National Academies Press: OpenBook

Traffic Signal Control Strategies for Pedestrians and Bicyclists (2022)

Chapter: Chapter 9 - Treatments Addressing Special Bicycle Needs

« Previous: Chapter 8 - Treatments Offering Added Information and Convenience
Page 131
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 131
Page 132
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 132
Page 133
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 133
Page 134
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 134
Page 135
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 135
Page 136
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 136
Page 137
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 137
Page 138
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 138
Page 139
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 139
Page 140
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 140
Page 141
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 141
Page 142
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 142
Page 143
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 143
Page 144
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 144
Page 145
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 145
Page 146
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 146
Page 147
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 147
Page 148
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 148
Page 149
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 149
Page 150
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 150
Page 151
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 151
Page 152
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 152
Page 153
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 153
Page 154
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 154
Page 155
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 155
Page 156
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 156
Page 157
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 157
Page 158
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 158
Page 159
Suggested Citation:"Chapter 9 - Treatments Addressing Special Bicycle Needs." National Academies of Sciences, Engineering, and Medicine. 2022. Traffic Signal Control Strategies for Pedestrians and Bicyclists. Washington, DC: The National Academies Press. doi: 10.17226/26491.
×
Page 159

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.

131   This chapter describes six treatments that address needs specific to bicyclists: C H A P T E R 9 Treatments Addressing Special Bicycle Needs Primary Function Section Treatment Name Improve bicycle safety by providing sufficient clearance time 9.1 Minimum Green and Change Interval Settings for Bicycle Clearance Reduce bicycle delay by providing enhanced progression 9.2 Signal Progression for Bicycles 9.3 Two-Stage Left-Turn Progression for Bicycles Offer signal timing techniques made possible by bicycle detection 9.4 Bicycle Detection Provide information to increase cyclist compliance 9.5 Bicycle Wait Countdown Reduce bicycle delay 9.6 Easing Bicycle Right Turn on Red Restrictions Minimum Green and Change Interval Settings for Bicycle Clearance (Section 9.1) addresses critical timing settings for vehicular phases to ensure that cyclists have sufficient clearance time at the end of a phase. It also addresses timing settings for bicycle-specific phases. Signal Progression for Bicycles (Section 9.2) provides signal timing guidance for providing green waves for bicycles, which enables bicycles to arrive on green at successive inter- sections, thereby reducing bicycle delay and stops. Two-Stage Left-Turn Progression for Bicycles (Section 9.3) addresses the traffic signal timing aspect of two-stage left turns with the aim of minimizing delay for left-turning cyclists. While two-stage left turns are a safe mode of turning and are the only practical way of making left turns from protected bike lanes, delay can be large if a cyclist has to wait a long time after the first crossing stage for the phase serving the next stage. Bicycle Detection (Section 9.4) describes technologies that can be used to detect bicycles and refers to timing techniques that become possible with bicycle detection. Bicycle Wait Countdown (Section 9.5) aims to improve cyclists’ red-light compliance by counting down the time until the start of green for bicycle phases, optionally displaying the word “Wait” while the countdown is active. Easing Bicycle Right Turn on Red Restrictions (Section 9.6) aims to reduce cyclist delay and promote equity by legalizing the widespread practice of bicycles turning right on red without stopping (while still yielding to pedestrians and cross traffic).

132 Traffic Signal Control Strategies for Pedestrians and Bicyclists 9.1 Minimum Green and Change Interval Settings for Bicycle Clearance 9.1.1 Basic Description 9.1.1.1 Alternative Names Bike minimum green; bicycle red clearance interval; bicycle yellow change interval; green extension for bike clearance. 9.1.1.2 Description and Objective The objective is to ensure that bicyclists crossing a street have sufficient time to clear the intersection. For bicycles departing from a standing start at the onset of green, this need can be met by using a sufficiently long minimum green interval, called bike minimum green, combined with the bicycle yellow change interval and bicycle red clearance interval. For bicycles departing near the end of the green on a rolling start, this need can be met by using a sufficiently long red clearance interval, called bike red clearance interval. It has also been proposed, though almost never applied, that the clearance need for rolling starts could be met using green extension for bike clearance. For bicycle crossings governed by bicycle signals, this section also covers timing for the bicycle yellow clearance interval. 9.1.1.3 Variations Not applicable for this treatment. 9.1.1.4 Operating Context Bike minimum green should be considered for all signal phases used by bicycles, including left-turn phases. As a practical matter, attention is needed only for actuated phases when cross streets are wide and have a short minimum green, since pretimed and coordinated phases typi- cally have more green time than needed for bicycle clearance. Bike red clearance should be considered for all signal phases used by bicycles. As a practical matter, it is of concern only with long crossings, where the clearance time—which is set based on considerations of vehicle safety—may not be enough for bike clearance. Bicycle yellow interval timing applies wherever there are bike signals, whether for exclusive bicycle phases or for bike phases that run concurrently with a parallel vehicular or pedestrian phase. 9.1.2 Applications and Expected Outcomes 9.1.2.1 National and International Use In the Netherlands, Germany, and several other European countries, bike minimum green and bike red clearance have been routine and required parts of signal timing practice for several decades. In the United States, bike minimum green is recommended in the California Manual on Uniform Traffic Control Devices (CA MUTCD) (State of California, 2014), and it is used in many (but not all) cities in that state. Some U.S. cities outside California also use bike minimum green. However, it has not yet become part of mainstream practice.

Treatments Addressing Special Bicycle Needs 133   Bicycle detectors have been used in some locations in California to apply bicycle minimum green only when a bike is detected by using the stop-line detector in the bike lane that is already present for call detection. Bike red clearance is rarely applied in the U.S. While the practice is recommended in the National Association of City Transportation Officials (NACTO) bikeway design guide, it is not mentioned in the AASHTO Guide for the Development of Bicycle Facilities nor NCHRP Report 812: Signal Timing Manual, 2nd Edition (STM2). There is a common belief that long crossings would demand very long red clearance times, which would be impractical and unsafe to implement from the viewpoint of vehicle operations. A contributing factor is that the stan- dard red clearance formula used in the U.S. makes conservative assumptions that lead to con- siderably more red clearance time than clearance time formulas used in Europe. Bicycle Crossing Time from a Standing Start. A bicycle’s needed crossing time from a standing start is given by Equation 9-1: = + + (9-1)BXT D L v StartupOffsetstanding where BXTstanding = bicycle crossing time from a standing start (s); D = crossing distance (ft) from the queuing position used by bicycles to the end of the most distant travel lane; L = bicycle length (ft), usually taken as 6 ft; v = final bicycle speed (ft/s); and StartupOffset = startup offset (s), incorporating reaction time and acceleration delay. Startup offset represents the extra time needed compared to if acceleration were instantaneous, and thus incorporates both reaction time and acceleration delay. To provide sufficient clearance time for the vast majority of cyclists, one can use either a near- average speed combined with a high-percentile startup offset, or a near-average startup offset combined with a low-percentile speed. Shladover et al. (2011) studied six California intersections and found that median and 15th-percentile speeds were approximately 12 mph and 8.5 mph, respectively, and that median and 90th-percentile startup offsets were approximately 3.5 s and 5.5 s, respectively. The CA MUTCD recommends default values of v = 10 mph (14.7 ft/s) and StartupOffset = 6 s. If, instead, a speed of 8.5 mph (12.5 ft/s) is combined with a less extreme startup offset of 4.5 s, calculated crossing times differ by less than 0.5 s for crossing distances within the range 75–160 ft. The AASHTO manual’s crossing time equation has a different form but essentially reduces to the CA MUTCD equation. Running grade is usually ignored because crossings are typically level; however, where there is a significant upgrade or the road being crossed has a sharp crown, another second or two may be needed. Where the street being crossed has fast and heavy traffic, cyclists may wait further back—particularly where there is poor visibility toward cross traffic—increasing crossing time (Shladover et al., 2009; Shladover et al., 2011). Bike Minimum Green. Two formulas that may be considered for bicycle minimum green are given in Equations 9-2 and 9-3: = − − (9-2)BikeMinGreen BXT Y RClearstanding = − − + − (9-3)BikeMinGreen BXT Y RClear PET tstanding entry

