Click for next page ( 12


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 11
11 CHAPTER THREE Contents Of Policies And Programs This chapter includes a detailed literature search regard- staffing and organizational structure for administering the ing the typical contents of access management policies and program. NCHRP Report 348 (6, p. 9) identifies key compo- programs at the state DOT and local levels, including ACS nents of an access management program as follows: and various key access management features, including the following: Classifying road systems, based on their areawide importance, into a logical functional hierarchy Spacing for traffic signals Planning, designing, and maintaining roadway sys- Spacing for unsignalized driveways and street tems based on criteria, such as road geometry, and connections functional or access classification Corner clearance Defining allowable access levels and spacing for each Median treatments roadway class that do not degrade its function in the Median openings hierarchy (this involves identifying when and where Access management on interchange crossroads access can be permitted and setting appropriate crite- ria for the spacing of access points) Also discussed are other typical program elements such Applying appropriate geometric design criteria and as the following: traffic engineering analysis to each allowable access Using driveway permit procedures and regulations to Access permit processes (including access denial) ensure that decisions are reasonably enforceable and Traffic impact studies that the government agency can manage effectively Purchase of access rights roadway design and operation Access design concepts (including frontage roads, Providing a mechanism for granting variances when right- and left-turn lanes, and alternative left-turn reasonable access cannot be provided treatments) Establishing a means for enforcing criteria. Each agency has unique circumstances to address The chapter concludes with the survey results concern- ing various state DOT and local agency program elements, including a general overview of access management pro- Literature Search grams, ACS, access management techniques, access permit processes, and traffic impact studies. The TRB Access Management Manual indicates that access management includes both systemwide and corridor-based programs. Systemwide programs involve the development Background and implementation of a comprehensive access management program for all roadways under state or local jurisdiction. The scope and content of access management programs in Corridor-based programs focus on the development and the United States vary widely. Some agencies have a compre- implementation of corridor AMPs. Corridor-based pro- hensive code that establishes a sound legal basis for access grams are useful for retrofitting problem areas or addressing management decisions. Other agencies establish guidelines the needs of high-priority corridors and often are combined or desirable practices as a step in the process of develop- with a systemwide approach. Some systemwide programs, ing a more extensive program. Formal agency regulations, for example, contain or authorize corridor-based solutions however, are more legally defensible and ultimately more (1, p. 7). (Corridor AMPs are discussed in more detail in effective in implementing an access management program chapter four.) (1, pp. 4041). The following elements of a comprehensive, systemwide All programs need to address a variety of technical and access management program are described in this section: administrative factors. These factors range from the type of access features that will be managed, to the appropriate ACS (access classification system)

OCR for page 11
12 Access features denied (or highly restricted) for higher-level arterial class Access management techniques roadways, although access may be provided where no rea- Access permit process sonable alternative access is available. Direct property Traffic impact studies access often is permitted for lower-level arterials and collec- Purchase of access rights tors, although the number and location of access points may Access design concepts be limited. Direct property access typically is allowed on local roadways and frontage roads, subject to safety consid- Access Classification System erations, such as maintaining proper sight distances. An ACS typically is used to establish the level of allowable Transportation agencies use a range of ACSs. Two access for roadways of varying levels of importance in a ACSs--North Carolina DOT's Strategic Highway Corridors state highway system. As stated in the Access Management initiative and Indiana DOT's ACS development--are high- Manual (1, p. 72), an ACS is a hierarchy of access categories lighted in chapter six as profiles of contemporary practice. that forms the basis for the application of access manage- ment. Although the structure of an ACS may vary widely Access Features among different agencies, establishing an ACS involves three basic actions: For each roadway classification that is established, an agency must determine the access features that will be managed and Defining access management categories how they will be managed. Access management standards Establishing whether access should be permitted, and for these features are assigned to roadways through the related access spacing and design criteria for each access categories (although access in the vicinity of inter- category changes typically is addressed through statewide standards, Assigning an access management category to each AMPs, or interchange areas management plans) (1, p. 42). roadway or roadway segment Access features to manage include the following: Each access category sets forth criteria governing the Traffic signals (minimum spacing distances or through access-related standards and characteristics for correspond- bandwidth) ing roadways. These access categories ultimately define Driveway and street connections, and corner clear- areas where access can be allowed between private devel- ance (minimum spacing distances, location, allowable opments and the roadway system, and where it should be movements, and design) denied or discouraged. The categories also define spacing Medians (to manage left turns and direct access) and standards for signalized and unsignalized intersections, and median openings (minimum spacing distances and where turning movements should be restricted. Defining design) access categories typically involves consideration of the fol- Interchanges and access in the vicinity of lowing factors: interchanges Level of importance of the roadways within the overall The following section provides an overview of each of network hierarchy --The foundation of an ACS may these access features. The access management techniques be the functional classification system (i.e., arterial, that are related to these features are presented in the Access collector, and so on) or another similar hierarchy that Management Techniques section. reflects the general purpose of each roadway within the transportation system. Spacing for Traffic Signals Roadway characteristics --Roadway characteristics associated with geometric design (e.g., number of Establishing traffic signal spacing criteria for arterial road- lanes, design speed, and median treatment) and traffic ways is one of the most important and basic access manage- operations (e.g., volume and speed) may be considered ment techniques. The same criteria for signal spacing apply in defining access categories. to both signalized driveways and signalized public roadway Degree of urbanization and land use controls --Fac- intersections. tors such as the intensity of existing and planned devel- opment, intersection frequency, parcel size, and need Effects of Signal Spacing for a supporting circulation system can be used to help define the degree of urbanization and could be consid- The spacing of traffic signals, in terms of frequency and ered in defining access categories. uniformity, governs the performance of urban and suburban highways. Traffic signals account for most of the delays that Typically, direct property access is prohibited from free- motorists experience. Closely or irregularly spaced signals ways and expressways. Direct property access typically is reduce arterial travel speeds, thereby resulting in an exces-

OCR for page 11
13 sive number of stops even under moderate traffic volume minimal negative effects on the progression. The through conditions. Signals also can increase crash frequency (7, p. bandwidth measures how large a platoon of vehicles can pass 22). through a series of traffic signals without stopping for a red light. Bandwidth may be expressed in terms of the number of NCHRP Report 420 identified that each traffic signal seconds per cycle or the percent of cycle length that the traffic added per mile to a roadway reduces travel speed by about could flow within a platoon. Further guidelines for through 2 to 3 mph. Table 1 shows the percentage increases in travel bandwidth are contained in NCHRP Report 348 (6, pp. 5658) times as signal density increases, using two traffic signals and in the Access Management Manual (1, pp. 140149). per mile as a base. For efficient traffic flow, NCHRP Report 348 indicates that new signals should be limited to locations where the TABLE 1 progressive movement of traffic will not be impeded signifi- PERCENTAGE INCREASES IN TRAVEL TIMES AS SIGNAL cantly. The "optimum" distance between signals depends on DENSITY INCREASES the cycle length and the prevailing progression speed. At the Percent Increase in Travel Time optimum distance, bandwidth is not lost. When signals are Signals per Mile (compared with 2 Signals per mile) placed at nonoptimal locations, bandwidth is lost and delay 2 0 increases (6, p. 58). 3 9 4 16 Spacing for Unsignalized Driveways and Street Connections 5 23 6 29 Access points--commonly referred to as driveways or street 7 34 connections--introduce conflicts and friction into the traf- 8 39 fic stream. They are, in effect, intersections and should be designed consistent with their intended use. A Policy on Source: Gluck et al. (7, p. 28). Geometric Design of Highways and Streets (i.e., AASHTO's "Green Book") indicates that the number of crashes is dis- Signal Spacing Criteria proportionately higher at driveways than at other intersec- tions. Therefore, driveway design and location merit special Signal spacing is a function of progression speed and signal consideration (8, pp. 729731). cycle lengths. The spacing distances for various combina- tions of progression speeds and cycle lengths are shown in Effects of Unsignalized Access Spacing Table 2. Nearly 50 years of research efforts have documented the basic relationships between access frequency and safety. The TABLE 2 methods of analyses and the resulting relationships among SIGNALIZED INTERSECTION SPACING FOR VARIOUS PROGRESSION SPEEDS AND CYCLE LENGTHS individual studies vary, but the patterns are generally similar. Roadways with properly managed access have lower crash Speed (mph) Cycle rates than other roadways. Arterial roadways with many length(s) 25 30 35 40 45 50 55 driveways and signals often have double or triple the crash 60 1,100 1,320 1,540 1,760 1,980 2,200 2,420 rates of roadways with wide spacing between access points 70 1,280 1,540 1,800 2,050 2,310 2,570 2,820 or of roadways where access is fully controlled. Crash rates generally increase with greater frequencies of intersections 80 1,470 1,760 2,050 2,350 2,640 2,640 2,640 and driveways. Table 3 lists a sample of the many studies that 90 1,630 1,980 2,310 2,640 2,640 2,640 2,640 have considered how crash rates are related to spacing (9). 120 2,200 2,640 2,640 2,640 2,640 2,640 2,640 Source: Adapted from Gluck et al. (7). NCHRP Report 420 presented information that had been Note: Spacing distances are in feet. synthesized from other studies to arrive at the composite predictors of crash rates for ranges of unsignalized and sig- A spacing of 2,640 ft is shown where the computed spacing in the table exceeds 2,640 ft. nalized access densities. The report presented the results of an analysis of crash data from around the nation, includ- ing a series of three graphs for quantifying the relationship When signalized driveways and intersections are placed between crash rates and signalized and unsignalized access at these distances, signal progression can be maintained and densities. All three figures have been incorporated into the green bandwidth (through bandwidth) is not lost. Small devi- 2004 AASHTO A Policy on Geometric Design of Highways ations in the signal location (e.g., less than 10%) will have and Streets (8).

