Guidelines for Ramp and Interchange Spacing (2011)

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.

Chapter 4 Operational and Safety Considerations This chapter provides an overview of the various operational considerations and elements that affect choices and decisions about ramp and interchange spacing. In general, traffic operations and safety evaluations should be included as an integral part of the initial geometric evaluations of potential ramp configurations. 4.1 TRAFFIC OPERATIONS OVERVIEW Professionals need to conduct an appropriate level of traffic operations analysis commensurate with the stage of the project development process (planning, location, or final design) to support ramp and interchange spacing decisions. The analysis should be consistent with the data available, and data should be consistent with the analysis tools applied. Operational considerations should begin at the earliest stage of project development and integrated with the geometric design considerations. This means evaluating lane numbers and arrangements along the highway and ramp series and considering the types of analyses that should be performed (i.e., ramp merge and diverge, mainline capacity, and weaving sections). In addition, this means considering ramp terminal intersection operations and understanding how predicted operations (lane numbers, arrangements, queuing, deceleration, and stopping sight distance to the back of queue) may influence the ramp and interchange layout. Operational analysis for basic planning applications can likely be conducted with limited data that is often available during the planning stages of a project. Hourly ramp and freeway volumes, operating speeds, lane numbers, and ramp configurations should be assessed to guide geometric design decisions. The Highway Capacity Manual ( HCM) and the ITE Freeway Handbook provide planning-level operational analysis procedures. Many state highway agencies provide general guidance concerning ramp and freeway service volumes and the number and arrangements of needed lanes. This guidance can be used to aid design decisions as ramp and interchange configurations are being developed. Some basic freeway-related capacity thresholds from the HCM are shown in Table 4-1. 41

Table 4-1 Approximate Capacity of Freeway-related Roadway Elements, 2010 HCM (5) The ITE Freeway Handbook also provides a tool for assessing merges and diverges. For both entry ramps and exit ramps, the handbook includes charts that require only freeway volume, ramp volume, and number of lanes as an input to determine if a ramp-freeway junction is below, near, or over capacity. Planning-level operational considerations can influence ramp and interchange configurations and, therefore, ramp spacing values. 4.2 HIGHWAY CAPACITY MANUAL PROCEDURES These Guidelines are not intended to replace the use of the HCM in the interchange planning and design process. Instead, they are intended to bring traffic operations considerations into the planning process at an early stage to consider the operational considerations and associated influence on ramp spacing values. The HCM addresses interchange and ramp spacing and density in discussions related to basic freeway segments, weaving segments, and ramp-freeway junctions. On a basic freeway segment, estimated free-flow speed decreases as the number of ramps per mile (total ramp density) increases. In a weaving section, speeds and level of service (LOS) decrease as the segment shortens. At a ramp-freeway junction, the presence of an adjacent ramp influences the density and LOS in some cases. An overview of 2010 HCM procedures from the three chapters most relevant to ramp and interchange spacing are presented in the following subsections. Although not detailed in these Guidelines, service interchange ramp terminal intersections should also be evaluated using HCM and other traffic engineering procedures to understand how ramp terminal intersection operations influence lane numbers and arrangements and queue lengths. Three HCM chapters are most relevant to ramp and interchange spacing: basic freeway segments, freeway weaving segments, and freeway merge and diverge segments. Professionals could include total ramp density as a factor in evaluating interchange forms and ramp configurations. Element Service Volume Freeway Lane 2,250 – 2,400 passenger cars per hour Single-Lane Ramp* 1,800 to 2,200 passenger cars per hour Merge Influence Area (on-ramp plus right two lanes of freeway) 4,600 passenger cars per hour Diverge Influence Area (off-ramp plus right two lanes of freeway) 4,400 passenger cars per hour * Basic ramp segment only, does not consider ramp terminal operations. 42 Guidelines for Ramp and Interchange Spacing