134 Traffic Signal Control Strategies for Pedestrians and Bicyclists where BikeMinGreen = bike minimum green (s); Y = yellow time (s); RClear = red clearance time (s); PET = post-encroachment time (s); and tentry = time needed for the first vehicle released in the next phase to reach the con- flict zone (s). Equation 9-2, used both by the AASHTO guide and the CA MUTCD, aims to ensure that the design bicyclist has reached the end of the most distant travel lane before conflicting traffic is released. STM2, which provides the same guidance regarding bike minimum green as the CA MUTCD, has a table showing the total minimum phase length needed (minimum green plus yellow plus red clearance) as a function of crossing length (see Exhibit 6-7 in STM2). Equation 9-3 demands a shorter minimum green because it accounts for time needed by the first vehicle released in the following phase to enter the conflict zone, a standard consideration in German and Dutch practice. McGee et al. (2012) found that the average entry time was 4.1 s, including an average reaction time of 1.1 s. For design, suggested default values are PET = 1.0 s and tentry= 2.8 s. The latter is based on a near-worst-case scenario in which the bike crossing is only 15 ft from the lead car, the lead vehicle’s acceleration is 9.2 ft/s2 (which is 40% greater than the average acceleration from a start found by Long [2000] for passenger cars), and reaction time is 1.0 s. With these default values, bike minimum green will be 1.8 s shorter using Equation 9-3 versus Equation 9-2. Bicycle minimum green is constraining only when it is greater than the minimum green that would have been applied based on auto needs. To illustrate, consider two crossings: one has a length of 80 ft, yellow time of 3 s, and red clearance time of 2 s, while the other has a length of 120 ft, yellow time of 3 s, and red clearance time of 3 s. Using CA MUTCD default values, bike minimum green would be 6.9 and 8.6 s, respectively. Most agencies use 6 s as the least minimum green for through phases (though many agencies use 10 s), so bike minimum green presents a very minor constraint on signal operations. Bike Red Clearance. The three formulas in Equations 9-4 through 9-6 may be considered for bike red clearance: = + (9-4)BikeRClear D L v ( )= + + + −2 (9-5)BikeRClear D Lv t vd Yreaction ( )= + + + − + −2 (9-6)BikeRClear D Lv t vd Y PET treaction entry where BikeRClear = bike red clearance time (s); treaction = reaction time for a cyclist reacting to a signal turning yellow (s); and d = bike deceleration rate at a traffic signal (ft/s2). Suggested default values are: L = 6 ft; v = 12.5 ft/s (8.5 mph);

Treatments Addressing Special Bicycle Needs 135   treaction = 1.0 s; d = 10 ft/s2 (as suggested in Ontario’s bicycle traffic signals manual); PET = 1.0 s; and tentry = 2.8 s. Equation 9-4 is the clearance time formula typically used for vehicles. It aims to ensure that no conflicting vehicle is released until a bicycle entering the intersection at the last moment of yellow has cleared the most distant travel lane. This formula can lead to rather long bicycle red clearance times; for example, for crossings of 80 ft and 120 ft, the needed clearance times would be 6.9 s and 10.1 s, respectively. Equation 9-5 allows some of the yellow time to count toward bicycle clearance, based on the idea that at the onset of yellow, cyclists who can stop (because they are far enough upstream of the stop line) should stop, consistent with the most common legal meaning of yellow. NACTO recommends using this concept when determining bike clearance time need, as does Ontario’s bicycle traffic signals manual (Province of Ontario, 2018). The time from the start of yellow to the moment the last cyclist who could not stop enters the intersection is ( )+ 2t vdreaction . When using the suggested default values, this is 1.6 s; the balance of the yellow can be used toward needed crossing time. (Faster bikes may enter later in the yellow but need less red clearance time because of their greater speed.) Using this formula reduces needed clearance by 1.4 s if yellow time is 3 s, and by 2.4 s if yellow time is 4 s. Equation 9.6 takes the additional step of accounting for the time needed for the first car released in the following phase to reach the conflict zone, while still allowing a post-encroachment margin (as discussed earlier). Using the suggested default values, the equation further reduces needed clearance by 1.8 s. An additional reduction in needed clearance can be gained by treating the entry point to the intersection for bicycles to be the curb line of the intersecting street rather than the stop line. This can be appropriate where cyclists routinely stop at that curb line rather than at the stop line. (One advantage of the protected intersection layout is that it places the bicycle stop line in this advanced position, reducing needed bicycle clearance.) Together, reductions in needed bicycle red clearance gained by more precisely accounting for cyclists’ clearance need can be substantial. If the yellow time is 3 s and the stop line is set back 18 ft from the intersecting curb line, those reductions would amount to 4.6 s. For an 80 ft crossing, needed bicycle clearance falls from 6.9 to only 2.3 s; for a 120 ft crossing, it falls from 10.1 to 5.5 s. Bicycle red clearance affects a signal’s operation only when it exceeds the red clearance time needed by autos; that difference can be called extra clearance time. Clearance time needed for vehicles is determined by local policy, but a common policy is to equate it to crossing length plus vehicle length (usually taken as 15 ft) divided by the speed limit (in miles per hour). With that policy and a speed limit of 30 mph, red clearance time needed by vehicles is 2.2 s for an 80 ft crossing and 3.1 s for a 120 ft crossing. Continuing the earlier examples, the extra red clearance time needed for bicycles would be 0.1 s for the 80 ft crossing and 2.4 s for the 120 ft crossing. Making Extra Phase Time for Bikes Bike-Actuated. The extra minimum green or red clearance needed to enable bikes to clear an intersection can be provided every cycle or only in cycles in which a bicycle is detected at specific times. In the latter case, minimum green can be provided if a bicycle is detected during the red interval, and red clearance can be provided if a bicycle is detected during the last few seconds of green or the first few seconds of yellow. The extra minimum green or red clearance needed to enable bikes to clear an intersection can be

136 Traffic Signal Control Strategies for Pedestrians and Bicyclists provided every cycle; alternatively, minimum green can be provided only in cycles in which a bicycle is detected during the red interval and red clearance can be provided in the last few sec- onds of green or the first few seconds of yellow. If the extra time needed for bikes is large, bike actuation can allow signal cycles to be more efficient. Except where bikes are in an exclusive path, bike actuation requires selective detection (i.e., the ability to detect a bike as distinct from a motor vehicle). As described later in Section 9.4, there is at least one commercial system using video detection and offering this capability. A few cities use bike detection to trigger longer minimum green intervals. Green Extension for Bike Clearance. To provide safe clearance for bikes arriving on a stale green, the AASHTO bike guide suggests extending the green rather than providing a longer red clearance time. The advantage of this method is that it leaves the red clearance interval unchanged. However, it can only be applied with bike detection; moreover, it cannot use a stop- line detector and instead requires a special upstream detector, located so that a cyclist who has not quite reached the detector at the onset of yellow has enough distance to stop before entering the intersection. Bike Yellow. For bike signals, the appropriate length of the yellow interval can be deter- mined using the standard yellow time formula but with performance parameters that pertain to bikes, as stated in Equation 9-7: = + 2 (9-7)Y t v dreaction Because the length of the yellow interval is for enforcement, including self-enforcement (i.e., giving cyclists feedback on whether they correctly decided whether to stop at the onset of yellow), its calculation should use a high-percentile speed. Using v = 14 mph (20.5 ft/s) and d = 10 ft/s2 yields a bike yellow time of 2.0 s. At the same time, a short bicycle yellow applied to a bike phase that runs concurrently with a vehicular phase leaves more time for bicycle red clearance. This is one of the primary reasons that Amsterdam, Netherlands, recently reduced the yellow time for its bicycle signals from 3 to 2 s citywide (S. Linders, personal communication, August 1, 2018). 9.1.2.2 Benefits and Impacts Bicyclists will benefit from safer and less stressful crossings; benefits will be particularly strong for those who ride slowly, such as children, older adults, tourists, and people new to cycling. Benefits depend on the number of cyclists, so crossings used by high volumes of cyclists or slower cycling populations should be prioritized. Impacts to traffic depend on whether bike clearance settings are applied in all cycles or only in those for which bikes have been detected during the red interval (for bike minimum green) or during the last few seconds of green or first few seconds of yellow (for bike red clearance). With detector-based actuation, the impacts of both bike minimum green and bike red clearance will be trivially small except where bike volume is high, in which case the benefits will be large. Without detector-based actuation, the impact to traffic of applying bike minimum green to through phases will usually be negligible. This is because vehicle green times required for the through phases are typically already longer than the bike minimum green (for most crossings, bike minimum green is 10 s or less). The impact of applying it to minor-street through move- ments could be greater, especially with long crossings and low left-turn volumes. However, even at very long crossings, there will usually be no impact in peak periods because traffic volume is usually large enough that vehicle green times run longer than bike minimum green. During low- traffic periods, the impact will be very small because there is plenty of excess capacity, therefore

