Planning for Traffic Signals by Jeffrey W. Buckholz, PhD, PE, PTOE A SunCam online continuing education course PLANNING FOR TRAFFIC SIGNALS

Transcription

1 PLANNING FOR TRAFFIC SIGNALS This Traffic Signal Design course addresses basic procedures used in planning traffic signal installations. Included in this course is a discussion of warrants for traffic signal installation, traffic study guidelines, signal phasing selection, and flashing analysis. TRAFFIC SIGNAL WARRANTS A traffic signal should not be installed unless one or more of the 8 traffic signal warrants described in the Manual on Uniform Traffic Control devices (MUTCD) are satisfied. However, the satisfaction of a warrant is not in itself justification for installing a signal. A signal should only be installed after an engineering study has been completed and a finding made that a signal is the most appropriate form of intersection control. This study must include a comprehensive warrant analysis and should be "signed and sealed" by a registered traffic engineer (in those states that have such a designation) or a registered Professional Engineer (P.E.) with expertise in traffic signal operations. Most warranted traffic signals are justified under warrant 1, the Eight-Hour Vehicular Volume warrant. This warrant requires that a certain volume of traffic exist on both the major street and the intersecting minor street for 8 hours of an "average" day. The same 8 hours must be used when comparing volumes from the major and minor street. The MUTCD defines an average day as "a day representing traffic volumes normally and repeatedly found at a location. Where volumes are primarily influenced by employment, the average day is typically a weekday. When volumes are primarily influenced by entertainment or recreation, the average day is typically a weekend day". If this is a location that, for a significant portion of the year, is influenced by school traffic or touristrelated traffic then it is prudent to conduct the warrant study during periods when school is in session or the tourists are in town. The volume of traffic required to satisfy warrant 1 varies depending on the number of travel lanes on the major and minor streets. The more approach lanes there are, the more traffic is required to satisfy this warrant. When two intersecting streets are roughly equivalent in size and traffic volume it can be difficult to decide which is the major street and which is the minor street. When this is the case, engineering judgment (based on such factors as the functional designation of the road and route continuity) must be exercised. The volume of traffic on the major street required to satisfy warrant 1 is obtained by adding together the approaching volumes on the major street. The volume of traffic on the minor street required to satisfy warrant 1 is obtained by taking the highest minor street approaching volume. It should be Copyright 2010 Jeffrey W. Buckholz Page 1 of 24

2 kept in mind that the minor street approach having the highest volume may change depending on the time of day. In order to satisfy warrant 1 condition A (often referred to as warrant 1A), the Minimum Vehicular Volume warrant, traffic volumes on both the major and minor streets must be relatively high. For example, at the intersection of a 2 lane major street and a 1 lane minor street, warrant 1A requires that the major street have at least 600 vehicles per hour and the minor street have at least 150 vehicles per hour, and that these volumes occur for at least 8 hours. WARRANT 1A 1 EXAMPLE B C D A FOR 8 HOURS: A+B 600, C D 150 AND Warrant 1 condition B (often referred to as warrant 1B), the Interruption of Continuous Traffic warrant, is satisfied when traffic volumes on the major street are so heavy that traffic on the minor street suffers excessive delay or hazard in crossing the major street. In comparison to warrant 1A, the required volumes for warrant 1B are higher on the major street, but lower on the minor street. For example, at the intersection of a 2 lane major street and a 1 lane minor street, warrant 1B requires that the major street have at least 900 vehicles per hour and the minor street have at least 75 vehicles per hour, and that these volumes occur for at least 8 hours. Copyright 2010 Jeffrey W. Buckholz Page 2 of 24

