearance except for increased travel times for vehicles taller than the available vertical clearance. Vertical clearance guidelines do not directly impact safety for the majority of vehicles, although in cases where the rec- ommended vertical clearance is not provided, advanced warning and alternate route designation become important mitigation strategies for avoiding possible crashes involving tall vehicles. Vertical clearance crashes can have severe impacts

38 on operations by damaging overpasses or other structures that result in extended road closures. Special attention is given to vertical clearance on Interstate freeways to maintain the integrity of the system for national defense purposes. On rural Interstate freeways, vertical clear- ance at structures of at least 16 ft is maintained. In urban areas, 16 ft of clearance is maintained for at least one Inter- state routing through the urban area, with other urban Inter- state routes having vertical clearance of at least 14 ft. The 16-ft vertical clearance for Interstate freeways in rural areas and for the single routing in urban areas applies to the entire roadway width, including the usable shoulder width and the ramps and collector-distributor roadways at Interstate-to-Interstate interchanges. 2.13 Horizontal Clearance/ Lateral Offset The controlling criterion known in current FHWA policy as horizontal clearance has been renamed lateral offset in the 2011 edition of the Green Book (5) to avoid confusion about the definition of this criterion. Lateral offset deals with the distance from the edge of the traveled way, face of curb, shoulder, or other designated point to a vertical roadside element or obstruction (7). Lateral offset can be thought of as an operational offset; vertical roadside elements are offset (1) so that they do not affect a driver’s speed or lane position and (2) so that adequate clearance to vertical roadside elements is provided for overhangs or mirrors of trucks and buses and for opening curbside doors where on-street parking is provided. Lateral offset as a controlling criterion is primarily of inter- est for roads with curb-and-gutter sections, such as urban and suburban arterials. For roads without curbs, the mini- mum shoulder widths generally take care of providing a min- imum lateral offset from the traveled way. Design criteria in the 2004 Green Book (4) specify a mini- mum lateral offset of 1.5 ft to address operational concerns for all roadway conditions and classifications. The 2011 Green Book (5) does not state an explicit lateral offset, but makes reference to the AASHTO Roadside Design Guide (RDG) (39). The 2006 edition of the RDG (39), as well as previous edi- tions, incorporated the same 1.5-ft lateral offset as the 2004 Green Book (4). The 2011 edition of the RDG (40) encour- ages wider lateral offsets, particularly on urban and suburban arterials (see Section 2.13.3 below). A design exception is required when the specified mini- mum lateral offset is not provided. It is important to note that the controlling criterion for lateral offset does not include the provision of clear recovery zones. Lateral offset is an opera- tional criterion and, as explicitly stated by FHWA policy, does not address clear-zone width (2). 2.13.1 Rural Two-Lane Highways Design Criteria Relatively few rural two-lane highways have curb-and-gutter sections, so the minimum shoulder-width criteria generally provide the minimum lateral offset needed for operational reasons. Traffic Operational Effects Chapter 15 (Two-Lane Highways) of the HCM provides guidance for estimating the free-flow speed for two-lane high- ways. Although the LOS boundaries are not directly adjusted for lateral clearance, Table 5 provides an adjustment to free-flow speed based on lane and shoulder widths. As shown in Table 5, a 6-ft shoulder on a rural two-lane highway provides sufficient lateral clearance that there is no effect on vehicle speeds. Traffic Safety Effects Chapter 10 (Rural Two-Lane Highways) of the HSM does not contain any CMF for lateral offset. However, the CMF for shoulder width on two-lane highway segments presented in Table 13 and Figure 4 implicitly reflects, at least in part, the safety effects of lateral offset. 2.13.2 Rural Multilane Highways Design Criteria Relatively few rural multilane highways have curb-and- gutter sections, so the minimum shoulder-width criteria generally provide the minimum lateral offset needed for oper- ational reasons. Traffic Operational Effects Chapter 14 (Multilane Highways) of the HCM provides guidance for estimating free-flow speed for multilane high- ways. Although the LOS boundaries are not directly adjusted for lateral clearance, Table 16 provides an adjustment to free- flow speed based on the sum of the lateral clearance on the left side of the roadway (maximum of 6 ft) and the right side of the roadway (maximum 6 ft). Traffic Safety Effects Chapter 11 (Rural Multilane Highways) of the HSM does not contain any CMF for lateral offset. However, the CMF for shoulder width in Table 13 and Figure 4 for undivided roadways and in Table 17 for divided roadways implicitly reflects, at least in part, the safety effects of lateral offset.