Treatments Addressing Special Bicycle Needs 137   increasing a side street’s green duration—for example, from 6 to 10 s—will have a negligible effect on average delay. Using microsimulation, Shladover et al. (2009) found that along the very wide El Camino Real corridor in Palo Alto and Mountain View, CA, with 26 signalized intersec- tions, increasing minimum green times on the side streets from 7 to 11 s to provide a bicycle minimum green had negligible impacts on travel times and queue lengths. For left-turn phases that bikes share with vehicles, it is common to use a minimum green of 4 or 6 s, in which case the impact of imposing a bike minimum green can be greater. Still, the impact is expected to be small because large impacts can only occur when an intersection is operating near capacity, and during those periods, left-turn volume typically dictates a green interval that is longer than bike minimum green. In the U.S., the impact to traffic of providing bike red clearance time has generally been con- sidered so onerous that the concept has been entirely omitted from national manuals. However, as described earlier, more precisely accounting for bike clearance can reduce the extra red clear- ance time. One impact is in the realm of safety; for small increases in clearance time, there may be small improvements in safety. Schattler et al. (2003) found that while adding red clearance time to three intersections did not significantly change the number of related crashes, the number of late exits (i.e., vehicles not clearing an intersection until the next phase has started) significantly declined. However, for large increases in red clearance time, there is widespread concern that it may lead to a large increase in red-light running, with strong, negative safety impacts. The other impact is in efficiency. By introducing additional lost time into the signal cycle, longer red clearance times decrease capacity, increase vehicle delay, and—at intersections that are near capacity—may require a longer cycle length. However, the impact will often be small. Where a wide street is crossed by a side street that is not wide, only the side street would need additional clearance; and if the intersection is part of a coordinated system in which it is not critical, the capacity and delay impacts could be negligible. For crossings whose phase length is governed by pedestrian-crossing needs, using efficient pedestrian clearance settings (see Section 7.4) can make it possible to provide bike red clear- ance without adding lost time by ending the vehicular green earlier while leaving the pedestrian timing unchanged. Impacts can be limited by policy. For example, in Toronto, Ontario, providing bike red clear- ance is part of traffic signal policy, but extra red clearance time is limited to 1 s (Ontario, 2018; Toronto Transportation Services, 2015). And if bike red clearance is bike-actuated, impacts will most likely be negligible. 9.1.3 Considerations 9.1.3.1 Accessibility Considerations Minimum green and red clearance settings that ensure sufficient clearance time for bikes are particularly relevant for slower bicyclists, including children, older adults, and tourists. At crossings used by slower bicycling populations, it may be advisable to measure bike speeds and startup offsets for calculating clearance needs. 9.1.3.2 Guidance The AASHTO bike guide and STM2 provide guidance on minimum green and yellow change interval settings for bicycle clearance.

138 Traffic Signal Control Strategies for Pedestrians and Bicyclists 9.1.3.3 Relationships to Relevant Treatments Bicycle detection (see Section 9.4) can improve the efficiency of this treatment, particularly for providing bike red clearance at long crossings. It is critical that the detector sense bicycles reli- ably. Accuracy in filtering out actuations from motor vehicles (e.g., a right-turning vehicle that encroaches on a bike lane) is not as critical, but inaccuracies will lower the treatment’s efficiency. Using efficient pedestrian clearance settings (see Section 7.4) lowers the impact of increasing bike red clearance when pedestrians are served concurrently with bikes. 9.1.3.4 Other Considerations Bicycle crossing time can be affected by the slope of the approach roadways (which affects bicycle crossing speed); the grade to be overcome during the crossing, including grades due to a sharp crown in the surface of the road being crossed; and the ability of the cyclists to see cross traffic from their starting position. Also, where cross traffic is fast, cyclists often wait further back, increasing their crossing time (Shladover et al., 2009; Shladover et al., 2011). 9.1.4 Implementation Support 9.1.4.1 Equipment Needs and Features No equipment adjustments are needed to change minimum green or red clearance settings in every cycle. Some controllers have a built-in feature to enter a detector-actuated bicycle minimum green. When a bicycle call is detected, the controller will increase the minimum green in the next cycle to the bicycle minimum green. If that feature is not present, it can usually be added. Likewise, controllers will need to be programmed to apply detector-actuated bicycle red clearance. To apply detector-actuation for either treatment, the only detector needed is a standard stop- line call detector, unless it is actuated too frequently by motor vehicles. If that cannot be cor- rected by adjusting the detector’s sensitivity—for example, because the stop-line detector is in a shared travel lane rather than a bike lane or because right-turning vehicles routinely overrun the detector—then detector-actuated application may require a detector capable of distinguishing bikes from motor vehicles, such as a camera-based detector. 9.1.4.2 Phasing and Timing Not applicable for this treatment. 9.1.4.3 Signage and Striping Not applicable for this treatment. 9.1.4.4 Geometric Elements Geometric changes that shorten bike crossings reduce clearance time needs. Examples include advanced stop lines for bikes, corner bulb-outs, road diets (on the street being crossed), and pro- tected intersection layouts. Intersection layouts that guide cyclists to make two-stage left turns eliminate the need for incorporating bicycle clearance needs into left-turn phase timing. Bibliography American Association of State Highway and Transportation Officials. (2012). Guide for the Development of Bicycle Facilities, 4th Edition. Washington, DC.

Treatments Addressing Special Bicycle Needs 139   Curtis, E. J. (2015). Comprehensive on-Street Bicycle Facilities: An Approach for Incorporating Traffic Signal Opera- tional Strategies for Bicycles. [Doctoral dissertation, Georgia Institute of Technology]. Long, G. (2000). Acceleration Characteristics of Starting Vehicles. Transportation Research Record: Journal of the Transportation Research Board, 1737(1), 58–70. McGee, H., Sr., Moriarty, K., & Gates, T. J. (2012). Guidelines for Timing Yellow and Red Intervals at Signalized Intersections. Transportation Research Record: Journal of the Transportation Research Board, 2298(1), 1–8. Province of Ontario. (2018). Bicycle Traffic Signals: Traffic Manual Book 12A. Rubins, D. I., & Handy, S. (2005). Bicycle Clearance Times: A Case Study of the City of Davis. Presented at 84th Annual Meeting of the Transportation Research Board, Washington, DC. Schattler, K. L., Datta, T. K., & Hill, C. L. (2003). Change and Clearance Interval Design on Red-Light Running and Late Exits. Transportation Research Record: Journal of the Transportation Research Board, 1856(1), 193–201. Shladover, S. E., Kim, Z., Cao, M., Sharafsaleh, A., & Li, J.-Q. (2009). Bicyclist Intersection Crossing Times: Quantitative Measurements for Selecting Signal Timing. Transportation Research Record: Journal of the Transportation Research Board, 2128(1), 86–95. Shladover, S. E., Kim, Z., Cao, M., Sharafsaleh, A., & Johnston, S. (2011). Intersection Crossing Times of Bicyclists: Quantitative Measurements at Diverse Intersections. Transportation Research Record: Journal of the Transportation Research Board, 2247(1), 91–98. State of California. (2014). California Manual on Uniform Traffic Control Devices. Taylor, D. B. (1993). Analysis of Traffic Signal Clearance Interval Requirements for Bicycle-Automobile Mixed Traffic. Transportation Research Record: Journal of the Transportation Research Board, 1405, 13–20. Toronto Transportation Services. (2015). Traffic Signal Operation Policies and Strategies. Urbanik, T., Tanaka, A., Lozner, B., Lindstrom, E., Lee, K., Quayle, S., Beaird, S., Tsoi, S., Ryus, P., Gettman, D., Sunkari, S., Balke, K., & Bullock, D. (2015). NCHRP Report 812: Signal Timing Manual, 2nd Edition. Trans- portation Research Board, Washington, DC. Wachtel, A., Forester, J., & Pelz, D. (1995). Signal Clearance Timing for Bicyclists. ITE Journal, 65(3), 38–38. 9.2 Signal Progression for Bicycles 9.2.1 Basic Description 9.2.1.1 Alternative Names Green wave for bicycles. 9.2.1.2 Description and Objective Traffic signals can be coordinated so that bicycles arrive at successive signals during a green phase and therefore pass through without delay. This is accomplished by choosing signal offsets that closely match bicycle speed. A signal offset for a given intersection is how much later its green begins than the green of a reference intersection, which is usually the first intersection in the series and can be numbered Intersection 1. To determine offsets on a one-way street, a progression speed is first chosen, then offsets for each successive intersection ( j = 2, 3, . . .) are determined using the formula in Equation 9-8: offset d v1 j 1 j progression = (9-8) where offset1j = offset of signal j relative to signal 1 (the signal at Intersection 1) (s); d1j = distance from signal 1 to signal j (ft); and vprogression = progression speed (ft/s). For example, if signals are 360 ft apart and the progression speed is 18 ft/s (12.3 mph), then beginning with an offset of 0 s at the first intersection, offsets for successive intersections are 20 s, 40 s, 60 s, etc.