OCR for page 11
14 TABLE 3 SAFETY EFFECTS OF ACCESS Context Result Year and Source Minnesota--about 4/5 of the segments in the Greater access densities were accompanied by late 1940s study were two-lane highways lower speeds and higher crash rates. Kipp (10, pp. 3337) Excepting low volume two-lane roadways, Oregon--426 sections of urban state routes, all increasing commercial driveways per mile or 1959 with parallel parking increasing signal density was correlated with Head (11, pp. 4563) increasing crash rates. California--closing median openings at Resulted in lower numbers of total crashes in 1967 selected intersections in a corridor the corridor. Wilson (12) For most crash types, the crash rate tended to increase as the number of median openings (excluding intersections) increased. 1967 Cribbins et al. North Carolina--92 homogeneous urban and Excepting crashes at night on unlit sections, (13, pp. 825) rural multilane divided highway sections with crash rate increase associated with an posted speed limits ranging from 35 to 60 mph increase in the frequency of signalized 1967 Cribbins et al. intersections. (14, pp. 140157) Crash rates increased as access points increased. Indiana--examined 100 urban arterial sections, The crash rate would likely decrease when the 1967 Mulinazzi and Michael and also compared 15 pairs of similar sections number of access points or the number of traffic (15, pp. 150173) (the majority were two-lane sections) signals per mile was reduced. The driveway crash rate decreased as spacing Indiana--examined 100 urban arterial sections 1976 McGuirk et al. between driveways or between driveways and in cities with populations more than 30,000 (16, pp. 6672) street intersections increased. North Carolina--6-year study of 57 undivided Crash rate increased as access density 1983 Heimbach et al. (17) urban four-lane sections increased. Wisconsin--referenced a regional planning Crashes per mile dramatically increased when 1993 McGee and Hughes commission study of both county and state the average spacing between access connec- (18, pp. 287291) trunk highways tions was less than 300 ft. Florida--median modifications on 5.1-mi section of a four-lane divided arterial Collision rates decreased by 15% Before: 77 full median openings 1997 Wu (19) Injury rates decreased by 24% After: closed 16 openings and converted 42 full openings to directional openings Both urban two-lane and urban four-lane roadways, crash rates increased as either Minnesota--432 rural and urban segments public street or commercial access density increased. 1998 Preston et al. (20) from the state network Rural areas, the crash rate increased as access density increased. Arkansas--3 years of crash data from all rural The crash rate increased as access density and suburban four-lane highways (excluding 2005 Gattis et al. (21) increased. freeways) on the state system Source: Gattis (9). Figures 5 and 6 present crash rates by median type and 0.13 crashes per MVMT on highways with two-way left-turn total access density (both directions) for urban-suburban and lanes (TWLTLs) or nontraversable medians. In rural areas, rural roadways, respectively. In urban and suburban areas, each driveway added would increase the annual crash rate by the addition of each driveway would increase the annual 0.07 crashes per MVMT on undivided highways and by 0.02 crash rate by 0.11 to 0.18 crashes per million vehicle-miles crashes per MVMT on highways with TWLTLs or nontra- traveled (MVMT) on undivided highways, and by 0.09 to versable medians (7, p. 55).

OCR for page 11
15 FIGURE 5 Estimated crash rate by type of median--Urban and suburban areas. Source: Gluck et al. (7, p. 57). FIGURE 6 Estimated crash rate by type of median--Rural areas. Source: Gluck et al. (7, p. 57). FIGURE 7 Estimated crash rate by access density--Urban and suburban areas. Source: Gluck et al. (7, p. 58).

OCR for page 11
16 Figure 7 presents crash rates by driveway density and sig- lanes. As defined in NCHRP Report 420, the influence area nal density. Each unsignalized driveway may increase the associated with a driveway includes (1) impact length (the crash rate by approximately 0.02 crashes per MVMT at low distance back from a driveway that vehicles begin to be signal densities. At higher signal densities, each unsignal- affected by driveway traffic), (2) perception-reaction dis- ized driveway may increase the crash rate from 0.06 to 0.11 tance, and (3) vehicle length. crashes per MVMT. The spacing of driveways should reflect the impact NCHRP Research Results Digest 247 (35) compares the lengths and influence areas associated with motorists enter- crashes per MVMT found in NCHRP Report 420 (7 ) with ing or leaving a driveway. The impact length represents the research results from Minnesota. It also compares safety distance upstream when the brake lights of through vehicles indices set forth in NCHRP Report 420 and derived from the are activated or when one vehicle changes lane because of a Minnesota data. Both sets of results confirm that driveways turning vehicle ahead. merit special consideration. The impact lengths associated with motorists entering Research by Levinson (36 ) explored estimating the safety or leaving a driveway should be considered in establish- of arterial roads based on traffic volumes, access road (i.e., ing driveway separation distances. A Policy on Geomet- driveways and intersecting streets) volumes, and access den- ric Design of Highways and Streets includes a figure (see sity. The research applied the relationship between intersec- Exhibit 9-101) derived from NCHRP Report 420 that identi- tion crashes and the product of conflicting traffic volumes fies impact lengths relating to vehicles making right turns to estimate safety. A simplifying assumption--that access into driveways (8, p. 730). For example, at 30 mph speed, roads have roughly equivalent volumes--made it possible 20% of the right-lane through vehicles was affected at an to identify safety indexes that relate only to the change in approximate distance of 172 ft or more in advance of a drive- access density; these indexes generally are consistent with way. At 50 mph, 20% of the right-lane through vehicles was those reported in NCHRP Report 420. These indexes esti- affected a distance of 345 ft or more in advance of a right- mate that the increase in crashes is roughly equal to the turn location. Influence areas can be obtained by adding the square root of the increase in access density (36 ) (e.g., a dou- perception-reaction distance and vehicle length to the dis- bling in driveway density would increase the crash rate by tance shown in the AASHTO exhibit. The functional area of more than 40%). This is known as the "square root rule." an intersection should reflect these influence areas. Unsignalized Access Spacing Criteria Transportation NCHRP Report 420 provides the series of tables from Research Circular Number 456 (37 ) provides information which the AASHTO exhibit was derived, and these tables on the basic considerations that may be applied in the devel- can be used to establish connection spacing guidelines based opment of sound unsignalized access spacing criteria. The on the spillback expected to occur along a roadway section. report provides an overview of selected spacing guidelines (The report includes five tables, one each for posted speeds at the state and local levels of government. It indicates that of 35 mph to 55 mph, in 5-mph increments.) Spillback occurs jurisdictions that have adopted access management regula- when a through vehicle must brake in response to another tions have used different approaches for establishing unsig- vehicle making a right turn at an access connection. The nalized access spacing criteria. The Access Management spillback rate represents the percentage of through vehicles Manual (1, p. 150) lists the following possible approaches experiencing such an event. Table 4 presents, as an example, for establishing unsignalized connection spacing criteria: the table for a 35-mph roadway. Safety As noted in the Access Management Manual, the higher Stopping sight distance the roadway functional classification, the lower the accept- Intersection sight distance able spillback rate. The acceptable rate on a major roadway Functional area may be no more than 2%. A spillback rate of 5% may be Right-turn conflict overlap acceptable on a major collector serving commercial, indus- Influence distance trial, or large mixed-use areas, whereas 15% or more may Egress capacity be acceptable on major collectors in residential areas. For example, the minimum access connection spacing on a As noted in A Policy on Geometric Design of Highways 45-mph major urban arterial, assuming a 2% spillback rate, and Streets (8), driveways should not be located within the would be at least 530 ft. The minimum spacing on a 35-mph functional area of an intersection, or within the influence major collector (right-turn-in driveway volume of between area of an adjacent driveway. The functional area extends 30 and 60 vehicles per hour) would be 355 ft if a 5% spill- both upstream and downstream from the physical intersec- back rate is acceptable, and 280 ft if a 15% spillback rate is tion area and includes the longitudinal limits of auxiliary acceptable (1, pp. 153154).