These can influence overall ramp design configurations and ramp spacing values. 4.2.1 Basic Freeway Segments The mainline, or basic, freeway segment occurs between ramp merge and diverge areas and can include basic lanes, auxiliary lanes, or high-occupancy vehicle lanes. Free-flow speed (FFS) is the performance metric for basic freeway segments, and is defined as the average speed of passenger cars on a uniform freeway segment with moderate volume. Interchange spacing and ramp density influences a freeway’s estimated FFS. FFS on a basic freeway segment decreases as more ramps are added. However, the spacing between the ramps does not impact estimated FFS. FFS may be measured or estimated with equation 4.1 (5): FFS = 75.4 – fLW – fLC – 3.22TRD 0.84 To predict FFS, the HCM assumes a base free-flow speed of 75.4 mph and applies reductions based upon lane width (fLW), right-side lateral clearance (fLC), and total ramp density (TRD). TRD is expressed in terms of ramps per mile and is measured over a six-mile segment of freeway—three miles upstream and three miles downstream of the point on the freeway being studied. Both onramps and offramps are included. The researchers who developed the FFS prediction equation considered including interchange density, onramp density, and offramp density instead of TRD, but ultimately found that TRD best predicted FFS (25). Exhibit 4-1 shows the impact of ramp density on FFS using the prediction model in the 2010 HCM. A freeway with three-mile diamond interchange spacing would have two ramps every three miles, or 0.67 ramps per mile. This ramp density will decrease predicted FFS by 2.3 miles per hour compared to an “ideal” six-mile section of freeway that has no ramps. A freeway with one-mile diamond interchange spacing, as is common in many urban areas, will have two ramps per mile and decrease predicted FFS by 5.8 mph in comparison to an ideal segment. One-mile spacing of full cloverleaf interchanges (four ramps per mile) would decrease predicted FFS by 10.3 mph. Complete HCM analyses should be conducted as applicable reflecting the relative accuracy of the level of engineering detail available while using the analysis results to guide and influence geometric design considerations. According to the HCM, the number of ramps impacts FFS, but ramp spacing does not. Total ramp density has the greatest impact on FFS. Operational and Safety Considerations 43

Exhibit 4-1 Impact of Total Ramp Density on Basic Freeway Segment Free- Flow Speed (5). 4.2.2 Freeway Merge and Diverge Segments Merge and diverge areas exist at ramp-freeway junctions at which a lane is not added or dropped. Operationally, it is desirable to separate merge and diverge influence areas to the extent possible so the combination or overlap of these two areas does not cumulatively degrade mainline performance. Therefore, ramp and interchange spacing dimensions can be affected by attempting to separate or reduce the overlap of these influence areas. Exhibit 4-2 schematically presents these conditions. Exhibit 4-2 Merge and Diverge Influence Areas Merge and diverge areas are limited by definition to the right two lanes of the freeway and the associated acceleration or deceleration lane because studies have shown that this is where most turbulence occurs (5, 23). The methodology of the HCM identifies influence areas to be approximately 44 Guidelines for Ramp and Interchange Spacing

1,500 ft in length. However, certain designs could increase this length. For example, two-lane ramps which are developed with auxiliary lanes may exert an influence on the freeway as far upstream/downstream as the auxiliary lane extends. In moderate- to high-volume freeway or ramp conditions, increased ramp spacing may be considered to separate merge and diverge overlap areas. In locations of extreme volumes and low capacities, the ramp spacing may have little influence on traffic operations since the individual merge and diverge condition or freeway mainline capacity limitation may dominate the influence of ramp spacing values. The 2010 HCM provides separate merge and diverge procedures for four-, six-, and eight-lane freeways. On six-lane freeways, merge segment operations improve as the distance to an adjacent downstream offramp increases or as the distance to an adjacent upstream offramp decreases. Also, on a six-lane freeway, diverge segment operations improve as the distance to an upstream onramp or a downstream offramp increases. The following two paragraphs discuss the reasons for this. Although these basic relationships would intuitively apply to all freeways, data used to create the HCM procedures only indicated this trend on six-lane freeways. As a result, HCM merge and diverge analysis procedures only take adjacent ramps into consideration on six-lane freeways. The HCM does not specify an upper boundary for what is considered “adjacent,” but, in general, the HCM models are not sensitive to adjacent ramps that are more than one mile apart regardless of volume. For a merge, a greater distance to a downstream exit ramp results in fewer vehicles in the merge influence area (fewer vehicles have changed to right side lane in preparation for the exit) and improved ramp-freeway junction operations. Likewise, a shorter distance to an upstream exit ramp also results in fewer vehicles in the merge influence area and improved operations. In the case of the downstream exit ramp, the volume of the ramp plays a role in merge influence area operations (greater volume decreases operational performance) in addition to the distance to the ramp (5, 26) For a diverge, a greater distance to an upstream entrance ramp or a greater distance to a downstream exit ramp will result in fewer vehicles in the diverge influence area and thus improve operations. In both cases, greater volume on the adjacent ramp also decreases operational performance. 4.2.3 Weaving Segments Weaving segments are formed when a merge area is closely followed by a diverge area and the resulting lane configuration requires two or more traffic streams to cross. The means of determining what is “close” are discussed in the following paragraphs. The length of the weaving segment (the spacing between the entry and exit ramp) is one of many factors that determines the For six-lane freeways, HCM merge and diverge procedures are sensitive to adjacent ramps, and therefore, ramp spacing values become a key operational consideration. Ramp spacing is one of many factors that determine weaving segment LOS. Target LOS can influence ramp spacing dimensions. Operational and Safety Considerations 45