3 WARRANT 1B 2 EXAMPLE B C D A FOR 8 HOURS: A+B 900, C D 75 AND The 900 value is 50% greater than the 600 value required by warrant 1A whereas the 75 value is half that required by warrant 1A. In examining the volumes required to meet warrants 1A and 1B an interesting phenomenon is noted. An intersection which satisfies one of these warrants may no longer satisfy the warrant if the number of lanes is increased. Consequently, a viable alternative to installing a signal may be the addition of travel lanes. The MUTCD echo's this sentiment by stating that "It is desirable to have at least two lanes for moving traffic on each approach to a signalized location". When neither warrant 1A nor warrant 1B is satisfied, yet the numbers are such that both of these warrants are almost satisfied, a combination warrant (which can be referred to as warrant 1A-B) may apply. This warrant is satisfied when 80% of both warrant 1A and warrant 1B is met. However, the MUTCD cautions that adequate trial of other intersection control measures (such as a four way stop) should be carried-out before installing a signal under this warrant. The values required to meet warrant 1 are reduced to 70% of their normal value if one of two conditions exist. The first condition is that the 85th percentile speed on the major street exceeds 40 mph. The MUTCD recognizes that, other things being equal, intersecting traffic on high-speed facilities requires a higher level of control. The second condition for applying the 70% reduction is that the intersection lies within an isolated community of less than 10,000 people. The reason for Copyright 2010 Jeffrey W. Buckholz Page 3 of 24

4 this condition is not clear. Apparently, smaller communities like the prestige associated with having a traffic signal or two, even if they don't have the traffic volumes to warrant them. Warrant 2, the Four Hour Vehicular Volume warrant, is satisfied when, for any four hours of an average day, plotted points on the graph fall above the appropriate line. WARRANT 9 WARRANT 2 CHART FOUR HOUR VOLUME WARRANT In this example both the major street and the minor street have two approach lanes. The horizontal axis of the graph represents the combined major street approach volume while the vertical axis represents the highest volume minor street approach. The 70% reduction also applies to warrant 2 and a similar graph is used when either of the two previously discussed 70% reduction conditions is met. Copyright 2010 Jeffrey W. Buckholz Page 4 of 24

5 WARRANT 2 CHART 9 WITH WITH 70% REDUCTION FOUR HOUR VOLUME WARRANT Warrant 3, the Peak Hour warrant, is satisfied when, for any hour of an average day, the plotted point on the graph falls above the appropriate line or the following three criteria are met: 1. The total delay on the busiest approach of the minor street exceeds 4 hours for a one lane approach or 5 hours for a two lane approach. 2. The volume on the same approach exceeds 100 vph for a one lane approach or 150 vph for a two lane approach. 3. The total entering volume exceeds 800 vph for intersections with 4 or more approaches, or 650 for intersections with three approaches. Copyright 2010 Jeffrey W. Buckholz Page 5 of 24

6 WARRANT 3 11 PEAK HOUR VOLUME WARRANT Warrant 3 is intended only for use at locations that experience high minor street traffic volumes during brief periods of the day, such as factory or school exits. As with warrants 1 and 2, the 70% reduction also applies to warrant 3 and a similar graph is used when either of the two 70% reduction conditions is met. Warrant 4 is the Pedestrian Volume warrant. In order to satisfy this warrant, the MUTCD requires that each of the following three conditions be met: 1. The number of pedestrians crossing the major street must be 100 or more for four hours (or 190 or more for any one hour) 2. The number of gaps adequate for these pedestrians to cross must be less than 60 per hour. (Or, in general, the number of gaps must be less than the number of minutes in the observation period.) 3. The nearest existing traffic signal on the major street is greater than 300 feet away, unless the proposed signal will not restrict the progressive movement of traffic. Copyright 2010 Jeffrey W. Buckholz Page 6 of 24

7 WARRANT 4 EXAMPLE FROM 3:30 TO 4:00 PM 26 SUITABLE GAPS 2 6 G A P S < 3 0 M IN W A R R A N T IS M E T When the predominant pedestrian crossing speed is below 4 feet per second, condition 1 can be reduced by 50%. A speed of less than 4 feet per second is usually indicative of small children, physically-challenged individuals, or frail elderly. Where the major street is a divided street having a median of sufficient width for pedestrians to wait, condition 2 applies separately to each direction of travel. If a signal is installed under warrant 4 then it should be traffic actuated, should have pedestrian features (pedestrian heads, push buttons, etc.) and, if it is a mid-block signal, parking should be prohibited for 100 feet in advance of the signal and 20 feet beyond the signal. In addition, if the signal is located within a signal system then it should be coordinated with the other signals in the system: IF WARRANT 4 IS USED SETBACKS 20' 100' S S 100' 20' PED FEATURES COORDINATED Copyright 2010 Jeffrey W. Buckholz Page 7 of 24