140 Traffic Signal Control Strategies for Pedestrians and Bicyclists On two-way streets, one-way coordination can be applied to a favored direction, which can switch by time of day; the other direction usually gets poor progression as a result (e.g., bicycles can typically get through only a few intersections before they get a red). Alternatively, signals can be timed for two-way coordination. Offsets for two-way coordination cannot be determined by formula (except for special cases); they are instead determined using signal timing software. The quality of two-way progression depends on intersection spacing and cycle length. An ideal signal progression speed for bicycles is about 2 mph faster than average bicycle speed, or 12 to 13.5 mph (roughly 17.5 to 20 ft/s) on level ground. This avoids inhibiting faster bicycles and promotes signal compliance, while still enabling slower bicycles to stay in the green wave for a considerable distance. Grades should also be considered when determining ideal progression speed for bicycles. 9.2.1.3 Variations Coordination can be one-way or two-way, as described earlier. Small-zone coordination is a coordination scheme affecting only a few intersections, usually focused on a critical intersection at which bikes get only a short green period. Neighboring sig- nals are timed so that bicycles released from an upstream signal arrive at the critical intersection just in time for its green, and bicycles released from the critical intersection can pass through downstream intersections without stopping. Corridor- or grid-level coordination involves timing signals so that bicycles have a green wave along a corridor or along all the streets in a grid. 9.2.1.4 Operating Context Main bicycle routes along streets with closely spaced traffic signals are good candidates for corridor-level coordination. Grid-level coordination can be applied where there is a grid of closely spaced intersections, such as in some downtowns. Small-zone coordination can be a good treatment if an important bicycle route passes through a critical intersection with a long cycle length and a short green phase for bicycles that is near other traffic signals on the same route. There is little benefit to implementing coordination where signal spacing is more than 1,200 ft apart since bicycle platoons disperse due to bicyclists’ varying speeds. 9.2.2 Applications and Expected Outcomes 9.2.2.1 National and International Use Both Copenhagen, Denmark, and Amsterdam, Netherlands, have green waves for bicycles on major bicycle routes. Copenhagen has bicycle green waves on several streets, of which the best known is NØrrebrogade. This is a very heavily used bicycle route on a historic, narrow road that has been prioritized for bikes—through-auto traffic is prevented by closing several blocks to autos, allowing bicycles and buses only. Signals are timed for one-way coordination (inbound in the morning, outbound in the evening) with a progression speed of 20 km/h (12.5 mph) (Colville-Andersen, 2014). On Raadhuisstraat, a street in Amsterdam, 11 signals over a span of 650 m (0.4 mi) are timed with a progression speed of 18 km/h (11 mph) (Linders, 2013). They use one-way coordina- tion, with a green wave inbound in the morning and outbound in the evening. This street has a considerable amount of auto traffic, and bikes on this street have a conventional bike lane, with bike demand so great that bikes often spill out into the adjacent travel lane. The green wave not

Treatments Addressing Special Bicycle Needs 141   only reduces bicycle delay but also improves safety by eectively limiting motor vehicle speed to 18 km/h, removing the incentive for motor vehicles to pass bicycles. Copenhagen and Utrecht, Netherlands, have installed pilot projects that provide cyclists with advice on whether to speed up or slow down to make the next green phase. However, such sys- tems cost a lot more than retiming signals, and both Copenhagen and Amsterdam have found that when intersection spacing is not large, cyclists quickly learn what speed will enable them to stay within the green wave (Colville-Andersen, 2014; Linders, 2013). In the United States, New York City, San Francisco, CA, and Chicago, IL, have retimed streets for bicycle progression. New York’s rst application is the one-way couplet of Hoyt Street and Bond Street in Brooklyn, New York (see Exhibit 9-1), retimed in December 2018 with a progres- sion speed of 15 mph (Colon, 2019; D. Nguyen, personal communication, 2019). A safety reason for this treatment is that cyclists going downhill (on Hoyt) were reluctant to stop at red signals, and a few red-light runners crashed into vehicles. Timing the signals so that through bicycles arrive on green eliminates most of that safety problem. In San Francisco, Valencia Street and Folsom Street—both two-lane, two-way streets in a mixed-use context—were retimed for progression at 13 mph. e signal spacing is approxi- mately 600  with a signal cycle of 60 s. While ideal one-way progression can be achieved with any progression speed, only one progression speed results in ideal green waves for two-way progression, as shown in Equation 9-9: ( )=- 2 (9-9)ideal two way progression speed signal spacing cycle length For Valencia Street and Folsom Street, that speed is 600/30 = 20 /s or 13.5 mph. With this design, bicycles get green waves in both directions. For bikes, this is an ideal solution. To some degree, nding an ideal progression speed that matches cyclists’ speeds is lucky—if the cycle length had been 100 s instead or if signal spacing were 360 , the ideal progression speed would have been only 8 mph. Source: Kuntzman (2020). Exhibit 9-1. Hoyt Street in Brooklyn, timed for a bicycle green wave at 15 mph.

142 Traffic Signal Control Strategies for Pedestrians and Bicyclists For vehicles as well as bicycles, the “ideal” progression speed is 13.5 mph (calculated using the same formula as Equation 9-9); this means that if signals were timed for a speed better suited to vehicles—for example, 25 or 27 mph—coordination would not be ideal and vehicles would still have to stop frequently, except during light traffic conditions. Instead, they can advance at a slow but steady speed of 13.5 mph. This also meets the desired driving regime for Valencia Street, as it is not meant to support through traffic (nearby parallel arterials have that function). Rather, it supports local traffic circulation with slower speeds. In addition, San Francisco has recently retimed signals for bicycle green waves on seven streets, using a 15 mph progression speed. Six of the seven streets are two-way, and intersection spacing is such that bicycles get green waves in both directions (Stonehill, 2016). The downtown in Portland, OR, is a one-way grid with square blocks, 260 ft on each side. The signal cycle throughout the downtown is 56 s (60 s in peak hours). In a one-way grid, the progression speed that results in ideal one-way green waves in all four directions is given in Equation 9-10: =- (9-10)ideal progression speed (one way grid) block circumference signal cycle length For Portland’s downtown, this is 12.7 mph (and 11.8 mph in peak hours). Portland has timed its downtown signals with this progression speed for decades, well before bicycling was popular. This regime serves bicycles well. This timing regime, known as quarter cycle offsets, offers good service for vehicles too; drivers have to go slowly but, in return, they can drive north, south, east, and west with almost no signal delay. Portland also has an example of small-zone coordination for bicycles. On NE Broadway, a one-way street, bicycles receive a short green period at North Williams Avenue because the bicycle movement has to split time with a very heavy right-turn movement. To avoid the long bicyclist delay, the intersection is coordinated for bicycle progression with the upstream (NE Victoria Avenue) and downstream (North Vancouver Avenue) intersections. When bicycles are released from Victoria, a bicycle phase at Williams becomes scheduled to begin a few sec- onds later; it is actuated, so a bike phase at Williams is only called if a bicycle is detected just downstream of Victoria. The same combination of coordinated timing and actuation happens at Vancouver. Bicycles that pass Victoria on a stale green may not catch the green wave at Williams; however, once released at Williams, they will have a green wave through Vancouver. Overall, bicycles stop at most once through this set of three intersections. Littman and Furth (2013) show the phasing plan and a simulation video explaining how the coordination works. 9.2.2.2 Benefits and Impacts For bicycles, enhanced coordination results in less delay. In addition, crash types associated with red-light running can be expected to go down when coordination enables most bicycles to arrive on green. Small-zone coordination can be a critical component of safety projects. At intersections where the bicycle green phase is shortened to provide a protected-only turn phase to protect bicycles from turn conflicts, small-zone coordination can reduce bicycle delay and improve compliance, which also improves safety. For example, at Portland’s NE Broadway and North Williams Avenue intersection, bicycles had previously crossed Williams with the vehicular phase, which involved a permitted conflict with heavy right-turn flow. It was changed to a concurrent-protected crossing (see Section 6.2), which protects bicycles from the turn conflict but at the same time drastically reduces bicyclists’ green time. By providing a small-coordination zone for bikes, the Portland Bureau of Transportation kept the increase in bicycle delay small in spite of that greatly reduced green time, which helped ensure compliance and was critical to improving safety (Littman & Furth, 2013).