operation of the segment. Generally speaking, the speeds and LOS within a weaving segment decrease as the segment shortens. Historically, many weaving segments were between loop ramps and the length of a weaving section was defined in terms of criteria specific to loop ramp design (27). This length was measured from a point at the merge gore where the right edge of the highway shoulder lane and left edge of the merge lanes are 2 ft apart to a point at the diverge gore where the two edges are 12 ft apart. The maximum length for which the weaving analysis was conducted was 2,500 ft. Beyond this distance, the merge and diverge areas were considered separately regardless of traffic volume or other factors. Because this description was applied for many years, a number of transportation agencies may still include this method of measuring weaving length in their documentation. In reality many weaving segments today are between adjacent interchanges rather than loop ramps within the same interchange. Research conducted in the past few years and incorporated into the 2010 HCM focused on weaving between non-loop ramps and considered several ways of defining weaving segment length (27). Exhibit 4-3 and Table 4-2 present these different measurements, and the corresponding AASHTO event points. Exhibit 4-3 Definitions of Weaving Segment Length (4, 5) Table 4-2 Design Event Points in Resource Documents Description HCM AASHTO The distance between the end points of any barrier markings that prohibit or discourage lane changing. Short length “painted tip” or “painted noses” The distance between points in the respective gore areas where the left edge of the ramp travel lanes and the right edge of the highway travel lanes meet. Base length “painted tip” or “painted noses” The distance between physical barriers marking the ends of the merge and diverge gore areas. Long length “physical noses” These Guidelines and the 2010 HCM use a different definition of weaving segment length than past editions of the HCM 46 Guidelines for Ramp and Interchange Spacing

Of the three potential definitions of weaving segment length depicted above, the researchers developing the weaving procedures for the 2010 HCM found the “short length” best predicted weaving operations (27). The short length is measured between the end points of barrier markings (such as solid white stripes) that prohibit or discourage lane changes. This definition is different than the one used in these Guidelines. These Guidelines define ramp spacing as the distance between painted gores, which is shown in Exhibit 4-3 as the “base length.” If barrier markings do not exist, then short length is measured between the painted gores. In such a case, the spacing definition in the HCM and these Guidelines is the same. To provide planning-level tools to understand how weaving may or may not influence ramp spacing decisions, these Guidelines contain some high-level aids that provide insights about the presence of weaving using procedures from the HCM. To determine if ramps are “close enough” for a weaving segment to exist, one must count the number of lanes from which a weaving maneuver may be made with one or no lane changes (NW). In order to have weaving, there will always be two lanes (NW = 2). For major weaves, NW will equal two or three. Exhibit 4-4 depicts a weaving section where one of the weaving movements requires two lane changes. Exhibit 4-4 Weaving Segment With Two Lane Changes for One of the Weaving Movements If ramps get “far enough” apart, the freeway segment between them will operate as a basic segment rather than a weaving section. Exhibit 4-5, which was developed from the HCM equation that determines maximum weaving segment length, can be used to check if this condition exists. For example, if an entry ramp and exit ramp with an auxiliary lane are spaced 2,000 ft apart and there are 800 weaving vehicles out of 2,000 total vehicles in the section (ratio of 0.4), an HCM weaving analysis should be performed. Operational and Safety Considerations 47

Exhibit 4-5 Maximum Weaving Segment Length. Adapted from (5). If Exhibit 4-5 indicates that a weaving segment exists for the location under study, then a full weaving analysis should be conducted using the HCM’s methodology to evaluate the weaving segment operations. This aid can be used at the earliest planning level to understand the range of traffic operations analyses that might be needed to confirm ramp spacing decisions. 4.3 OTHER PLANNING-LEVEL OPERATIONAL GUIDELINES The HCM provides analysis procedures for many common ramp and interchange forms and designs, but not all possible situations. In particular, the HCM provides little guidance on how the spacing between two ramps impacts freeway speed. In developing these Guidelines, this relationship was HCM. 4.3.1 Simulation Modeling Simulation modeling of four freeway lanes in each direction (eight lanes total) with consecutive entry ramps and an entry ramp followed by an exit ramp (without an auxiliary lane) found that ramp spacing usually had little impact on freeway speeds at low to moderate freeway volumes (1,500 vehicles per hour per lane (vphpl) or less). Entry-exit ramp combinations when exit-ramp modeling scenarios, ramp spacing did have a significant impact (up to 15 mph) on freeway speed at low to moderate freeway volumes. At higher freeway volumes (1,750 vphpl), decreased spacing between ramps had a significant impact on freeway speeds. For entry-exit ramp combinations the impact was up to 15 mph, and for entry-entry ramp combinations the impact was up to 10 mph. Spacing impacts of entry-exit ramp combinations are shown in Exhibit 4-6. Planning-level tools can aid professionals in the types of operational analysis that might be needed to support informed ramp and interchange spacing decisions. investigated through simulation modeling and the limited information in the volumes are near capacity (1,750 vphpl) are an exception to this. During these 48 Guidelines for Ramp and Interchange Spacing