OCR for page 11
17 TABLE 4 PERCENTAGE OF RIGHT-LANE THROUGH VEHICLES INFLUENCED AT OR BEYOND ANOTHER DRIVEWAY: POSTED SPEED = 35 MPH Right-Turn-In Volume per Driveway, R (vph) R 90 Multiple Multiple Multiple Multiple Driveways, Driveways, Driveways, Driveways, No. At Least At Least At Least At Least Driveway Driveways Single Once per Single Once per Single Once per Single Once per Spacing per 1/4 Mi., Driveway 1/4 Mi., Driveway, 1/4 Mi., Driveway, 1/4 Mi., Driveway, 1/4 Mi., (ft) n P2 1- (1-P2)n P2 1- (1-P2)n P2 1- (1-P2)n P2 1- (1-P2)n 100 13.2 2.4% 27.3% 7.5% 65.2% 12.2% 82.1% 21.8% 96.1% 125 10.6 2.4% 22.5% 7.5% 56.0% 12.2% 74.7% 21.8% 92.5% 150 8.8 2.4% 19.1% 7.5% 49.5% 12.2% 68.2% 21.8% 88.5% 175 7.5 2.4% 16.4% 7.4% 44.0% 12.1% 62.1% 21.8% 84.0% 200 6.6 2.2% 13.9% 7.1% 38.3% 11.5% 55.4% 20.6% 78.1% 225 5.9 2.0% 11.2% 6.3% 31.8% 10.3% 47.2% 18.4% 69.7% 250 5.3 1.5% 7.7% 4.8% 22.7% 7.8% 34.7% 13.8% 54.5% 275 4.8 1.1% 5.3% 3.5% 15.9% 5.8% 24.8% 10.3% 40.7% 300 4.4 0.8% 3.6% 2.6% 11.1% 4.3% 17.6% 7.7% 29.6% 325 4.1 0.6% 2.6% 2.0% 8.0% 3.3% 12.8% 5.9% 22.0% 350 3.8 0.5% 1.8% 1.5% 5.6% 2.5% 9.0% 4.4% 15.6% 375 3.5 0.3% 1.2% 1.1% 3.7% 1.7% 6.0% 3.1% 10.5% 400 3.3 0.2% 0.7% 0.7% 2.3% 1.1% 3.7% 2.0% 6.6% 425 3.1 0.1% 0.5% 0.5% 1.4% 0.8% 2.4% 1.4% 4.2% 450 2.9 0.1% 0.3% 0.3% 0.9% 0.5% 1.5% 0.9% 2.6% 475 2.8 0.1% 0.2% 0.2% 0.5% 0.3% 0.8% 0.5% 1.4% 500 2.6 0.0% 0.1% 0.1% 0.2% 0.1% 0.4% 0.3% 0.7% 525 2.5 0.0% 0.0% 0.0% 0.1% 0.1% 0.2% 0.1% 0.3% Source: Gluck et al. (7, p.140). The Access Management Manual offers the following Gattis et al. performed NCHRP Project 15-35: Geo- guidance in selecting and applying unsignalized access metric Design of Driveways (5 ) to develop a driveway spacing (1, p. 155): design guide that addresses the needs of the various users in the driveway, roadway, and sidewalk area and reflects Longer spacing standards are generally applied to the requirements of the Americans with Disabilities Act roadways of a higher functional classification. (ADA). As part of this effort, the authors completed a lit- Higher classifications of roadways typically have erature search and review, compiled transportation agency higher speeds than roadways of a lower classification. design documents, and documented state of the practice. Higher classifications of roadways tend to carry higher This led to the identification of research needs, some of traffic volumes than roadways of lower classification. which were addressed in NCHRP Project 15-35. The rec- The interference with through traffic increases as ommendations for the geometric design of driveways will traffic volume increases. A small number of turning be useful to state DOTs, local governments, and consul- vehicles can interfere with a large number of through tants in preparing driveway design standards and practices vehicles on high-speed, high-volume suburban-urban that consider standard engineering practice and accessibil- roadways--especially during peak periods. A single ity needs, and that provide for safe and efficient travel by vehicle turning from a through lane can disrupt pla- motorists, pedestrians, bicyclists, and transit users on and tooned flow and traffic progression. in proximity to the affected roadway. Roadways with speeds 45 mph are typically more critical than those with speeds 40 mph.

OCR for page 11
18 Corner Clearance The PIEV + Maneuver distances for various speeds are tabulated in Table 5-13 of Transportation and Land Develop- As noted in A Policy on Geometric Design of Highways and ment. As an example, the desirable PIEV + Maneuver distance Streets (8, p. 729), driveways should not be located within for a speed of 30 mph would be 250 ft, whereas for a speed of the functional area of an intersection or in the influence 50 mph, the PIEV + Maneuver distance would be 570 ft. area of an adjacent driveway. The functional area extends both upstream and downstream from the physical intersec- Downstream clearance on major roadway--This tion area on both the major roadway and the intersecting distance is calculated as the greater of the upstream cross street, and includes the longitudinal limits of auxiliary clearance distance (see "Upstream clearance on major lanes. As a result, the functional area encompasses the area roadway") or the AASHTO stopping sight distance, where motorists are responding to the intersection, deceler- based on speed. ating, and maneuvering into the appropriate lane to stop or Upstream clearance on minor crossroad--The max- complete a turn. imum back-of-queue length. Downstream clearance on minor crossroad--Allow Additional guidance related to the computation of the drivers to clear the major roadminor road intersec- driveway influence area is available in NCHRP Report 420, tion (see Figure 6-23 of Transportation and Land and was presented in the Spacing for Unsignalized Drive- Development). ways and Street Connections section of this report. Another general guideline that applies to driveway location is that Median Treatments sight distance must be sufficient. AASHTO's "Green Book" (8, pp. 651677) contains detailed guidance related to the Left turns increase vehicular conflicts, as well as conflicts purpose and computation of sight distance. In addition, with pedestrians and bicyclists. They also result in increased driveways must be located so that they are conspicuous and crashes and delays, and complicate the signal timing and clearly delineated for the various users. phasing parameters at signalized intersections. These prob- lems are especially acute on major roadways (7 ). Therefore, Gattis et al. noted in NCHRP Project 15-35 (5) that one the presence (or absence) of a median has a substantial major objective of access management is avoiding driveway impact on roadway operations and safety, and on the provi- queuing that backs up into a public roadway. This is accom- sion of left-turn access to abutting properties (1, p. 199). plished through design of the throat length, internal circu- lation, and traffic control within a site. Queuing of traffic Effects of Median Treatments exiting a site does not affect the operation of the public road- way, but it could affect site circulation and parking lot opera- As noted in NCHRP Report 420 (7, p. 68), the treatment of tions. This internal queuing is affected by the throat length, roadway medians has an important bearing on how well number of egress lanes, and traffic control at the intersection roadways operate, their crash experience, and the access with the public roadway. they provide to adjacent developments. The basic choices for designing medians are as follows: In Transportation and Land Development, Stover and Koepke (38) provide extensive guidance on criteria to use Whether to install a continuous TWLTL to provide sufficient corner clearance at intersections both Whether to install a nontraversable (physical) median on the upstream and downstream sides of the major roadway on an undivided roadway and intersecting cross street. This guidance is summarized Whether and when to replace a TWLTL with a nontra- as follows: versable median Upstream clearance on major roadway--This The Access Management Manual contains definitions distance is calculated as follows: of these three cross-section types (1). An undivided road- way offers no control of, or refuge for, turning and cross- Upstream clearance = (PIEV distance + Maneuver distance) + Queue ing vehicles. A TWLTL has a flush-center lane that serves as refuge for left-turning vehicles. A nontraversable median Where: is depressed or raised, and actively prohibits crossing and PIEV distance = Distance traveled during Perception- turning movements. Although a traversable (or flush) paved Identification-Evaluation-Volition (commonly referred median is not intended to be crossed, it does not actively to as "Perception-Reaction distance"). restrict left-turn and crossing movements. Maneuver distance = Distance traveled while maneuvering and decelerating to a stop. Table 5 presents a selection of studies compiled by Gattis that together span half a century. Some of the studies com- Queue = Maximum back-of-queue length. pared vehicular crash rates among all three cross-section