Treatments Addressing Special Bicycle Needs 143   Lower progression speeds in combination with short cycle lengths (see Section 7.1) reduce speeding opportunities, thereby reducing extreme speeds, especially during periods of low traffic (Furth et al., 2018). This brings general safety benefits for pedestrians crossing the street (including those crossing away from the traffic signals) and for cyclists. On one-way streets that are independently timed (i.e., not timed as part of a grid), ideal pro- gression can be achieved for any desired progression speed. While bringing progression speed closer to bicycle speed on such streets will decrease bicycle delay by enabling bicycle progression, it will also increase delay for motor vehicles. Analysis conducted for this guidebook suggests that the delay increase to motor vehicles will typically exceed the delay reduction to bicycles. Exhibit 9-2 shows how bicycle delay and auto delay vary with progression speed on a one-way street whose cycle length is 90 s, with 40 s of green for the street of interest and with signalized intersections 0.1 mi apart. Bicycle speed is assumed to be in a range centered around 11 mph, and target auto speed is taken to be 25 mph. As progression speed is lowered, delay reduction for bicycles stays small until progression speed gets within approximately 6 mph of bicycle speed (i.e., 17 mph in this example). Mean- while, auto delay increases considerably when progression speed is reduced below the target speed. Thus, lowering progression speed to reduce bicycle delay on independently timed one- way streets does not appear to be a promising strategy unless a street has been prioritized for bicycling or there are other reasons that favor a reduction in progression speed (e.g., a goal of discouraging auto traffic). 9.2.3 Considerations 9.2.3.1 Accessibility Considerations Not applicable for this treatment. 9.2.3.2 Guidance Not applicable for this treatment. 0 2 4 6 8 10 12 14 16 18 20 0 5 10 15 20 25 30 35 A ve ra ge d el ay p er in te rs ec tio n (s ) Progression speed (mph) Bicycle Delay Auto Delay Exhibit 9-2. Through-bicycle and through-auto delays per intersection versus progression speed for an uncongested one-way street with 25 mph target speed.

144 Traffic Signal Control Strategies for Pedestrians and Bicyclists 9.2.3.3 Relationships to Relevant Treatments Bicycle coordination should be considered in combination with treatments that substantially reduce bicycle green time to eliminate turn conflicts (e.g., protected-concurrent crossings) and reduce bicycle delay. Combining low progression speeds with shorter cycle lengths reduces speeding opportunities on a street, reducing extreme speeding for vehicles. 9.2.3.4 Other Considerations Not applicable for this treatment. 9.2.4 Implementation Support 9.2.4.1 Equipment Needs and Features Not applicable for this treatment. 9.2.4.2 Phasing and Timing Section 9.2.2.1 discusses signal timing considerations for providing signal progression for bicycles. 9.2.4.3 Signage and Striping Not applicable for this treatment. 9.2.4.4 Geometric Elements Not applicable for this treatment. Bibliography Colville-Andersen, M. (2014, August 5). The Green Waves of Copenhagen. Copenhagenize. http://www. copenhagenize.com/2014/08/the-green-waves-of-copenhagen.html Colon, D. (2019, October 23). The ‘Green Wave’ is a Ripple—But Trottenberg Battles Car Entitlement. StreetsBlog NYC. https://nyc.streetsblog.org/2019/10/23/the-green-wave-is-a-ripple-but-trottenberg-finally-pushes- back-on-car-entitlement/ Furth, P. G., Halawani, A. T. M., Li, J., Hu, W., & Cesme, B. (2018). Using Traffic Signal Control to Limit Speeding Opportunities on Bidirectional Urban Arterials. Transportation Research Record: Journal of the Transporta- tion Research Board, 2672(18), 107–116. Kuntzman, G. (2020, June 4). DOT Expands ‘Green Wave’ for Cyclists as City Prepares to Reopen. StreetsBlog NYC. https://nyc.streetsblog.org/2020/06/04/dot-expands-green-wave-for-cyclists-as-city-prepares-to- reopen/ Linders, S. (2013). STOP: 100 jaar verkeer regelen in Amsterdam 1912–2012. City of Amsterdam, The Netherlands. Littman, M., & Furth, P. G. (2013). Protecting the Bike – Dual Right Turn Lane Conflict in Portland (NE Broadway & Williams) [Student project, Northeastern University]. https://sites.google.com/site/studentprojectstraffic/ home/protecting-the-bike-right-turn-conflict-in-portland-broadway-ne-williams Stonehill, L. (2016). Catch the Green Wave! Timing Signals for Bikes [Poster]. ProWalk ProBike ProPlace Con- ference, Vancouver, BC. 9.3 Two-Stage Left-Turn Progression for Bicycles 9.3.1 Basic Description 9.3.1.1 Alternative Names Pedestrian-style left turn; hook turn; box turn; Copenhagen left.

Treatments Addressing Special Bicycle Needs 145   9.3.1.2 Description and Objective To turn le at an intersection, bicyclists have the choice of making a vehicle-style turn, which typically involves shared use of a vehicular le-turn lane, or making a pedestrian-style turn, executed as a pair of simple crossings. In Exhibit 9-3, a northbound le (NBL) bicycle would rst make Crossing (a) when the northbound movement has the green, wait in the far corner, and then make Crossing (b) when the westbound movement has the green. Two-stage le turns are common in many European countries and are becoming popular in the United States. NACTO’s Urban Bikeway Design Guide (2012) has guidance on their geo- metric design, including how to mark two-stage queuing boxes—also known as bicycle turn boxes—where cyclists wait between simple crossings. is treatment focuses on the signal timing aspects of two-stage le turns. If there is good signal progression between the two phases of a turn, bicycle delay can be moderate; however, without any progression, delay to le-turning cyclists can be very large. Exhibit 9-4 provides an example with good progression for bicycles turning le from O Street to D Street. By providing a leading le turn for O Street and a lagging le turn for D Street, the through movement for O Street is immediately succeeded by the through movement for D Street. However, this phase sequence results in poor progression for left turns beginning on D Street; this makes it a good solution if le-turn demand from one street dominates, as might occur at a skew angle intersection. Where there is a full set of le-turn phases, it is not possible to provide good progression for all le turns as long as the crossings are unidirectional. Making bicycle crossings bidirectional gives cyclists a choice in how to make their crossing and creates new opportunities for providing le-turn progression, as explained in the Dutch bikeway design guide (CROW, 2016). In the example shown in Exhibit  9-5, le turns are sequenced (as in Exhibit 9-4) so that the through movements for one street (in this case, the east– west street) are immediately followed by the other street’s through movements. With bidirectional crossings (also shown in Exhibit 9-5), this phasing sequence creates good progression for all Exhibit 9-3. Unidirectional bicycle crossings for making a two-stage left turn: Cross at (a) then (b) for northbound left (NBL); (b) then (c) for westbound left (WBL); (c) then (d) for southbound left (SBL); and (d) then (a) for eastbound left (EBL).

Exhibit 9-4. Lead-lag progression for bicycles traveling from O Street to D Street. Exhibit 9-5. Bidirectional bicycle crossings, allowing for additional crossing movements.

Treatments Addressing Special Bicycle Needs 147   bicycle left turns, as every left turn becomes possible by first crossing with the east–west street and then finishing with the north–south street. Left-turning bicyclists approaching the intersec- tion at Crossings A and C can make their left turn by traveling counterclockwise (following the green arrows), and those approaching at Crossings B and D can make their left turn by traveling clockwise (following the blue arrows). However, cyclists are not obligated to follow the direc- tions with good progression; they minimize their delay by choosing the first crossing at their arrival corner that gets a green light. 9.3.1.3 Variations Not applicable for this treatment. 9.3.1.4 Operating Context This treatment is applicable wherever bicycles make two-stage left turns and the phasing plan includes left-turn phases. It is particularly important where a main bicycle route turns or switches from one side of a street to the other or switches from a bidirectional path on one side of a street to unidirectional paths on the other, forcing a large stream of bicyclists to make two- stage turns. 9.3.2 Applications and Expected Outcomes 9.3.2.1 National and International Use In Denmark, two-stage left turns are mandatory since vehicular-style left turns are prohib- ited by law, except between local streets that lack a centerline. In the Netherlands, while there is no such prohibition, road design guidelines require provision for two-stage left turns at traffic signals. Dutch traffic planning emphasizes the need to provide good progression for two-stage turns, and standard intersection analysis includes measuring delay for two-stage left turns. Intersections there often have phasing plans and bidirectional crossings for improving two- stage turn progression. Wagenbuur (2014) describes an example in ‘s-Hertogenbosch, the Netherlands, in which a bidirectional bike path on one side of a street transitions to unidirectional paths on either side of the street, forcing a large volume of bicyclists to make a two-stage turn. To reduce delay, crossings are bidirectional, giving bicyclists a choice of two routes: one beginning with a north- bound crossing and the other beginning with a westbound crossing. A dynamic display has been set up pointing bicyclists to the crossing that will next get a green signal. 9.3.2.2 Benefits and Impacts To assess the benefit of this treatment, one needs to be able to measure average delay for two-stage turns, something that is still new to American practice (see Section 3.3). Delay for a two-stage turn is not equal to the sum of the average delay of the two crossings that make up the turn—it may be considerably greater or lower, depending on the progression, and it can be far less where bidirectional crossings create two path options for people turning left. The North- eastern University Ped and Bike Crossing Delay Calculator is a freely available tool that can be used to determine average delay for two-stage turns, including those with bidirectional crossings (Furth & Wang, 2015). By using a phasing sequence that creates good progression for two-stage turns, bicycle delay can be substantially reduced. In one example, Furth et al. (2019) show that with bidirectional crossings and a favorable phasing sequence, two-stage left-turn delay can be comparable to delay for bikes making a single-stage crossing.