Exhibit 4-6 Impact of EN-EX Ramp Spacing on Freeway Speed. “Major” Impact defined as 5+ MPH In general, entry-exit ramp combinations have a greater impact on freeway speed than entry-entry ramp combinations. Also, for entry-exit ramp combinations, increased exit ramp volume has a greater impact on freeway speed than increased entry ramp volume. The HCM provides a procedure for analyzing closely spaced entry-exit ramp combinations with auxiliary lanes (weaving segments), but does not provide information on the performance of a weaving segment in comparison to a closely spaced entry-exit combination without an auxiliary lane. Simulation modeling identified that adding an auxiliary lane between an entry ramp and an exit ramp is operationally beneficial. Regardless of ramp spacing, adding an auxiliary lane generally improved freeway speed by 5 mph or more if at least one of the ramps had moderate to near-capacity volume (1,500-1,750 vphpl) as shown in Exhibits 4-7 and 4-8. Ramp spacing generally has little influence on freeway speeds when the freeway has low to moderate volumes and the entrance and exit ramps operate below capacity. Decreased ramp spacing generally has a significant impact on freeway speeds when the freeway is operating with high volumes. Operational and Safety Considerations 49

Exhibit 4-7 Benefit of Auxiliary Lane on Freeway Speed with 1000’ ramp spacing. “Major” Benefit defined as 5+ MPH Exhibit 4-8 Benefit of Auxiliary Lane on Freeway Speed with 2500’ ramp spacing. “Major” Benefit defined as 5+ MPH 4.3.2 Planning-Level Application of HCM Procedure In most cases, there is little flexibility with the spacing of exit-entry ramp combinations because the ramps are part of the same interchange and spacing is generally determined by interchange form. Of the remaining ramp combinations, entry-exit is the most common. Using the merge analysis procedures and traffic volumes for the freeway and ramps, the minimum ramp spacing needed for a desired LOS can be determined for six lane freeways. The HCM does not provide an analysis procedure for closely spaced entrance-exit ramps without auxiliary lanes. Simulation quantified the operation benefits of auxiliary lanes. Regardless of ramp spacing values, adding an auxiliary lane generally will improve freeway speeds compared to a no- auxiliary-lane condition. 50 Guidelines for Ramp and Interchange Spacing

freeways (three lanes in each direction). This chart can be used to consider ramp and freeway volumes to evaluate whether approximate entry-exit ramp spacing values will create a design that achieves a desired LOS. Such charts can be used in a project’s initial planning stages to quickly test whether conceptual designs are feasible from a traffic operations perspective. The chart shown in Exhibit 4-9 is for LOS D. Charts developed for LOS C, D, and E are included in Appendix B. Chart users should begin by finding the volume of the freeway being studied on the x axis. Users should then find the set of curves associated with the volume on the entry ramp. In Exhibit 4-9, curves are provided for entry ramp volumes of 500 vehicles per hour (vph) and 1,750 vph for ease of presentation. For example, with a one-direction, three-lane freeway volume of 3,000 vph and an entrance ramp volume of 1,750 vph, proposed ramp freeway regardless of the volume on the downstream exit ramp. However, with the same freeway and entrance ramp volumes, if the proposed ramp spacing was only 2,500 ft, LOS D or better operation would be achieved with a downstream exit ramp volume of 800 vph, but not with a downstream exit ramp volume of 1,200 vph or 1,750 vph. For entrance and exit ramp volumes not shown in Exhibit 4-9, users can interpolate between ramp volumes. These tools can assist users in quickly assessing various freeway interchange or ramp spacing alternatives and assess how these alternatives meet various levels of service targets. Planning-level tools can help correlate various target levels of service with ramp spacing values. spacing of 3,500 ft should result in LOS D or better operation on the Exhibit 4-9 is a chart developed from a HCM merge procedure for six-lane Operational and Safety Considerations 51

Exhibit 4-9 Example of Minimum Ramp Spacing Guidance Developed from HCM Ramp-Freeway Junction Procedure The charts shown in Exhibit 4-9 and Appendix B are not replacements for a complete HCM analysis. These charts are based on assumptions of some inputs into the ramp-freeway junction procedures (such as the peak-hour factor, the vehicle mix, and acceleration lane length). Exhibit 4-9 is intended to give planners and designers a means of assessing whether interchange concepts are likely to be feasible at the earliest stage of project development. Concepts that are likely feasible would warrant further investigation using complete HCM procedures. 4.4 MICROSIMULATION Microsimulation is the most detailed and data-intensive analysis that could be conducted for estimating the traffic operations on a highway and at an interchange. There are multiple types of microsimulation tools and some are more effective at estimating highway operations than others. Microsimulation tools require an extensive amount of data that is often not available for all types of projects. Therefore, these types of analysis tools should be applied appropriately, given the amount and type of data available and the specific needs of the project. Microsimulation tools can be helpful in investigating complex ramp sequencing scenarios and the queue interactions of ramp terminal intersection traffic control or ramp metering relationships. Charts for other target levels of service are presented in Appendix A. 52 Guidelines for Ramp and Interchange Spacing