OCR for page 11
19 types, while others compared two of them. The general trend NCHRP Report 395 compared the different outcomes is that nontraversable medians are associated with lower from a number of crash prediction models developed by dif- crash frequency. Continuous two-way left-turn lanes gener- ferent researchers. A composite finding suggested that, as ally are preferable to undivided roadways, but generally are traffic volumes exceed approximately 15,000 Average Daily not preferable to nontraversable medians (9). Traffic (ADT), a raised median is safer than a two-way left- turn lane. Both are safer than no median (i.e., an undivided roadway) for volumes at least as low as 10,000 ADT (30 ). TABLE 5 relative safety of cross-section design alternatives Relative Safety of Median Treatment (1 = best, 3 = worst) Context Restrictive Year and source Undivided TWLTL (e.g., raised, depressed) California--563 miles of 4-lane ADT 5,0009,000 3 - 1 without speed zones, development, or ADT 10,00014,999 1 - 3 1953, Telford et al. intersections (22, pp. 208231) ADT 15,00025,000 3 - 1 with speed zones, ADT 5,0009,999 1 - 3 roadside development ADT 20,00029,000 3 - 1 Illinois--Compare two suburban roadways 3 - 1 1968, Frick (23, pp. 1420) Nebraska--urban 4-lane 3 1 - 1986, McCoy et al. (24, pp. 1119) Georgia--82 urban 4- or 6-lane sections - 3 1 1989, Squires et al. (25) Florida--4-lane arterials urban 3 2 1 1993, Long et al. (26) rural 1 3 2 4 cities, 15 sites, 145.9 miles vehicular crashes 3 1 2 1994, Bowman/Vecellio central business district pedestrian crashes 3 2 1 (27, pp. 169179) suburban - vehicular crashes 2 3 1 Rural and urban; 4- and 6-lane - 3 1 1995 Hadi (28, pp. 169177) 4-lane suburban commercial with ADT < 1995, Marigiotta/ - 3 1 32,500; compare 11 TWLTL with 11 median Chatterjee (29) Phoenix, Omaha--urban, suburban Business or office; ADT > 10,000 2.5 2.5 1 1997, Bonneson/McCoy (30) Residential or ADT 15,00025,000 2 3 1 Industrial ADT > 25,000 3 2 1 4 states; 4-lane 3 - 1 1999, Council/Stewart (31) 5 states; 264 urban segments signalized and unsignalized 3 2 1 1999, Papayannoulis signalized density > 2/mile 3 2 1 et al. (32) signalized density < 2/mile 2 3 1 Springfield, MO--compare 3 multilane - 3 1 2000, Gattis/Hutchinson (33) commercial arterial sections Georgia--all divided highways - 3 1 2000, Parsonson (34) Overall Rankings (1 = best, 3 = worst) worst better best Source: Gattis (9).

OCR for page 11
20 Schultz et al. (39, pp. 1115) applied stepwise linear regression analysis to identify correlations between access management techniques and crash patterns. Their research indicated that the presence of a raised median corresponded to a reduction of 1.23 crashes per MVMT. In addition, raised medians were negatively correlated with right-angle colli- sions, while TWLTLs were positively correlated with oppo- site-direction collisions. A more recent median treatment option is to apply a "road diet" that converts a four-lane undivided roadway into three lanes (one through lane in each direction and a TWLTL). The fourth lane may be converted to bicycle lanes, sidewalks, or on-street parking. The Road Diet Handbook: Setting Trends for Livable Streets (40 ) is a practitioner's guidebook for FIGURE 8 Median decision choices. Source: Gluck et more information. al. (7, p. 69). Huang et al. (41, p. 1) performed safety research related to road diets and found the following: Selecting a median alternative will depend on many pol- icy, land use, and traffic factors. Factors identified in NCHRP Crash rates did not change significantly from the period Report 420 include the following: before the road diet to the period after the road diet. Although crash rates were lower at road diets than at Access management policy and access class for the comparison sites, road diets did not perform better or roadway under consideration worse (from the before period to the after period) rela- Types and intensities of the adjacent land use tive to the comparison sites. Supporting street system and the opportunities for Road diet conversion did not affect crash severity. rerouting left turns Road diet conversion did not result in a significant Existing driveway spacing change in crash types. Existing geometric design and traffic control features (e.g., left-turn provisions and proximity of traffic signals) In their conclusions and recommendations, the research- Traffic volumes, speeds, and crash patterns ers indicated a need for future safety and operational studies, Costs associated with roadway widening and recon- under a range of traffic volumes and other considerations, to struction (7, p. 85) identify the situations in which road diets would be appro- priate. They also noted that traffic operations and capacity Table 7 from NCHRP Report 420, based on the research per- must be considered fully at a given site before implementing formed for NCHRP Report 395, provides a comparative analy- a road diet or other lane reduction measures (41, p. 6). sis of the three midblock treatments, citing the strengths and weaknesses of each. NCHRP Report 395 contains more detailed Selecting a Median Type guidelines for selecting the midblock left-turn treatment based on benefit-cost comparisons for roadways serving various The basic decision process with respect to median type is land uses (either business and office land use or residential and illustrated in Figure 8. TWLTLs and medians improve traf- industrial land use). A series of tables in NCHRP Report 395 fic operations and safety by removing left turns from the may be used to help identify the following situations: through-traffic lanes. TWLTLs provide more access and maximize operational flexibility. Medians physically sepa- When an undivided cross-section should be converted rate opposing traffic, limit access and conflicts, and provide to a nontraversable median a better pedestrian refuge. Median design requires adequate When an undivided cross-section should be converted provisions for left turns and U-turns to avoid problems to a TWLTL associated with concentrating these movements at other When a TWLTL should be converted to a nontravers- locations (7, p. 68). Table 6 from the Access Management able median Manual, based on research conducted for NCHRP Report 395, provides a comparative evaluation of median treatments The Florida Department of Transportation (FDOT) consid- (1, p. 203). ers its median policy--which requires a restrictive median on new and reconstructed multilane highways--one of the more effective elements of its access management program (1, p. 42). The FDOT Median Handbook indicates that the department