148 Traffic Signal Control Strategies for Pedestrians and Bicyclists Changing phase sequence to improve bicycle turn progression will have no impact on traffic operations at many intersections, while at others it may affect arterial progression and thereby increase auto delay. 9.3.3 Considerations 9.3.3.1 Accessibility Considerations Not applicable for this treatment. 9.3.3.2 Guidance Not applicable for this treatment. 9.3.3.3 Relationships to Relevant Treatments Providing short cycle lengths (see Section 7.1) will help further reduce bicycle delay at two- stage left turns. Two-stage left turns with good progression can be an alternative to an exclusive diagonal bicycle phase (in which a bike path crosses from one side of the street to another or transitions between a bidirectional path and one-way bike lanes) (see Section 6.3). 9.3.3.4 Other Considerations Not applicable for this treatment. 9.3.4 Implementation Support 9.3.4.1 Equipment Needs and Features Not applicable for this treatment. 9.3.4.2 Phasing and Timing Section 9.3.2.1 provides phasing and timing guidance. 9.3.4.3 Signage and Striping Both NACTO and the Manual on Uniform Traffic Control Devices provide guidance on how to design two-stage turn boxes, with detailed information on pavement markings and striping. 9.3.4.4 Geometrics Bidirectional crossings must be wider than unidirectional crossings. Queuing areas in the corners of an intersection should be large enough to hold waiting bicyclists, including those waiting to make the second stage of their turn. Bibliography CROW. (2016). Design Manual for Bicycle Traffic. Ede, Netherlands. Furth, P. G., & Wang, Y. (2015). Delay Estimation and Signal Timing Design Techniques for Multi-Stage Pedes- trian Crossings and Two-Stage Bicycle Left Turns. Transportation Research Board 94th Annual Meeting (No. 15-5365). Furth, P. G., Wang, Y., & Santos, M. A. (2019). Multi-Stage Pedestrian Crossings and Two-Stage Bicycle Turns: Delay Estimation and Signal Timing Techniques for Limiting Pedestrian and Bicycle Delay. Journal of Trans- portation Technologies, 9, 489–503. National Association of City Transportation Officials. (2012). Urban Bikeway Design Guide, 2nd Edition. New York. Wagenbuur, M. (2014, October 9). Dynamic Sign to Indicate the Fastest Cycle Route. Bicycle Dutch. https:// bicycledutch.wordpress.com/2014/10/09/dynamic-sign-to-indicate-the-fastest-cycle-route/

Treatments Addressing Special Bicycle Needs 149   9.4 Bicycle Detection 9.4.1 Basic Description 9.4.1.1 Alternative Names Bicycle actuation. 9.4.1.2 Description and Objective In bicycle detection, the presence of bicyclists is made known to the signal controller, which can use that information to activate the signal for phases that serve bikes, including calling for service (see Section 7.5), extending the minimum green interval (see Section 9.1), extending red clearance (see Section 9.1), and extending the green for bicycle clearance (see Section 9.1). The objective is to make traffic signal operations more efficient by running or extending phases for bikes only when needed, rather than in every cycle. A secondary benefit is that bicycle detec- tors can be used to collect data on bicycle use that can be archived and used for many purposes, including monitoring bicycle-use trends and signal timing design. For some applications, calling for a phase that bicycles share with autos—that is, picking up both bicycles and autos without distinguishing them, also known as mixed detection— is sufficient. For actuating bicycle-specific phases and extensions, exclusive detection (detecting bicycles only) or selective detection (detecting bicycles from within a mixed traffic stream) is necessary. 9.4.1.3 Variations A variety of technologies can be used for bicycle detection. Inductive loops, buried in the pave- ment, detect the presence of metal. Their shape and sensitivity can be tailored to maximize the chance that a bicycle is detected while minimizing false calls from vehicles in an adjacent lane. In a lane shared with autos, they provide only mixed detection, while in a physically separated bike lane, they provide exclusive detection. In conventional bike lanes, their ability to provide exclusive detection depends on whether autos encroach in the bike lane, such as when making right turns. For extending minimum green and red clearance times, occasional encroachment can be tolerated. Video with image processing can be a powerful method for selective detection. At night, image processing algorithms oriented to autos rely on identifying headlights; different algo- rithms are needed to identify bicycles. Bicycles can also be detected using pushbuttons, although this method is inconvenient for bicyclists except when used as a backup for another automatic detection method. Bicycle detection can be considered near the stop bar or upstream from a given intersection (i.e., in advance). Stop-bar placement or advanced detection is appropriate for calling for ser- vice, extending minimum green, or extending red clearance; advanced detection is needed to provide green extension for bike clearance. 9.4.1.4 Operating Context Call detectors that can detect bicycles are needed wherever the phases that serve bikes are actuated, including both bicycle phases and mixed traffic phases. Bicycle detectors with exclusive or selective detection can be considered for all intersec- tion approaches used by bicycles in conjunction with the treatments that are based on bicycle clearance (bike minimum green, bike red clearance, and green extension for bike clearance), as described in Section 9.1. While the first two of these treatments can also be applied without detectors, using exclusive or selective detection makes it possible to apply those extensions only in cycles in which a bicycle is present to benefit from it.

150 Trafc Signal Control Strategies for Pedestrians and Bicyclists 9.4.2 Applications and Expected Outcomes 9.4.2.1 National and International Use Using bicycle-sensitive call detectors for phases shared by bicycles and autos is common in the United States. Where bikes share a lane with motor trac, a special loop layout resembling a gure eight is used to increase a detector’s sensitivity to the relatively small amount of metal in a bicycle. In keeping with an option described in the Manual on Uniform Trac Control Devices (MUTCD) (2009), a small bicycle silhouette with a broken line can be marked on the part of the detector with maximum sensitivity, accompanied by an explanatory sign (as shown in Exhibit 9-6). One weakness of this approach is that the combination of marking and sign is not well understood; another is that the small bike silhouette marking oen does not survive winter snow removal operations, and if it is not replaced, the sign becomes meaningless. Based on public outreach, Portland developed the marking shown in Exhibit 9-7 for use with inductive loops. In this design, the information is provided as part of the marking rather than in an accompanying sign. For detectors in bike lanes, STM2 recommends setting the detector back a short distance to minimize actuations from vehicles that are turning right. In the Netherlands, small inductive loops near the stop bar are used to detect bicycles in physi- cally separated bike lanes, with pushbutton backup (see photo in Section 8.2). A call indicator lights up when the bicycle is detected, so bicyclists know that if the loop detector senses them, they do not need to push the pushbutton. Some intersections have an upstream call detector as well so that, during periods of light trac, a bicyclist detected at the upstream detector may get a green light without having to stop. In the U.S., Portland has been a leader in developing and applying inductive loop detectors for separated bike lanes. Portland also use pushbuttons (without loops) for bicycle detection, and has used educational campaigns to help cyclists better understand how to be detected. In the Netherlands, an app called “Schwung” has been developed that enables signal control- lers to detect approaching bikes using cyclists’ smartphones several seconds before they arrive Source: MUTCD (2009), Figures 9B-2 and 9C-7. Exhibit 9-6. MUTCD marking and sign for inductive loop bicycle detectors.