4.5 SAFETY This section provides an overview of relationships between ramp spacing and safety. A discussion of a substantive safety approach to ramp spacing. These Guidelines are consistent with safety analysis approaches in AASHTO’s Highway Safety Manual (HSM). This section compares interchange spacing and ramp spacing dimensions in the context of safety analysis. Also, this section includes a description of observed and modeled relationships between ramp spacing and safety for three ramp scenarios: (1) an entrance ramp followed by an exit ramp, (2) two consecutive entrance ramps, and (3) an exit ramp followed by an entrance ramp. Freeway interchange ramps, by definition, coincide with increased vehicle lane changing, acceleration, and deceleration adjacent-to and on the freeway mainline. Observable operational measures, including density, average speed, and speed differentials, as well as the higher cognitive and decision-making demands on drivers at, near, and between interchange ramp locations are often used as surrogates to deduce lower expected levels of safety on freeway segments with increased ramp presence. Historically, expected crash patterns at ramp locations, including crash frequencies, crash severities, and crash types, were relatively unknown. 4.5.1 Traditional View of Ramp Spacing and Safety The transportation profession has traditionally taken a nominal approach to safety analysis; a design alternative either meets all geometric design criteria or does not. Acceptable safety performance, measured by having low crash risk, is presumed to result from attaining desired design criteria. If the criteria is achieved, the design is presumed “safe.” If minimum values are not attainable, the design is presumed “unsafe.” This idea, applied to ramp spacing, is illustrated in Exhibit 4-10. the AASHTO Green Book (see Exhibit 3-13). The AASHTO values are intended to represent general guidance that should be supplemented with more detailed geometric, operational, safety, and signing analyses. However, the values are often applied as “absolute minimums” in early stages of interchange planning. The binary result (i.e., above minimum or below minimum, safe or unsafe) of a nominal safety approach is interpreted as an indicator of acceptable or unacceptable design. Unacceptable designs are associated with visions of poor driving performance and high frequencies and severities of crashes. These generalizations oversimplify driver behavior and complex interactions between roadway geometrics, traffic operations, and safety. They also oversimplify the definition of safety itself and the trade-offs that often exist between crash frequencies and severities. traditional, criteria-based view of the ramp spacing/safety relationship is followed by a Actual values for minimum spacing have been based on recommendations in Operational and Safety Considerations 53

Exhibit 4-10 Nominal Safety Approach to Ramp Spacing. Adapted from (25). 4.5.2 Defining Substantive Safety The HSM assists professionals in taking a substantive approach to safety, where expected crash frequencies and outcomes for different design alternatives can be predicted and analyzed. These substantive safety measures result in more informed decision making, but a more complex decision making environment. Instead of a traditional binary approach (“safe” or “unsafe”), designers now have a continuously changing safety function readily available for their use. Exhibit 4-11 illustrates a non-binary approach to considering the safety continuum. It is difficult to recommend absolute minimum dimensions from the expected safety outcomes themselves, but it is possible to conduct more meaningful trade-off analysis that considers a variety of important transportation, environmental, societal, and cost factors. Exh ibit 4-11 Substanti ve Safety Approach to Ramp Spacing. Adapted from ( 28 ). These Guidelines provide a means of investigating the “continuum of safety” associated with ramp spacing v alues. Traditional approach to investigating the safety of ramp spacing decisions followed an all or nothing philosophy. Either ramp spacing values w ere acceptable or not. These Guidelines present a substantive safety discussion of ramp spacing. 54 Guidelines for Ramp and Interchange Spacing