8 Warrant 5 is the School Crossing warrant. In order to satisfy this warrant, the MUTCD requires that the following condition be met: During the time children are crossing, the number of gaps in the traffic stream adequate for children to safely cross the street must be less than the number of minutes in the same period, and there must be a minimum of 20 students during the highest crossing hour. In addition, the nearest existing traffic signal on the major street must be greater than 300 feet away unless the proposed signal will not restrict the progressive movement of traffic. For example, if during the 3:00 to 4:00 pm school crossing period there were 25 students crossing but only 36 suitable gaps in the traffic stream were available, then a signal would be warranted since 25 is greater than 20 and since 36 is less than 60 (the minutes of the crossing period). As with the Minimum Pedestrian Volume warrant, when a signal is installed under the School Crossing warrant then it should be traffic actuated, should have pedestrian features (pedestrian heads, push buttons, etc.) and, if it is a mid-block signal, parking should be prohibited for 100 feet in advance of the signal and 20 feet beyond the signal. When a signal used for crossing school children is installed at an intersection, it is prudent to prohibit conflicting right turns on red during the school crossing periods. This can be done either through traditional signs that indicate the times during which right turn on red is prohibited or, better yet, via fiber optic or LED blank-out signs that are activated only during the restricted periods. Copyright 2010 Jeffrey W. Buckholz Page 8 of 24

9 Even with signalized control, crossing guards still provide a valuable service, especially when young children are involved. Inattentive motorists, motorists trying to turn right on red, and confused or daredevil children are all situations that can be appropriately handled by well-trained crossing guards. Warrant 6, the Coordinated Signal System warrant, is used to justify the installation of signals at locations that will foster the proper grouping of vehicles within a coordinated signal system. The idea is that, without such signals, platoons might spread out too much to maintain adequate progression. The MUTCD indicates that a signal should not be installed under this warrant if the resultant signal spacing would be less than 1000 feet. This is the only warrant with no numerical guidelines. Warrant 7 is the Crash Experience warrant. In order to satisfy this warrant, the MUTCD requires that each of the following conditions must be met: 1. Within a 12 month period, 5 or more accidents of the type correctable by signal control (rightangle accidents) must have been reported. Each of these accidents must involve either personal injury or property damage % of warrant 1A, 1B or 4 is met 3. Less restrictive remedies (such as increased enforcement) have been tried to correct the accident problem with no success. Copyright 2010 Jeffrey W. Buckholz Page 9 of 24

10 If the accident pattern that is to be corrected by installation of a signal involves left turning vehicles colliding with thru vehicles, then left turn phasing should be provided as part of the new signal installation. If a signal is installed based on the accident warrant then, after the signal has been operational for awhile, a follow-up accident study should be conducted to make sure that the accident problem has been corrected. Warrant 8 is the Roadway Network warrant. It is applicable where two major routes intersect. In order to satisfy warrant 8 each of the following two conditions must be met: 1. The intersection has a total entering volume (either existing, or projected for the near future) of at least 1000 vehicles during the peak hour of a typical weekday. 2. The intersection has 5 year projected traffic volumes which meet warrants 1, 2, or 3. WARRANT 8 WARRANT 7 B C D A FOR WEEKDAY PEAK HOUR For Weekday Peak Hour: A+B+C+D A + B + C + D > > 1000, and AND Warrant 1, 2 or 3 WARRANT 1, 2, 8, 9 OR 11 Met within 5 Years MET WITHIN 5 YEARS Copyright 2010 Jeffrey W. Buckholz Page 10 of 24