Treatments Addressing Special Bicycle Needs 151   at an intersection. At the same time, controller logic has been modied to use this information to more quickly change phases in favor of bikes and to extend bike phases (Wagenbuur, 2018). 9.4.2.2 Benets and Impacts e eectiveness of inductive loop detectors in shared lanes depends on whether bicycles position themselves where the detector has greatest sensitivity. Research has found that many bicyclists do not understand the meaning of the MUTCD’s “9C-7” marking and corresponding sign shown earlier in Exhibit 9-6 (Boot et al., 2013; Bussey, 2013). Making matters worse, the zone of heightened sensitivity in the standard layout of a bike-sensitive loop detector is the middle of a lane, while cyclists tend to ride near the right edge of a lane. Boudart et al. (2016) found that 60% of survey respondents understood the meaning of Portland’s marking (shown earlier in Exhibit 9-7), a 30% improvement over locations that did not have the explanation provided as part of the pavement marking. A literature review done by Portland found that advanced bicycle detection (using an auto- mated, passive system) reduces bicycle delay but was not associated with any safety impact (City of Portland, 2010). Bicycle detection using active infrared, video image processing, and cell phone app technolo- gies are developing rapidly, but at this time they have not been well documented. 9.4.3 Considerations 9.4.3.1 Accessibility Considerations Pushbuttons intended for bicyclists should be reachable without requiring that they deviate from their path. Path deviations that require bicyclists to shi laterally are dicult for those with limited strength or balance and for those who are carrying children. Pushbuttons should be set back enough from the curb that the bicycle will not encroach on the conicting roadway (for more detail, see Section 8.3). Source: Maus (2016). Exhibit 9-7. A bicycle loop detector marking developed in Portland.

152 Traffic Signal Control Strategies for Pedestrians and Bicyclists 9.4.3.2 Guidance Not applicable for this treatment. 9.4.3.3 Relationships to Relevant Treatments Pedestrian recall versus actuation (see Section 7.5) discusses considerations that are also relevant to bicycle phases being actuated. When implementing bicycle detection, it is desirable to provide call indicators (see Section 8.2) to assure bicyclists that they have been detected, which may improve compliance. Pushbuttons should be located for safety, convenience, and accessibility (see Section 8.3). Minimum green and change interval settings (see Section 9.1), including bicycle red clearance, can be applied more efficiently using bicycle detection. Flashing indicators for permitted conflicts (see Section 6.9) could be dynamic, that is, activated only when a bicycle is detected. 9.4.3.4 Other Considerations Real-time applications of active infrared detection and video image processing are not well documented or researched. As the technology for these methods advances, there will be more opportunities to include them for bicycle detection at signalized intersections instead of just for bicycle counting purposes. 9.4.4 Implementation Support 9.4.4.1 Equipment Needs and Features Most modern traffic signal controllers allow for bicycle detection inputs. 9.4.4.2 Phasing and Timing Not applicable for this treatment. 9.4.4.3 Signage and Striping Section 9.4.2.1 provides examples of signing and striping for bicycle detection. 9.4.4.4 Geometric Elements Not applicable for this treatment. Bibliography Boot, W., Charness, N., Stothart, C., Fox, M., Mitchum, A., Lupton, H., & Landbeck, R. (2013). Aging Road User, Bicyclist, and Pedestrian Safety: Effective Bicycling Signs and Preventing Left-Turn Crashes. Final Technical Report No. BDK83 977-15. Florida Department of Transportation, Tallahassee. Boudart, J., Foster, N., Koonce, P., Maus, J., & Okimoto, L. (2016). Improving the Bicycle Detection Pavement Marking Symbols to Increase Comprehension at Traffic Signals. Presented at 95th Annual Meeting of the Transportation Research Board, Washington, DC. Bussey, S. W. (2013). The Effect of the Bicycle Detector Symbol and R10-22 Sign on Cyclist Queuing Position at Signalized Intersections. City of Portland. (2010). Portland Bicycle Plan for 2030. Grembek, O., Kurzhanskiy, A. A., Medury, A., Varaiya, P., & Yu, M. (2018). Introducing an Intelligent Inter- section. ITS Reports, 2018(13).

Treatments Addressing Special Bicycle Needs 153   Manual on Uniform Trac Control Devices for Streets and Highways. (2009). FHWA, U.S. DOT. http://mutcd. wa.dot.gov/ Maus, J. (2016). Want More Green? Help the Transportation Bureau Make Signals Better for Cycling. BikePortland. https://bikeportland.org/2016/02/16/want-that-green-light-help-us-tell-pbot-what-works-best-171624 Urbanik, T., Tanaka, A., Lozner, B., Lindstrom, E., Lee, K., Quayle, S., Beaird, S., Tsoi, S., Ryus, P., Gettman, D., Sunkari, S., Balke, K., & Bullock, D. (2015). NCHRP Report 812: Signal Timing Manual, 2nd Edition. Trans- portation Research Board, Washington, DC. Wagenbuur, M. (2018, March 20). Get a Green Light Quicker With Schwung. Bicycle Dutch. https://bicycledutch. wordpress.com/2018/03/20/get-a-green-light-quicker-with-schwung/ 9.5 Bicycle Wait Countdown 9.5.1 Basic Description 9.5.1.1 Alternative Names Wait signal; bicycle red countdown. 9.5.1.2 Description and Objective Bicycle wait countdown devices indicate the time remaining in the bicycle red period, oen in combination with displaying the word “Wait.” ey aim to improve cyclist red-light compliance and to reduce the anxiety associated with waiting. 9.5.1.3 Variations e countdown can have a gurative display, which shows an arc of lights becoming suc- cessively shorter (see Exhibit 9-8), or it can have a digital display, which shows the number of seconds remaining in the red period (see Exhibit 9-9). e gurative display is more appropriate with fully actuated control since the time remaining in the red interval is not known. Source: Peter J. Koonce. Exhibit 9-8. Wait signals in Amsterdam (left) and Portland (right), with time shown as a shrinking arc. “Wacht” is Dutch for “Wait.”

154 Trafc Signal Control Strategies for Pedestrians and Bicyclists 9.5.1.4 Operating Context Bicycle wait countdown signals may be an appropriate treatment for bicycle signals when there is a desire to increase cyclist compliance with the red signals. 9.5.2 Applications and Expected Outcomes 9.5.2.1 National and International Use Amsterdam introduced wait signals in order to reduce noncompliance with red signals. ey are now used at more than 50 intersections in the city. Some use numeric countdowns; others show a shrinking arc. Digital countdowns stop 5 s before bicycle signals turn green so that bicy- clists will look to the bicycle signal, not the countdown, for their cue to depart. In North America, the only known application of this treatment is in Portland (see Exhibit 9-8). At this intersection, the red period varies in length from cycle to cycle, so the “time remaining” indicator remains at the “full” position during the cross street’s green and declines during the cross street’s yellow interval and red clearance interval. 9.5.2.2 Benets and Impacts According to Fong et al., a 2003 international scan sponsored by FHWA noted that the use of shrinking-arc wait countdowns in the Netherlands reduced bicycle red-light running by 25% to 30%. Additionally, the same scan revealed that 60% of users thought their waiting time was shorter, and 78% of users found the bicycle wait countdown information helpful. However, Amsterdam ocials report, based on an internal study, that wait countdowns led to no net change in bicycle red-light running. While there was some reduction in bicycles departing early during the red period, it was oset by an increase in bicycles departing late during the red period. Nevertheless, there is a general perception that shrinking-arc wait countdowns improve compliance; therefore, the transportation department oen receives requests to install them (S. Linders, personal communication, 2018). 9.5.3 Considerations 9.5.3.1 Accessibility Considerations Not applicable for this treatment. Source: Colville-Andersen (2014). Exhibit 9-9. Numerical wait countdown in Copenhagen.

Treatments Addressing Special Bicycle Needs 155   9.5.3.2 Guidance Not applicable for this treatment. 9.5.3.3 Relationships to Relevant Treatments Not applicable for this treatment. 9.5.3.4 Other Considerations When a wait countdown device is visible to waiting motorists, there is concern about whether motorists might use this information and what its impact might be. In Amsterdam and Portland, using a shrinking-arc display—or in the case of a numeric display, halting the countdown at “5”—seems to have avoided any such issue. 9.5.4 Implementation Support 9.5.4.1 Equipment Needs and Features The wait/countdown device should be mounted next to the bicycle signal, so that its meaning and intended audience (i.e., bicyclists) is clear. A wait/countdown device can be smaller if it is mounted next to a bicycle signal that is on the near side of an intersection, as bicycle signals are in Europe. In North America, where bicycle signals are far-side, a larger device is needed unless the crossing has a supplemental near-side bicycle signal. Bicycle wait countdown displays are not available in the U.S. market. The one located in Portland was imported from Europe, and its wiring was adapted for American AC (alternating current) voltage and frequency. 9.5.4.2 Phasing and Timing Not applicable for this treatment. 9.5.4.3 Signage and Striping Not applicable for this treatment. 9.5.4.4 Geometric Elements Not applicable for this treatment. Bibliography Colville-Andersen, M. (2014, August 5). The Green Waves of Copenhagen. Copenhagenize. http://www. copenhagenize.com/2014/08/the-green-waves-of-copenhagen.html Fong, G., Kopf, J., Clark, P., Collins, R., Cunard, R., Kobetsky, K., Lalani, N., Ranck, F., Seyfried, R., Slack, K., Sparks, J., Umbs, R., & Van Winkle, S. (2003). Signalized Intersection Safety in Europe (No. FHWA-PL-03-020). FHWA. 9.6 Easing Bicycle Right Turn on Red Restrictions 9.6.1 Basic Description 9.6.1.1 Alternative Names Bicycle turn on red.