These Guidelines present a substantive safety discussion of ramp spacing, with safety defined as: The number of crashes, or crash consequences, by type and severity, expected to occur on an entity during a specified time period (29). This safety definition is comprised of three main components that are consistent with the HSM: • crash number, • crash type, and • crash severity. Professionals should consider all three components when assessing alternate ramp and interchange concepts. For example, knowing that the total number of predicted crashes increased is not enough to say safety has decreased; changes in crash severities must also be known. Additionally, knowledge of crash types is necessary to link safety outcomes with specific design decisions and to identify effective safety improvements. Relationships between ramp spacing and safety are most meaningful and informative when discussed in terms of crash numbers, crash types, and crash severities. Crash number is the total number of crash events, regardless of crash type or severity. The total number of crashes serves as a baseline to compute crash type and crash severity proportions. It can be used as a safety performance measure with the idea that design decisions or safety countermeasures that reduce the total crash count are effective (i.e., if a crash does not occur, then there is no chance of a traffic fatality or injury). However, total crash counts alone do not provide a complete understanding of the safety associated with alternative ramp spacing values, particularly in the ramp spacing context where there are complex interactions between design features, traffic operations, and safety. Crash type refers to the manner of vehicle collision. At the highest level, crash types are classified by the number of vehicles involved in the crash. Single-vehicle crash examples include overturn and fixed object collisions. Multiple-vehicle crash examples include same-direction-sideswipe, opposite- direction-sideswipe, rear-end, head-on, and angle collisions. Multiple-vehicle crashes become more prevalent than single-vehicle crashes as traffic volumes and levels of congestion increase. They are also common at locations where conflicting traffic movements interact, including entrance ramps, exit ramps, and weaving areas. Head-on and angle collisions are generally associated with higher crash severities (increased likelihoods of occupant injuries and fatalities). The severity of sideswipe and rear-end Operational and Safety Considerations 55

collisions depends on the impact speeds of the involved vehicles and the presence of vehicle occupants near the impact location. Crash severity is a measure of the crash outcome with respect to occupant health following the collision. The recorded crash severity refers to the most severe injury to any vehicle occupant involved in the collision. For example, a collision involving two vehicles, each vehicle having a driver and two passengers (i.e., six total occupants involved in the crash), will be recorded in a crash database as an injury crash if only one occupant sustained an injury and the remaining five occupants were unharmed. Fatal and injury crashes are often combined into one crash category, referred to as severe crashes or fatal-plus-injury crashes. Crash severity is strongly related to the change in speed a vehicle and its occupants experience during a collision (see Exhibit 4-12). Understanding this phenomenon is critical to analyzing and interpreting ramp spacing and safety relationships. A given crash is more likely to be severe during free- flowing conditions (i.e., “better” levels of service) when vehicles are traveling at higher speeds. Crashes during congested conditions, often associated with high volumes and short ramp spacing, are more likely to result in property- damage-only within the boundaries of the congestion. However, crashes are more likely to be severe at the end of queues formed upstream of these areas, where there is an abrupt transition from high to low speeds. Crash severity is also strongly linked to driver and vehicle factors, particularly occupant ages and the weights of vehicles involved in the crash (30). Exhibit 4-12 Relationship between Crash Severity and Change in Vehicle Speed during a Collision (31). These guidelines provide tools to assess crash number, crash type, and crash severity for various ramp spacing dimensions. 56 Guidelines for Ramp and Interchange Spacing

Safety Analysis Elements 4.5.3 Interchange or Ramp Spacing when Discussing Safety Interchange spacing, defined from cross-street-centerline to cross-street- centerline, is not as meaningful as ramp spacing, defined from painted gore to painted gore, from a safety modeling and analysis standpoint. For a given interchange spacing, the freeway segment between the cross streets may have different numbers, types, combinations, and spacings of interchange ramps. In addition, cross streets associated with some interchange ramps are difficult to identify for atypical interchange types, and may not be centered between exit and entrance ramps. As a result, the safety discussions in these Guidelines focus on relationships between ramp spacing and safety. The relationships can be aggregated to determine interchange spacing effects for different interchange forms if desired. 4.5.4 Relationships between Ramp Spacing and Safety Past studies of ramp spacing and safety have generally indicated an increase in the total number of crashes (of all types and severities) as ramp spacing decreased, all else being equal. Findings on crash severity have been inconclusive, but hinted that the proportion of crashes resulting in a fatality or injury decreased as ramp spacing decreased. Tools for analyzing the relationship between ramp spacing and safety have been developed through research leading to these Guidelines. Research was conducted for the following ramp combinations: • an entrance ramp followed by an exit ramp (EN-EX), both with and without an auxiliary lane connecting the ramp terminals; • an entrance ramp followed by another entrance ramp (EN-EN); and • an exit ramp followed by an entrance ramp (EX-EN). The EN-EX scenario was studied at the greatest level of depth in developing these Guidelines. This is a commonly occurring ramp-sequence scenario and one in which operational analyses are frequently conducted and safety information is frequently needed. The Guidelines present safety performance functions (SPFs) and ramp spacing accident modification factors (AMFs) for the EN-EX ramp combination. The EN-EN and EX-EN were explored less vigorously; trends between crash frequencies and ramp spacing for different volume levels were developed without controlling for other potential safety influencing features. The safety analysis tools to conduct quantitative assessments of ramp spacing on freeway mainline safety are presented in Section 5.3.3, and their application is illustrated in the Case Studies. The remainder of this section summarizes the general findings of the research conducted to develop these Guidelines, including observed trends between ramp spacing and crash frequencies, types, and severities. Ramp spacing provides a more meaningful evaluation than interchange spacing. These Guidelines focus on ramp spacing value, safety relationships. Research indicates crashes generally increase as ramp spacing decreases. However, fatal and injury crash trends are less clear. Operational and Safety Considerations 57