OCR for page 11
21 has a 1993 Multi-lane Facility Policy that essentially directs all The research for NCHRP Report 524 investigated the department multilane projects over 40 mph in design speed to safety and operational effect of U-turns at unsignalized have a restrictive median. It also directs designers to find ways median openings. The safety performance of typical median to use restrictive medians in all multilane projects, even those opening designs was documented, and guidelines for the use, below the 40 mph design speed (42, ch. 1, p. 9). location, and design of unsignalized median openings were developed. The research included unsignalized median open- The FDOT Median Handbook (42) is a valuable resource ings on all types of divided highways, but the focus was on to guide decisions related to the design of medians and the urban-suburban arterials because these present the greatest location of median openings. It indicates that: current challenge to highway agencies in access management. Among the research conclusions were the following (43): Full median openings serve a "Major" transition function. This means that on arterial roads they should only be provided at arterial junctures of the road system For urban arterial corridors, average median opening as defined for the public street or internal circulation accident rates are slightly lower for conventional three- systems. (ch. 2, p. 2) leg median openings than for conventional four-leg median openings. Median Openings For urban arterial corridors, average median opening accident rates for directional three-leg median openings Figure 9, from FDOT's Median Handbook (42, ch. 2, p. 19), are about 48% lower than the accident rates for conven- illustrates components in the spacing of median openings, tional three-leg median openings. including deceleration length, queue storage, perception- For urban arterial corridors, average median opening reaction distance or full width of median, and turn radius. accident rates for directional four-leg median openings TABLE 6 COMPARISON OF TYPES OF MIDBLOCK LEFT-TURN TREATMENTS Comparison Nontraversable Median TWLTL Undivided Operational effects 1. Reduced delay to major roadway traffic 2. Enhanced capacity 3. Reduced delay to major roadway left turns 4. Reduced delay to minor roadway left turns a. Low-volume major roadway b. Highway-volume major roadwaya Safety effects 1. Reduced vehicular crashes 2. Pedestrian refuge 3. Positive guidanceb Other effects 1. Aesthetics 2. Snow removal 3. Construction costc Source: Access Management Manual (1, p. 203). Note: = most effective/preferable; = somewhat effective/somewhat preferable; = least effective/ least desirable. a Very low capacity for direct left turns due to an absence of gaps in traffic on the major roadway. A nontraversable median has a relatively high capacity for "left turns" that can be made by a right-turn followed by a U-turn. b Effective communication to motorist. c NCHRP Project 3-49 concluded that the difference between a raised, curbed median and a TWLTL is negligible. Florida DOT reports a slightly lower cost for a "flush grass" median (exclusive of landscaping) than for a TWLTL.

OCR for page 11
34 because inconsistent or infrequent application of stan- ards to a minimum. The encroachment permit may contain dards makes them vulnerable to legal challenges. specific terms and conditions providing for the expiration of the waiver if, in the future, the grounds for the waiver no In a review of variance considerations for access manage- longer exist (59). ment, Eisdorfer and Siley (58) noted the following: Two contemporary profiles relating to access permit An exception which is granted to a standard has the effect of lowering that standard. Because agencies are processes--Minnesota DOT's Permit Process and Oregon obligated to act consistently, agency staff should be wary DOT's Central Highway Approach/Maintenance Permit of recommending approval of any variance that they are System (CHAMPS)--are highlighted in chapter six. not prepared to grant every time a similar circumstance arises. To achieve consistency, an agency must consider future decisions based on a record established through Traffic Impact Studies past decisions. This requires tracking of all exceptions which have been requested and noting the disposition As noted in the Access Management Manual (1, pp. 224 and reasoning behind each outcome. . . . Variances that 225), a site TIS assesses the effects that a proposed develop- are routinely granted should eventually be authorized as accepted practice. (pp. 289297) ment will have on the surrounding transportation network, the ability to get traffic on and off the site, and the need for off-site mitigation. A TIS is an essential part of the develop- Eisdorfer and Siley (58, pp. 289297) suggest the follow- ment review process to assist developers and public agencies ing hierarchy for variance decision making that reflects the in making land use decisions. The studies are appropriate relative importance of the access feature: not only during access permitting, but also during requests for subdivision, rezoning, and other development activities Safety (e.g., sight distance, etc.) when a proposal may have a substantial adverse impact on Spacing of interchanges transportation operations. A well-prepared TIS helps the Spacing of traffic signals developer and permitting agency accomplish the following: Spacing of driveways Corner clearance Forecast the traffic impacts created by proposed devel- Number of driveways on one property opment based on accepted practices, not perception Edge clearance between the driveway and property Determine improvements needed to accommodate the sidelines proposed development Allocate funds more efficiently For example, using this concept, review staff would place Relate land use decisions with traffic conditions somewhat less emphasis on compliance with driveway spac- Evaluate the number, location, and design of access ing in a case in which a variance is needed to maintain ade- points quate sight distance for safe operations (58, pp. 289297). Update traffic data Identify needed roadway improvements The South Carolina DOT's 2008 Access and Roadside Provide a basis for determining the developer's respon- Management Standards included new provisions regarding sibility for specific off-site improvements access waivers (variances). The request for an access waiver should describe the undue hardship that will be placed on the Small developments (typically fewer than 100 trips per applicant if a waiver is not granted. A waiver will be granted hour) usually are exempted from preparing a TIS, because only if the following is determined: the impact of these developments generally will be limited to the vicinity of the access connection. However, a site access Denial of the waiver will result in loss of reasonable and circulation review can be conducted to ensure that access access to the site. connections are safely located. Principal elements of this The waiver is reasonably necessary for the convenience review include sight distance, driveway geometry, driveway and welfare of the public. throat length, and provisions for bicycles and pedestrians. All reasonable alternatives that meet the access requirements have been evaluated and determined to For all other developments (typically those that generate be infeasible. 100 trips or more in the peak hour), some type of TIS gener- Reasonable alternative access cannot be provided. ally is required as part of the access permit review appli- cation (60 ). The type of analysis can depend on the size, When a waiver is approved, the reasons for granting the impact, and complexity of the development. Typically, the waiver and any recommendations given by the South Caro- larger the development (as measured by the number of trips lina DOT need to be clearly stated and included in depart- generated) the larger the area that may experience a mea- ment files. Restrictions and conditions on the scope of the surable traffic impact caused by the development. Table 12, permit are imposed as required to keep potential safety haz- from the TRB Access Management Manual (1, p. 226, Table

OCR for page 11
35 TABLE 12 SUGGESTED REQUIREMENTS FOR VARIOUS TYPES OF TRAFFIC IMPACT STUDIES Trip Generation Threshold Access Location Small Medium Large & Design Development: Development: Development: Review Traffic Impact Traffic Impact Regional Traffic Assessment Statement Analysis T 1,000 Hour Trips Peak-Hour Trips Peak-Hour Trips Peak-Hour Trips Preapplication meeting or discussion Analyses of roadway issues Existing condition, analysis within study area Sight distance evaluation Nearby driveway locations ? Existing traffic conditions at nearby intersections and driveways Future road improvements ? Crash experience in proximity to site ? Trip generation of adjacent development ? Trip distribution analysis Background traffic growth ? Future conditions analysis at nearby intersection ? Mitigation identification and evaluation ? ? Site issues Traffic generation Traffic distribution ? Evaluation number, location, and spacing of ? access points Evaluate access design, queuing, etc. Evaluate site circulation Other analyses Gap analysis for unsignalized locations ? ? TSM/TDMa mitigation measure (car-or vanpool- ? ing, transit etc.) - transit agency participation Effect on traffic signal progression analysis of b b ? proposed signal locations Source: Access Management Manual (1, p.226, Table 12-1). Note: = required; ? = may be appropriate on a case-by-case basis. a TSM/TDM = transportation system management/transportation demand management b Not signalized 12-2), identifies basic analyses suggested for inclusion in Four profiles of contemporary TIS practices--Louisiana the TIS. As shown in Table 12, the scope and complexity of DOT's TIS process, Caltrans' Equitable Share Responsibil- analysis to be conducted should be determined based on the ity Calculations, as well as New Jersey DOT's vehicle-use projected number of peak-hour trips. limitations and transit-trip credit methodologies--are pre- sented in chapter six.