39 2.13.3 Urban and Suburban Arterials Design Criteria The design criterion for lateral offset on urban and sub- urban arterials in the 2006 RDG (39), and previous editions, is 1.5 ft. The 2011 RDG (40), which is referred to explic- itly in Chapter 7 (Arterials) of the 2011 Green Book (5), states that a lateral offset of 3 ft from the face of the curb to obstructions should be provided at intersections and drive- way openings, while a minimum lateral offset of 1.5 ft should be used elsewhere. However, the new RDG also presents a targeted design approach for high-risk urban roadside corridors: • For locations with vertical curbs, provide a 6-ft offset from the face of curb to obstacles on the outside of curves, because obstacles on the outside of curves are hit more often, and provide a 4-ft offset elsewhere • For locations without a vertical curb, 12-ft offsets to obstacles on the outside of curves and 8-ft offsets on tangent sections are recommended as reasonable goals where the clear-zone widths in RDG Chapter 3 cannot be achieved. Traffic Operational Effects Chapter 17 (Urban Street Segments) of the HCM includes a procedure for estimating free-flow speeds, but neither lateral offset nor shoulder width is considered as part of that procedure. Traffic Safety Effects Chapter 12 (Urban and Suburban Arterials) of the HSM does not include a CMF for either lateral offset or shoulder width. There is currently no quantifiable safety effect for these design elements. 2.13.4 Freeways Design Criteria Lateral offset is not generally relevant on freeways because minimum shoulder widths should always provide the mini- mum lateral offset from the traveled way. Traffic Operational Effects Chapter 11 (Basic Freeway Segments) of the HCM includes criteria for estimating the effect of shoulder width on free- flow speed (see Table 19). Traffic Safety Effects There are no CMFs for lateral offset on freeways, as freeway shoulders are usually wide enough to provide the minimum lateral offset. The results of NCHRP Project 17-45 include a CMF for right (outside) clearance (25). This is essentially a CMF for clear-zone width on freeways, which incorpo- rates an adjustment for right (outside) shoulder width. The NCHRP Project 17-45 methodology also includes CMFs for right (outside) roadside barriers on freeways. Neither of these CMFs appears applicable to lateral offset on freeways because the shoulder-width CMFs from NCHRP Project 17-45, pre- sented in Equations 8 through 11, should account for the effect of lateral offset on safety. 2.13.5 Mitigation Strategies The primary mitigation strategy for lateral obstructions within the minimum lateral offset that cannot practically be removed is to delineate such obstacles with reflectors or reflective sheeting so that they become more visible, particu- larly at night (7). 2.14 Summary of Traffic Operational Effects Table 44 summarizes which traffic operational effects for the 13 controlling criteria have been quantified and where in this report the information concerning each of those known effects can be found. 2.15 Summary of Traffic Safety Effects Table 45 summarizes which traffic safety effects for the 13 controlling criteria have been quantified and where in this report the information covering each of those known effects can be found.

40 Design criterion Roadway type Traffic operational effects Design speed All No direct effects.a Lane width Rural two-lane highways See Table 5 (based on HCM Exhibit 15-7) and Equation 1. Rural multilane highways See Table 7 (based on HCM Exhibit 14-8) and Equation 3. Urban and suburban arterials No quantified effects. Freeways See Table 10 (based on HCM Exhibit 11-8) and Equation 4. Shoulder width Rural two-lane highways See Table 5 (based on HCM Exhibit 15-7) and Equation 1. Rural multilane highways See Table 16 (based on HCM Exhibit 14-9) and Equation 3. Urban and suburban arterials No quantified effects. Freeways See Table 19 (based on HCM Exhibit 11-9) and Equation 4. Bridge width Rural two-lane highways Bridge roadway widths less than the approach roadway width do not appear to increase crash frequency or severity. Rural multilane highways No quantified effects directly applicable to bridge width; related effects for lane and shoulder width are known (see Sections 2.2.2 and 2.3.2). Urban and suburban arterials No quantified effects. Freeways No quantified effects directly applicable to bridge width; related effects for lane and shoulder width are known (see Sections 2.2.4 and 2.3.4). Structural capacity All No relationship to traffic operations; controlling criterion is based on risk of structural failure. Horizontal alignment Rural two-lane highways See Table 21. Rural multilane highways See Equation 18. Urban and suburban arterials See Equation 21. Freeways No quantified effects. Vertical alignment (sag vertical curves) Rural two-lane highways No quantified effects. Rural multilane highways No quantified effects. Urban and suburban arterials No quantified effects. Freeways No quantified effects. Grade Rural two-lane highways See Tables 23 through 31 (based on HCM Exhibits 15-9, 15-10, 15-11, 15- 12, 15-13, 15-16, 15-17, 15-18, 15-19) and Equation 34. Rural multilane highways See Tables 33 through 36 (based on HCM Exhibits 14-12, 14-13, 14-14, 14-15) and HCM Equation 14-4. Urban and suburban arterials No quantified effects. Freeways See Tables 39 through 42 (based on HCM Exhibits 11-10, 11-11, 11-12, 11-13) and HCM Equations 11-2, 11-3, and 11-4. Stopping sight distance All No quantified effects. Cross slope All No quantified effects. Superelevation All No quantified effects. Vertical clearance All No quantified effects. Horizontal clearance/lateral offset Rural two-lane highways Effect discussed in shoulder-width section (see Section 2.3.1). Rural multilane highways Effect discussed in shoulder-width section (see Section 2.3.2). Urban and suburban arterials No quantified effects. Freeways Effect discussed in shoulder-width section (see Section 2.3.4) a For indirect effects, see lane width, horizontal alignment, vertical alignment, and stopping sight distance. Table 44. Summary of traffic operational effects of the 13 controlling criteria for design.