4.5.4.1 ENTRANCE RAMP FOLLOWED BY EXIT RAMP (EN-EX) As discussed in Chapter 1, ramp spacing is defined as the distance between the painted tips of the entry ramp and the exit ramp. However, entering and exiting vehicles can cross striped areas, so any safety analysis limited to the area between painted tips might omit crashes that occurred between the physical gores and the painted tips. Safety analysis conducted as part of the development of these Guidelines considers all crashes that occur between physical gores, while still defining ramp spacing as the distance between painted tips. This is illustrated in Exhibit 4-13. Safety analysis tools do not address rear-end crashes that may occur far upstream of the entrance gore as a result of queue formation during congested conditions. Exhibit 4-13 Freeway Segment Definition for Safety Analysis of an Entrance Ramp Followed by an Exit Ramp 4.5.4.1.1 EN-EX Ramp Spacing versus Crash Frequency The expected number of crashes increases as ramp spacing decreases, all else being equal. This trend is illustrated in Exhibit 5-5. The sensitivity of total crashes to ramp spacing becomes higher as ramp spacing decreases. For example, reducing ramp spacing from 1,400 to 1,200 ft (200 foot reduction), is associated with a larger increase in crash frequency than reducing ramp of total crashes to ramp spacing becomes close to negligible for spacing values greater than about 2,600 ft; in other words, the safety performance of the segment between the ramps approaches that of a basic freeway segment with no ramps. Tools to estimate the impact of ramp spacing on crash frequency are provided in Section 5.3.3.1. 4.5.4.1.2 EN-EX Ramp Spacing versus Crash Type The expected number of crashes involving more than one vehicle increases as ramp spacing decreases. As ramp spacing decreases and lane change intensity increases, a crash is much more likely to involve at least two vehicles and be a sideswipe or rear-end collision. The rate of increase is higher than for total crashes. In other words, the percentage of total crashes classified as multiple vehicle increases as ramp spacing decreases. This trend is quantified in Exhibit 5-7. Safety analysis tools to quantitatively assess ramp spacing alternatives are presented in Section 5.3.3 The sensitivity of total crashes to ramp spacing becomes higher as ramp spacing decreases. The sensitivity of total crashes to ramp spacing becomes close to negligible for spacing values greater than about 2,600 ft. spacing from 2,400 to 2,200 ft (also a 200 foot reduction). The sensitivity Tools to estimate the percentage of total crashes that are multiple vehicle crashes as a function of ramp spacing are provided in Section 5.3.3.1. The 58 Guidelines for Ramp and Interchange Spacing

4.5.4.1.3 EN-EX Ramp Spacing versus Crash Severity The expected number of crashes resulting in a fatality or injury to at least one vehicle occupant increases as ramp spacing decreases. T he rate of increase is lower than for total crashes. In other words, the percentage of total crashes that result in at least one fatality or injury decreases as ramp spacing decreases as quantified in Exhibit 5-7. T he general speed/severity trend is illustrated in Exhibit 4-12. Traffic speeds are likely to be lower in urban areas with higher volumes and shorter ramp spacing values. While the likelihood of crashes increases in these areas, the probability of a severe crash decreases due to the slower speeds. T his finding is applicable to crashes occurring on the freeway mainline between the physical gore of the entrance ramp and the physical gore of the exit ramp. It is not applicable to rear-end crashes that may occur far upstream of the entrance gore as a result of queue formation during congested conditions. These crashes are likely to be severe when the relative speed differences between colliding vehicles are large. Tools to estimate the percentage of total crashes expected to result in a fatality or injury to at least one vehicle occupant as a function of ramp spacing are provided in Section 5.3.3.1. T he percentage of total crashes expected to be severe is approximately 20% at a 600 ft ramp spacing value. The proportion reaches 30% at a ramp spacing of 2,000 ft and remains at approximately 30% for all larger spacing values. 4.5.4.1.4 EN-EX Ramp Safety with Auxiliary Lanes The presence of an auxiliary lane between an entrance ramp and an exit ramp corresponded to approximately 20% fewer expected crashes for a given ramp spacing and traffic volume level. The expected 20% overall reduction is the result of a reduction in multiple vehicle collisions. T he presence of an auxiliary lane has no effect on single vehicle collisions. The reduction applies almost equally to both fatal plus injury crashes and property damage only crashes. E quation 5.1 and Exhibit 5-7 can be used to estimate crash frequencies and severities with and without an auxiliary lane. T he process is demonstrated in Case Study 3 4.5.4.2 ENTRANCE RAMP FOLLOWED BY ENTRANCE RAMP (EN-EN) Like entry-exit safety analysis, entry-entry safety analysis defines ramp spacing as the distance between painted tips but also considers crashes that occur outside of the area. Specifically, entry-entry safety analysis considers all A n auxiliary lane between an entrance and an exit ramp corresponded to about 20% fewer crashes. percentage of total crashes involving more than one vehicle is approximately 90% at a 600 ft ramp spacing value. The proportion reaches 65% at a ramp spacing of 3,000 ft and remains at approximately 65% for all larger spacing values. Operational and Safety Considerations 59