OCR for page 11
36 Purchase of Access Rights with current policies, legal and real estate literature, and other publications that address this subject. The findings focus on the Access control by the acquisition of property rights has been three main areas: acquisition, management, and disposal. Les- used on the Interstate Highway System since it was man- sons learned and information gaps are explored. dated by the Federal Aid Highway Act of 1956. A growing number of agencies are recognizing the benefits of acquiring Access Design Concepts property rights to control access on other important arterial highways to preserve safety and mobility. The purchase of Access design concepts complement the access features and property rights can prevent undesirable accesses at the loca- spacing criteria presented in this chapter. These concepts tions where the property rights were acquired (61, p. 5). include the following: The purchase of access rights may be expensive and Alternative left-turn treatments time-consuming compared with access regulation, but the Frontage roads purchase of access rights is a stronger and longer-lasting Left-turn lanes solution. Regulations can change with political administra- Right-turn lanes tions and attitudes (6, p. 10). Access rights may be purchased to achieve the following: Appendix D contains a retrofit toolbox, adapted from South Dakota DOT (62), for implementing access manage- Limit access to designated locations or side streets ment techniques (available on web version only). Control access and sight distance at intersections or interchanges Limit access to designated highways or new facilities Business Turns to Access Management Principles and bypasses Introduce long-term or permanent access control One of the largest overnight package delivery companies Improve locations with high crash experience (1, p. 314) has spent years studying and refining its vehicle routing to improve efficiency, operations, and the bottom line. Access rights may be acquired through negotiation, pur- The company decided to implement a routing strategy chase, or the power of eminent domain, and is recorded in consistent with basic access management principles. the county of record. The purchase of access rights offers the By routing trucks to make right turns and minimize the following advantages: number of left-turn movements, the company is saving millions of dollars on its gasoline bill. It recognized that Provides long-term assurance of access control, the time the trucks spent in left-turn lanes leads to more Avoids concerns over property rights and regulatory engine idling, fuel consumption, and traffic delays. The takings by compensating property owners for access company also recognized that left turns are not as safe rights, and as right turns. This conclusion was reached based on Avoids the expense of purchase or condemnation, if it the extensive experience of its drivers and reconfirms is achieved through negotiated dedication. the crash analysis findings in the access management research. The purchase of access rights may have the following Source: http://compass.ups.com/features/article.aspx?id=340 and disadvantages: http://solveclimate.com/blog/20080422/ups-goes-left-turn-diet- slims-down-its-carbon-footprint (April 22, 2008). Cost may be prohibitive, It may be difficult to establish a dedicated funding Alternative Left-Turn Treatments source in light of other needs, An effective tracking mechanism is required for NCHRP Report 420 indicates that U-turns are being used enforcement, and increasingly as an alternative to direct left turns to reduce Condemnation is required when a negotiated purchase conflicts and to improve safety along arterial roads. is unsuccessful (1, p. 314). U-turns make it possible to prohibit left turns from drive- way connections onto multilane highways and to eliminate NCHRP Synthesis 351 (61) provides additional information traffic signals that would not fit into time-space (progres- regarding access rights. It was prepared for state transporta- sion) patterns along arterial roads. When incorporated tion agency personnel, as well as for others who are involved into intersection designs, U-turn provisions enable direct in acquiring access rights along roadways other than freeways. left turns to be rerouted and signal phasing to be simpli- It documents the state of the practice with the intent to limit fied (7, p. 97). the amount of access to the roadway to better manage highway safety and mobility. Successful practices are documented along

OCR for page 11
37 The operational and safety issues related to direct left- The prohibition of direct left turns from existing drive- turn movements have been noted by an overnight delivery ways may transfer the displaced left turns to the nearest company. traffic-signal-controlled intersection unless intermediate U-turn lanes are provided. The increased left-turn volumes Cities and states use various approaches for reducing the at public road intersections would require longer left-turn number of conflicts involving left turns along their arterials. phases, which could reduce the green time and capacity One approach is to provide dual left turns at intersections for through movements. U-turn provisions are especially with collector streets, with the innermost lane accommo- important along roadways with relatively few median open- dating U-turns. Another approach is to prohibit left-turn ings. Several approaches have evolved for accommodating exits onto major arterials and to provide midblock U-turn the diverted left-turn volumes by providing U-turn lanes in lanes to accommodate these movements. New Jersey uses advance of, at, or beyond intersections. The U-turns may jughandles along multilane divided highways. Michigan be made from conventional left-turn lanes or via jughan- uses U-turn channels on highways with wide medians and dles from the right (curb) lanes (7, p. 97). Illustrative treat- prohibits all left turns at signalized intersections. Most ments from NCHRP Report 420 are shown in Figure 15, states do not have standards, however, and handle U-turn and are as follows: provisions on a case-by-case basis (7, p. 97). FIGURE 15 U-turns as an alternative to direct left turns. Source : Gluck et al. (7, p. 99).

OCR for page 11
38 Left-turn lanes can be provided for U-turning vehi- in advance of downstream signalized intersection. However, cles in advance (i.e., upstream) of signalized intersec- if U-turns are provided at a signalized intersection, vehicles tions. This avoids concentrating development-related making a right turn followed by a U-turn could encounter turning traffic at signalized junctions of major longer delay and travel time than those making direct left crossroads. turns at a driveway. Considering the safety benefits, the Dual left-turn lanes can be provided at signalized longer delays related to this treatment are not considered intersections, with the inner lane dedicated to U-turns. unacceptable, when the left-turn traffic demand at a drive- Many states now provide these lanes. However, they way is not so high. However, if the left-turn traffic demand still require multiphase traffic signal controls. at a driveway is relatively high--for example, greater than Left- and U-turn lanes can be provided downstream of 150 vehicles per hour--relocating the left-turning vehicles signalized intersections, thereby allowing two-phase to a downstream signalized intersection could constitute an traffic signal controls. operational concern. The increased numbers of U-turning vehicles also have some adverse impacts on the capacities of From 2001 to 2005, a series of research projects regard- signalized intersections. Therefore, when the left-turn traffic ing the safety and operational effects of U-turns was con- demand at a driveway is high, consideration should be given ducted by the University of South Florida for Florida DOT. to providing U-turn opportunities in advance of the down- Two basic research approaches were employed by Liu et al. stream signalized intersection (63, pp. 1314). (63, pp. 23) to evaluate the safety and operational effects of various driveway left-turn alternatives. The research This analysis found that providing U-turn locations at a approaches include traffic conflict technique and opera- special unsignalized location before the traffic signal has tional data analysis to compare the safety and operational many positive safety and operational impacts. Finding an performance of three driveway left-turn alternatives that appropriate location for this midblock U-turn median open- are widely used in Florida, and nationally. These driveway ing upstream of a traffic signal in built-out areas sometimes left-turn alternatives include direct left turns from a drive- can be difficult, however, because of the tight geometric con- way, right turns followed by U-turns at a median open- ditions. For this condition, Liu et al. (63, pp. 1314) recom- ing, and right turns followed by U-turns at a signalized mend the consideration of providing a median opening for intersection. U-turns after the signalized intersection. Lu and Williams (64, p. 61) performed a safety analysis The research performed by Zhou et al. (65, p. 78) devel- for 258 sites in seven Florida counties to identify the safety oped a methodology to quantify the operational effects of benefit of this access control treatment. The results indi- U-turns as alternatives to direct left turns from driveways. cated that this treatment could lead to a statistically signifi- The researchers noted that many concerns guide the deci- cant reduction of total crashes in both the crash frequency sion about which type of median opening should be used, and crash rate on major arterial roadways with nontravers- indicating that safety considerations are the first priority, able medians, high traffic volumes and speeds, and moder- followed by the operational efficiency of the highway, and ate to high driveway and side-street volumes. Although the the delay of vehicles at the driveway. Their research dem- property-damage-only average crash numbers were similar onstrated that U-turns as alternatives to direct left turns between the direct left turns and this treatment, the injury- provide better safety with regard to traffic conflicts and fatality crash rate of right turns followed by U-turns was fewer effects on through-traffic operations on a major much lower. These results indicate that the U-turn concept highway. NCHRP Report 524 (43) provides additional has a beneficial impact on safety relative to the typical full research related to the safety of U-turns at unsignalized median opening design. median openings. The conflict data analysis results from Liu et al. (63, pp. A different potential treatment to combat congestion 1314) show that indirect left turns generally are safer than and safety problems at intersections is the Median U-Turn direct left turns from driveways. Vehicles making a right Intersection Treatment, which sometimes is referred to as turn onto the major street and a U-turn at a downstream the "Michigan U" or "Michigan Boulevard" treatment, median opening were shown to result in 47% fewer conflicts because it has been used extensively in Michigan for many than those making direct left turns from a driveway. Vehi- years. It also has been implemented successfully in Florida, cles making a right turn followed by a U-turn at a signalized Maryland, and New Jersey (66, p. 1). This treatment gen- intersection were shown to result in about 26% fewer con- erally is applied along a corridor that involves a multilane flicts than those making direct left turns. roadway with a nontraversable median, where left turns are not allowed at or between intersections. Figure 16 illustrates The delay and travel time comparison results show that an how left turns are made in this treatment. indirect left-turn movement does not result in longer delay or travel time, if U-turns are provided at a median opening