41 Design criterion Roadway type Traffic safety effects Design speed All roadway types No direct effects.a Lane width Rural two-lane highways See Equation 2 and Table 6 (based on HSM Equation10-11 and Table 10-8). Rural multilane highways For undivided sections, see Equation 2 and Table 8 (based on HSM Equation 11-13 and Table 11-11); for divided sections, see Equation 2 and Table 9 (based on HSM Equation 11-16 and Table 11-16). Urban and suburban arterials Lane width does not appear to affect crash frequency or severity. Lanes narrower than 12 ft may not be desirable on streets where substantial volumes of bicycles, trucks, or buses are present. Rural freeways See Equations 5 and 6. Shoulder width Rural two-lane highways See Equation 7 and Tables 13 and 14 (based on HSM Equation10-12 and Table 10-9 and 10-10). Rural multilane highways For undivided sections, see Equation 7 and Tables 13 and 14 (based on HSM Equation10-12 and Table 10-9 and 10-10); for divided sections, see Table 17 (based on HSM Table 11-17). Urban and suburban arterials No quantified effects. Freeways See Equations 8 through 13. Bridge width Rural two-lane highways No quantified effects directly applicable to bridge width; related effects for lane and shoulder width are known (see Sections 2.2.1 and 2.3.1). Rural multilane highways No quantified effects directly applicable to bridge width; related effects for lane and shoulder width are known (see Sections 2.2.2 and 2.3.2). Urban and suburban arterials No quantified effects. Freeways No quantified effects directly applicable to bridge width; related effects for lane and shoulder width are known (see Sections 2.2.4 and 2.3.4). Structural capacity All roadway types No relationship to traffic safety; controlling criterion is based on risk of structural failure. Horizontal alignment Rural two-lane highways See Equation 15 (based on HSM Equation 10-13); potential updated effects are presented in Equations 16 and 17. Rural multilane highways See Equations19 and 20. Urban and suburban arterials No quantified effects. Freeways See Equations 22 through 25. Vertical alignment (sag vertical curves) Rural two-lane highways See Equations 30 through 33. Rural multilane highways No quantified effects. Urban and suburban arterials No quantified effects. Freeways No quantified effects. Grade Rural two-lane highways See Table 32 (based on HSM Table 10-11) and Equation 35; potential updated effects are presented in Equations 16 and 17. Rural multilane highways No quantified effects. Urban and suburban arterials No quantified effects. Freeways No quantified effects. Stopping sight distance Rural two-lane highways No effect on safety unless a hidden horizontal curve, intersection, or driveway is present. Rural multiline highways No quantified effects. Urban and suburban arterials No quantified effects. Freeways No quantified effects. Cross slope All roadway types No quantified effects. Superelevation Rural two-lane highways See Equations 36 through 38 (based on HSM Equations 10-14 through 10-16). Rural multilane highways No quantified effects. Urban and suburban arterials No quantified effects. Freeways No quantified effects. Vertical clearance All roadway types No quantified effects. Horizontal clearance Rural two-lane highways Only known effects are based on shoulder width (See Section 2.3.1). Rural multilane highways Only known effects are based on shoulder width (See Section 2.3.2). Urban and suburban arterials No quantified effects. Freeways Only known effects are based on shoulder width (See Section 2.3.4). a For indirect effects, see lane width, horizontal alignment, vertical alignment, and stopping sight distance. Table 45. Summary of traffic safety effects for the 13 controlling criteria for design.

Next: Section 3 - Design Exception Practices »
Evaluation of the 13 Controlling Criteria for Geometric Design 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 783: Evaluation of the 13 Controlling Criteria for Geometric Design describes the impact of the controlling roadway design criteria on safety and operations for urban and rural roads.

  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!