Exhibit 4-14 Freeway Segment Definition for Safety Analysis of an Entrance Ramp Followed by an Entrance Ramp Data and sample sizes are relatively limited for the entry-entry ramp combinations compared to the entry-exit combinations. Safety analysis tools consider only freeway and ramp volumes in addition to segment length and ramp spacing; variable definitions are provided in Section 5.3.3.2. The remainder of this section summarizes observed relationships between ramp spacing and safety for the EN-EN scenario. 4.5.4.2.1 EN-EN Ramp Spacing versus Crash Frequency Crash frequency is defined by the total number of crashes (all severities and types) expected to occur between the physical gore of the first entrance ramp to the end of the acceleration lane taper of the second entrance ramp. The expected number of crashes increases as ramp spacing decreases. The sensitivity of crash frequency to ramp spacing for the EN-EN is similar to, but slightly less than, the sensitivity for the EN-EX combination. The trend is illustrated in Exhibit 5-8. The sensitivity of total crashes to EN-EN ramp spacing becomes close to negligible for spacing values greater than about 2,200 ft; in other words, the safety performance of the segment approaches that of a basic freeway segment with no interchange ramps. The value is slightly less than the 2,600 ft value for the EN-EN. The relative differences are in agreement with the geometric analysis, which showed a slightly smaller not feasible (see Section 5.3.1). Tools to estimate the impact of EN-EN ramp spacing on crash frequency are provided in Section 5.3.3.2. 4.5.4.2.2 EN-EN Ramp Spacing versus Crash Type A majority of crashes on segments with an entry-entry ramp combination involve more than one vehicle. The percentage of total crashes classified as The expected number of crashes increases as ramp spacing decreases, with the sensitivity become negligible for spacings greater than 2,200 feet. Regardless of the spacing dimension, about three quarters of crashes at entry-entry ramp combinations involve multiple vehicles. The percent of crashes that are severe at entry- entry ramp combinations increases as ramp spacing decreases. dimension for the EN-EN than the EN-EX in which geometrics are likely crashes that occur between the physical gore of the first entry ramp and the end of the acceleration lane taper of the second ramp. This is illustrated in Exhibit 4-14. 60 Guidelines for Ramp and Interchange Spacing

4.5.4.2.3 EN-EN Ramp Spacing versus Crash Severity The expected number of crashes resulting in a fatality or injury to at least one vehicle occupant increases as ramp spacing decreases. The rate of increase is lower than for total crashes. In other words, the percentage of total crashes that result in at least one fatality or injury decreases as ramp spacing decreases. The magnitude and direction of this relationship is the same as that for the EN-EX ramp combination. Therefore, the fatal plus injury curve in Exhibit 5-7 can be used to predict the percentage of the total crashes on the EN-EN segments expected to be severe. Explanations and limitations of these findings are the same as for the EN-EX combination (see Section 4.5.4.1.3). 4.5.4.3 EXIT RAMP FOLLOWED BY ENTRANCE RAMP (EX-EN) AND EXIT RAMP FOLLOWED BY EXIT RAMP (EX-EX) Research conducted to develop these Guidelines did not show an increase in crashes associated with a decrease in ramp spacing for the EX-EN ramp combination. Most data were from EX-EN combinations within the same interchange, which is different than an exit ramp followed by an entrance ramp servicing grade separated ramps (ramp braids). The geometric analysis should be a primary factor in the spacing assessment until additional safety information becomes available (see Section 5.3.1.4). consistent with the EX-EN results (i.e., no relationship between ramp spacing and safety). Without quantitative safety findings, the geometric analysis (Section 5.3.1.3) and signing considerations (Section 5.3.4) are the primary factors for the EX-EX spacing assessment. EN-EX ramp combinations do not show an increase in crashes as ramp spacing decreases. The safety characteristics of the EX-EX combination are expected to be multiple-vehicle varies between 70% and 80% and is relatively insensitive to the ramp spacing dimension. Operational and Safety Considerations 61