OCR for page 11
39 FIGURE 16 Vehicular movements at a Michigan U-turn intersection treatment. Source : Rodegerdts et al. (67, p. 243). The treatment involves the elimination of direct left Although this is typically a corridor treatment, the con- turns at signal-controlled intersections from major or minor cept has been used successfully for isolated intersec- approaches. Drivers desiring to turn left from the major road tions to improve traffic operations and safety. onto an intersecting cross street must first travel through the "Loons" (extra widening of the pavement beyond the at-grade, signal-controlled intersection and then execute a normal shoulder) can be installed to accommodate U-turn at the median opening downstream of the intersec- larger U-turning vehicles, so this treatment can be fea- tion. These drivers then can turn right at the cross street. For sible for corridors with narrow medians. drivers on the side street who want to turn left onto the major road, they must first turn right at the signal-controlled inter- Directional median crossovers provide better operational section and then execute a U-turn at the downstream median and safety benefits compared with bidirectional median opening and proceed back through the signalized intersec- crossovers. tion. This arrangement can be implemented with and with- out signal control at the median openings on the major road The reduction in signal phases at intersections where this (66, p. 2). Because of the additional right-turning volume, treatment is applied provides increased capacity in com- right-turn lanes may need to be added on the approaches parison to the conventional intersections. The capacity to the intersection. Because the Michigan U-turn treatment increases are typically in the range of 20% to 50%. accommodates only through and right-turning traffic at the The total network travel time savings can and usually intersection of the major and minor roads, this arrangement do outweigh the additional travel time required for left- requires only a two-phase traffic signal, which can reduce turning vehicles from the major road and cross street cycle lengths and improve signal coordination. for corridors with this treatment compared with con- ventional intersections. The literature review and synthesis conducted by Jaganna- The safety performance of this treatment is better than than (66, p. 13) summarized the advantages and disadvantages conventional intersections because they have fewer of the "Michigan U" compared with conventional, at-grade vehicle-vehicle conflict points. Typical total crash signal-controlled intersections with left turns permitted from reductions range from 20% to 50%. all approaches. It offered the following major conclusions: Head-on and angle crashes that have high probabilities of injury are significantly reduced compared with con- Michigan and other states have used this treatment ventional intersections. successfully for more than 4 decades without major problems related to traffic operational failures or safety In research sponsored by FHWA, Jagannathan (66, p. 2) hazards. applied a traffic simulation model to analyze the performance Positive guidance communicated through additional of three New Jersey Department of Transportation (NJDOT) signs and pavement markings at sites with this treat- jughandle design configurations as shown in Figures 1719. ment may be beneficial in reducing driver confusion This included a comparison with conventional intersections and enhancing traffic safety. for a variety of traffic flows and signal settings. With respect to driver expectancy, this treatment should not be mixed with other indirect and direct left- The generalizations drawn from the research conclusions turn strategies on corridor-level implementations. (66, p. 11) were as follows:

OCR for page 11
40 FIGURE 17 Typical geometry of the jughandle intersection--Case "A." Source : Jagannathan (66 ). FIGURE 18 Typical geometry of the jughandle intersection--Case "B." Source : Jagannathan (66 ).

OCR for page 11
41 FIGURE 19 Typical geometry of the jughandle intersection--Case "C." Source : Jagannathan (66 ). Jughandles have lower average intersection delays com- The vehicular capacity for left turns on the major road of pared with conventional intersections for near-saturated the jughandle decreases as the ramp offsets decrease. traffic conditions. The magnitude of difference ranges from 15% to 35% for forward-forward jughandles, from 20% NCHRP Report 348 (6, p. 76) illustrates a treatment at a to 40% for forward-reverse jughandles, and from 25% to large activity center that involves a "directional" design of 40% for reverse-reverse jughandles. The jughandles had access roads to separate major conflicting left-turn move- similar traffic performance compared with conventional ments. This treatment, illustrated in Figure 20, may achieve intersections for undersaturated traffic conditions. high capacities because it permits two-phase signal operations The jughandles have higher intersection capacities at each intersection. It requires a divided highway and works compared with conventional intersections for satu- well in cases in which dual left-turn entry lanes are provided. rated traffic conditions. The magnitude of the differ- ence ranges were 20% to 25%, 25% to 30%, and 25% "Superstreets" present another alternative left-turn treat- to 40% for the forward-forward, forward-reverse, and ment. Instead of allowing left-turn and through movements reverse-reverse jughandles, respectively. from side streets to be made directly through a two-way The reverse-reverse jughandles had the highest inter- median opening, a "superstreet" redirects these movements section capacity, followed by the forward-reverse and 500 to 1,000 ft downstream on the main street to a one-way the forward-forward jughandles. The changing of the median opening. As shown in Figure 21, a left turn from a left-turn gap acceptance maneuver (forward jughandle side street will be made by a right turn and then a U-turn. ramp) to a right-turn merge maneuver (reverse jughan- A through movement from a side street would be made by a dle ramp) yields a 5% to 15% increase in intersection right turn, a U-turn, and then another right turn. capacity based on the distribution of turning movement percentages on all approaches. Hummer and Jagannathan (68) prepared a paper on "super- The travel times and number of stops per vehicle for street" implementation and research, which concentrates on jughandles are lower compared with conventional safety and reviews the performance of several "superstreets" in intersections only for near-saturated traffic conditions. North Carolina and Maryland. In North Carolina, a rural "super- For other traffic scenarios, jughandles are comparable street" application resulted in a reduction in the total crash rate of or have slightly higher travel times and stops compared approximately 36%, and a reduction in the fatal or injury crash with conventional intersections. rate of approximately 55%. In addition, a signalized suburban

OCR for page 11
42 FIGURE 20 Directional access treatment. Source : Koepke and Levinson (6, p. 76). application in North Carolina resulted in a crash rate below the between the intersections along the main roadway facilitates statewide average for that roadway type. In Maryland, signifi- the design of auxiliary lanes for deceleration and accelera- cant safety improvements were noted at the rural applications. tion. Thus, frontage roads segregate through and local land- service traffic, thereby protecting the through-traffic lanes Advantages of a "superstreet" include a reduction in the from encroachment, conflicts, and delays (7, p. 121). number of conflict points, reduction in the number of traffic signal phases, improved signal progression (since the pro- NCHRP Report 348 indicated (as adapted for NCHRP gression in one direction has little to do with the progression Report 420 ) that frontage roads must be designed care- in the reverse direction), and potential for safer pedestrian fully to avoid increasing conflicts at junctions and delays crossings (pedestrians must cross an intersection on a diago- on intersecting roads. The following planning and design nal as shown in Figure 21). Disadvantages and issues related guidelines should be considered when installing arterial to a "superstreet" include how to address high side-street frontage roads in both new developments and retrofit situ- volumes, the need for a wide median, potential driver confu- ations (6, p. 68): sion related to a new design, and perceived adverse impacts on roadside businesses. Hummer and Jagannathan recom- Frontage roads, especially for "retrofit" situations, mend that further research be done on "superstreets" to help should operate in one direction and should enter or answer questions related to safety, efficiency, environmental leave the mainline lanes as merging or diverging move- benefits, design details, business impacts, and other aspects. ments. Preferably, these merging and diverging loca- tions should not be signalized, as shown in Figure 22. The separation of frontage roads at cross streets should be maximized to ensure sufficient storage for crossroad traffic between the frontage roads and the arterial. The absolute minimum separation should be 150 ft, where two-way frontage roads are provided. This dimension is about the shortest acceptable length for placing signs and other traffic control devices. Greater distances are needed to provide adequate FIGURE 21 "Superstreet" schematic. Source : Hummer and left-turn storage and to separate operation of the two Jagannathan (68 ). intersections. Spacing of at least 300 ft (preferably more) enables turning movements to be made from the Frontage Roads main lanes onto the frontage roads without seriously disrupting arterial traffic and thereby minimizes the NCHRP Report 420 identifies the frontage road as an access potential of wrong-way entry onto the through lanes management technique that reduces the frequency and sever- of the predominant highway. ity of conflicts along the main travel lanes of a highway. Direct "Reverse" frontage roads, with developments along property access is provided from the frontage roads and pro- each side, are desirable in developing urban areas. A hibited from the main travel lanes. The resulting spacing desirable separation distance is 600 ft with a minimum FIGURE 22 Arterial frontage road concept for retrofit conditions. Source : Gluck et al. (7, p. 125).