156 Trafc Signal Control Strategies for Pedestrians and Bicyclists 9.6.1.2 Description and Objective is treatment encompasses two provisions that can be applied independently. One is exempting bicycles from no turn on red (NTOR) restrictions; the other is requiring bicyclists to only yield, not stop, when turning right on red. e objectives are to reduce bicycle delay and to promote equity by legalizing safe, common behaviors. 9.6.1.3 Variations For spot applications, exemptions from an NTOR restriction can be implemented by posting bike exception signs, as in Exhibit 9-10. For systematic application, state law can be changed to exempt bicycles from NTOR restrictions. In New York City and on the island of Montreal, Québec—the only places in the United States and Canada, respectively, where RTOR is prohibited by default—spot application of an exemption could be applied using signs such as those used in France to allow RTOR and to allow cyclists at the top of a “T” intersection to go through on red, with an obligation to yield to pedestrians (see Exhibit 9-11). Removing a bicyclist’s obligation to stop when turning right on red (while retaining the obli- gation to yield to pedestrians and to vehicles that have lawfully entered the intersection) can be done through legislation that applies only where bikes are allowed to turn right on red. Such a law could also be applied universally, encompassing a universal NTOR exemption. 9.6.1.4 Operating Context Exempting bikes from NTOR restrictions can be considered at any intersection with an NTOR restriction. It can be considered for specic locations by using signs or universally by means of legislation. Allowing bikes to turn right on red without stopping, but with an obligation to yield, can probably only be implemented statewide, through legislation. It could be limited to intersections Source: Peter Furth. Exhibit 9-10. Bicycle exemption from NTOR restriction, Cambridge, MA.

Treatments Addressing Special Bicycle Needs 157   where bicycles are already allowed—by existing laws—to turn right on red, or it could be applied universally, in which case it would encompass a universal exemption to NTOR restrictions. 9.6.2 Applications and Expected Outcomes 9.6.2.1 National and International Use In the U.S., bicyclists routinely violate RTOR restrictions. Yielding, but not necessarily stop- ping, when turning right on red is also common practice. Since 1982, Idaho law has allowed bicyclists to turn right on red with an obligation to yield but not to stop. Starting in 1982, the obligation to stop was lifted at both Stop signs and at red traffic signals. In 2005, the law was amended to reinstate the obligation to stop at red traffic signals, except for bikes turning right on red. (Recently adopted “Idaho stop” laws in Delaware, Arkansas, Oregon, and Washington State apply only at Stop signs, without any provision regarding RTOR.) Posting bicycle exemptions to NTOR signage is rare. Cambridge has posted exemptions to NTOR restrictions at a few intersections where the restriction appeared to be hindering cyclists. For example, the intersection in Cambridge shown previously in Exhibit 9-10, at Broadway approaching Galileo Galilei Way, has concurrent-protected crossings (see Section 6.2), with right turns for vehicles allowed only during a short phase that overlaps with a left-turn phase. With the NTOR exemption, bicycles—which approach in a separated bike lane along the right curb—are allowed to turn right during other phases as well, particularly during the through phase. Due to a railroad crossing, the bicycle and vehicle stop lines are set back about 60 ft from the curb of the cross street, which led to more bicyclists than usual stopping for a red light rather than informally turning right on red, as they do at most other intersections (P. Baxter and C. Seiderman, personal communication, September 2018). In the Netherlands, where RTOR for vehicles is almost unknown and enforcement of bicycle laws is stricter, bicycles may legally turn right at most signalized intersections. The most common mechanism—which applies where both intersection streets have cycle tracks (i.e., separated bike Source: Mieux se Déplacer à Bicyclette. Exhibit 9-11. Signs in France permitting bikes to turn right on red (left) and go through on red but only at the top of a T-intersection (right).

158 Traffic Signal Control Strategies for Pedestrians and Bicyclists lanes)—is that the stop line for bicycles is at the curb of the cross street, so that bicyclists turning right from one cycle track to another never pass the stop line and therefore are not regulated by the traffic signal. This is known colloquially as “right turn past red” (Wagenbuur, 2012). At inter- sections lacking this layout, a sign permitting bicyclists to turn right on red is sometimes posted. 9.6.2.2 Benefits and Impacts Exempting bicycles from RTOR restrictions and allowing them to turn on red without stop- ping (provided that they yield to pedestrians and other road users who have legally entered the intersection) will reduce bicycle delay. Additionally, easing these restrictions will legalize the behavior of many bicyclists who commonly ignore these restrictions. It is unlikely that there will be any negative safety impact from easing RTOR restrictions on bicyclists because these restrictions are often violated already. Furthermore, the safety rea- sons behind most NTOR restrictions from motor vehicle considerations either do not apply to bicycles or apply too weakly to merit a restriction. These reasons include turning onto high- speed roads (which usually have a shoulder or bike lane into which a bicycle can turn safely) and preventing turning vehicles from blocking a crosswalk while waiting to turn on red (this is not a problem for turning bicycles as they usually wait beyond the crosswalk, and their ability to use bike lanes and shoulders means they rarely have to wait). RTOR restrictions prevent vehicle collisions with crossing pedestrians, but pedestrian collisions are far less likely with bicycles compared to motor vehicles because of their smaller size and mass and their better lateral vis- ibility (i.e., no A-pillar obstructing their view). Easing RTOR restrictions on bicycles is similar to easing restrictions at Stop signs. A study of the safety impact of Idaho’s stop law found that the law has been beneficial or had no negative effect (Meggs, 2010). 9.6.3 Considerations 9.6.3.1 Accessibility Considerations Not applicable for this treatment. 9.6.3.2 Guidance Not applicable for this treatment. 9.6.3.3 Relationships to Relevant Treatments Not applicable for this treatment. 9.6.3.4 Other Considerations Not applicable for this treatment. 9.6.4 Implementation Support 9.6.4.1 Equipment Needs and Features Not applicable for this treatment. 9.6.4.2 Phasing and Timing Not applicable for this treatment. 9.6.4.3 Signing and Striping Not applicable for this treatment.

Treatments Addressing Special Bicycle Needs 159   9.6.4.4 Geometric Elements Not applicable for this treatment. Bibliography Caldwell, J., O’Neil, R., Schwieterman, J. P., & Yanocha, D. (2016). Policies for Pedaling: Managing the Tradeoff between Speed & Safety for Biking in Chicago. Chaddick Institute for Metropolitan Development, DePaul University. Meggs, J. N. (2010). Bicycle Safety and Choice: Compounded Public Cobenefits of the Idaho Law Relaxing Stop Requirements for Cycling. https://bikeportland.org/wp-content/uploads/2019/06/idaho-law-jasonmeggs- 2010version-2.pdf Wagenbuur, M. (2012). Cycling Past Red Lights; It’s Often Legal in the Netherlands [Blog post]. bicycledutch. wordpress.com

Next: Chapter 10 - Techniques for Multistage Crossings »
Traffic Signal Control Strategies for Pedestrians and Bicyclists Get This Book
×
 Traffic Signal Control Strategies for Pedestrians and Bicyclists
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

In the United States, traffic signal timing is traditionally developed to minimize motor vehicle delay at signalized intersections, with minimal attention paid to the needs of pedestrians and bicyclists. The unintended consequence is often diminished safety and mobility for pedestrians and bicyclists.

The TRB National Cooperative Highway Research Program's NCHRP Research Report 969: Traffic Signal Control Strategies for Pedestrians and Bicyclists is a guidebook that provides tools, performance measures, and policy information to help agencies design and operate signalized intersections in a way that improves safety and service for pedestrians and bicyclists while still meeting the needs of motorized road users.

Supplemental to the report are presentations of preliminary findings, strategies, and summary overview.

READ FREE ONLINE

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!