OCR for page 11
43 distance of 300 ft. The frontage road may operate in (6 ), provides examples of each option and shows when each one or two directions. should be considered. Frontage roads that can be terminated at each block operate well with respect to the arterial roadway and Left-turn lanes offer important benefits by removing the cross street. This type of design should be consid- the turns from the through-traffic lanes. As a result, they ered in cases in which continuity of the frontage road reduce rear-end collisions and increase capacity. In addi- is not needed. tion, they improve the visibility of oncoming traffic for In cases in which major activity centers front an arte- vehicles turning left. This helps to reduce right-angle col- rial roadway, frontage roads should be incorporated lisions (7, p. 88). into the ring road or otherwise eliminated. A minimum outer separation of 20 ft should be used to Left-turn lanes normally are provided by offsetting the provide space for pedestrian refuge and safe placement centerline or by recessing the physical (or painted) median. of traffic control devices and landscaping. Examples of single and dual left-turn lanes are shown in Pedestrian and bicycle movements should use the Figure 23. A typical shared-lane treatment is shown for frontage roads. Parking may be permitted where the comparison purposes (6, p. 71). frontage roads traverse residential areas. NCHRP Project 3-91: Left-Turn Accommodations at The following sections discuss left-turn and right-turn Unsignalized Intersections is investigating what conditions lanes. NCHRP Project 03-98 is performing research related to warrant the installation of a left-turn lane or some other the development of guidelines on the use of auxiliary through accommodation, such as provision for a left-turn bypass lanes at signalized intersections. The research objective is to or indirect left turns. The objectives of the NCHRP Project provide guidelines and procedures to analyze, justify, and 3-91 research are as follows: design auxiliary through lanes at signalized intersections. Develop an objective and clear process for the selection of Left-Turn Lanes left-turn accommodations at unsignalized intersections Provide guidance on the design of these As indicated in NCHRP Report 420, left turns may pose prob- accommodations lems at driveways and street intersections. They may increase conflicts, delays, and accidents and often complicate traffic Left turns should be removed from the through-traffic signal timing. These problems are especially acute at major lanes wherever possible. Therefore, provisions for left suburban highway intersections where heavy left-turn move- turns (i.e., left-turn lanes) have widespread application. ments take place, but also occur where left turns enter or leave Ideally, left-turn lanes (or jughandles) should be provided driveways serving adjacent land development (7, p. 88). at driveways and street intersections along major arterial and collector roads wherever the turns are permitted. This The following illustrate these problems: is essential to improve safety and preserve capacity (7, p. 95). More than two-thirds of all driveway-related accidents involve left-turning vehicles (69, pp. 3740). The design of left-turn lanes is straightforward. The In cases in which more than six left turns are made per lanes should be shadowed (protected) from the through- traffic signal cycle, virtually all through vehicles in the traffic lanes and transitions around the lanes for through shared lane may be blocked by the left-turning vehicles traffic (where required) should be gradual. The storage (70, pp. 4552). lengths should be maximized by keeping entry tapers In cases in which left-turn lanes are provided along relatively short (7, p. 95). Stover and Koepke (38 ) provide multilane highways, each opposing left-turning vehicle extensive guidance on criteria to use in designing left-turn reduces the through-vehicle capacity by the number of lanes. through lanes it crosses (e.g., 100 left turns/hour across three traffic lanes reduce the through vehicle capacity The Access Management Manual (1, pp. 172173) notes by about 300 vehicles) (70, pp. 4552). that it is important for turn lanes on roadways of a high functional classification to be of sufficient length to store The treatment of left turns has an important bearing on the all arriving vehicles most of the time. For example, at an safety and movement along arterial roadways, and is one of intersection on a major arterial, it is probable that the stor- the major access management concerns. Left-turn movements age length will be sufficient to store all left-turning vehicles at driveways and street intersections may be accommodated, on an acceptable percentage of the cycles at least 95% of prohibited, diverted, or separated depending on specific cir- the time. On roadways of lesser importance, a lower likeli- cumstances (7, p. 88). Table 13, from NCHRP Report 348 hood of storing all arriving vehicles may be acceptable.

OCR for page 11
44 TABLE 13 TREATMENTS OF LEFT-TURNS AT INTERSECTIONS AND DRIVEWAYS Option Condition Application Considerations Limit to minor roads or places where Shared Lane R/W is not available for left-turn lane Provide Left-Turn Lane Protected or permissive phasing Dual Left-Turn Lane Protected phasing only Full Time Requires alternate routes Prohibit Peak Periods Only Requires alternate routes Divided highways at minor roads Jug-Handle (signalized junctions only) Divert Modified Jug-Handle 6-lane divided highways Divided highways with wide median - Michigan "U" Allows two-phase signals Directional Design Very heavy turns in one direction Separate Left-Turn Flyover Very heavy turns in one direction Through Lane Flyover Major congestion points Source: Koepke and Levinson (6, p. 70). Right-Turn Lanes Right-turn maneuvers from a two-lane arterial street at an unsignalized intersection or driveway can delay NCHRP Report 348 indicates that right-turn deceleration through traffic by 0 to 6 s per through vehicle, where lanes remove turning vehicles from the through traffic, no right-turn deceleration lane is present, depending thereby reducing the speed differences in the main travel on through volume, right-turn volume, and major- lanes and the frequency and severity of rear-end collisions. road traffic speed. Delays to through traffic due to Right-turn lanes also increase capacity at signalized inter- right turns in the same situation on a four-lane arterial sections and may allow refinements in phasing (6, p. 80). are substantially lower, in the range from 0 to 1 s per through vehicle, because through vehicles can change Right-turn lanes may be provided at a single access point, lanes to avoid delay. or they can be extended to accommodate several nearby Pedestrians at unsignalized intersections or driveways driveways. To operate as intended, the continuous lane can have a substantial impact on delay to through should not be longer than 0.25 mi (6, p. 80) to avoid addi- vehicles because of slowing by right-turning vehicles tional conflicts that would be introduced with both vehicular yielding to pedestrians. Provision of a right-turn lane and bicycle traffic. can reduce delays to through traffic by 0.4 to 2.1 s per through vehicle at a pedestrian volume of 50 pedes- The objective of the NCHRP Project 3-72 research by trians per hour, by 0.6 to 3.1 s per through vehicle at Potts et al. (71, p. 1) was to develop design guidance and a pedestrian volume of 100 pedestrians per hour, and criteria to address the safety and operational trade-offs for by as much a 6 s per through vehicle at a pedestrian motorists, pedestrians, and bicycles to channel right turns volume of 200 pedestrians per hour. and use right-turn deceleration lanes at driveways and unsig- An economic analysis procedure can identify areas nalized intersections. The scope of the project was limited to where provision of right-turn lanes at unsignalized urban and suburban arterials with speeds of 45 mph or less intersections and major driveways is cost-effective. (70, p. 2). As noted in this study, new access points, particu- The economic analysis procedure can be used to larly busy commercial driveways, often contribute notice- develop plots indicating combinations of through- ably to congestion and reduced outside travel-lane capacity. traffic volumes and right-turn volumes in cases in Several states have established application and design crite- which the provision of a right-turn lane would be war- ria for right-turn deceleration lanes for driveways and inter- ranted. Examples of such plots are presented in the sections, but the criteria vary widely from state to state. NCHRP 3-72 report. Following are three of the research conclusions from NCHRP Project 03-89: Design Guidance for Channel- NCHRP Project 3-72 (71, p. 116): ized Right-Turn Lanes is developing a process to determine