11 or, alternatively, the following condition must be met: The intersection has a total entering volume (either existing, or projected for the near future) of at least 1000 vehicles for each of any 5 hours on a Saturday or Sunday....OR FOR 5 HOURS ON A SATURDAY OR SUNDAY A+B+C+D > 1000 The installation of an unwarranted traffic signal can cause many problems, including: 1. Excessive motorist delay 2. Disobedience of the signal 3. An increase in rear-end accidents 4. The use of less adequate routes to circumvent the signal Probably the worst effect of installing an unwarranted signal is that it fosters motorist disrespect for the entire traffic control system, and those who design it. This makes the motorist less likely to obey other traffic control devices and, over time, results in a decrease in the safety level of the entire roadway system. From a practical standpoint, it should be understood that the installation of a traffic signal can be a highly political matter. Unfortunately, many citizens mistakenly view traffic signals as the solution to a variety of transportation problems, including speeding and accidents. More than one unwarranted traffic signal has been installed by a traffic engineer afraid to lose his or her job. Copyright 2010 Jeffrey W. Buckholz Page 11 of 24

12 However, it should be kept in mind that the installation of unwarranted signals is a dangerous practice that also exposes the agency responsible for the installation to a significant legal liability risk. In practice, the removal of an unwarranted traffic signal is more difficult to accomplish than keeping the signal from being installed in the first place. The detailed signal removal procedures required by the Traffic Control Devices Handbook support this contention. Suffice it to say that there is a ratchet-effect with respect to signalization and, once a signal is in, it can be very difficult to remove. Before a signal is removed, a "signed and sealed" signal removal study should be prepared and submitted by a registered professional engineer. This study should discuss the reasons for removing the signal and should include a warrant analysis. Peculiar site conditions, such as a sight distance problem, may justify keeping a signal even though it is not warranted. PRE-DESIGN TRAFFIC ENGINEERING STUDY Prior to designing and installing a traffic signal, it is important to conduct a thorough traffic engineering study of the intersection in question. This study will be important in selecting design features and developing timings for the new signal. As was just discussed, a signal warrant study should also be conducted - either independently or as part of this traffic study. If it turns out that a signal is not warranted, then the traffic study should identify the best form of intersection control. A wide variety of data is typically collected as part of a pre-design traffic engineering study. This data includes: Manual Turning Movement Counts (TMC s) Automatic Traffic Recorder Counts (ATR s), also referred to as "machine counts" Accident Reports Approach Speeds Minor Street Approach Delay Major Street Gap Distribution Physical Layout of the Intersection Sight Distance from Minor Street Approaches Copyright 2010 Jeffrey W. Buckholz Page 12 of 24

13 MANUAL TMC s Manual turning movement counts (TMC's) are usually conducted during the 8 busiest hours of a typical weekday. TMC's are not typically conducted on Fridays (and sometimes Mondays) because of the atypical traffic characteristics of this day. Although 8 hours is a typical duration for a TMC, 10 and 12 hour TMC's are also relatively common. TMC's totals are usually recorded every 15 minutes. This period has been chosen since it is short enough to accurately depict time-of-day variations in traffic flow, but not so short that unmanageable mounds of data are obtained. Once the TMC is complete the data is summarized and both the AM and PM peak hour traffic volumes are determined. An example PM peak hour summary is shown in this graphic: TMC SUMMARY :00-5:00 PM Peak hour volumes are very important since they are used to guide the operational design of the new traffic signal. During a TMC, every vehicular movement (thru's, left turns, and right turns) is recorded by a human observer. In the past, simple pencil & paper tally sheets were used to keep track of the turning movements. However, most TMC's are now carried-out using computerized counting boards: Copyright 2010 Jeffrey W. Buckholz Page 13 of 24

14 Once they are mastered, the computerized boards are easier to use and allow much quicker tabulation and summary of the data than the old manual method. During TMC's it is common to record commercial vehicle movements separately from passenger car movements. Although the definition may vary from area to area, to keep things simple a commercial vehicle is usually considered to be any vehicle having more than four wheels. Pedestrian and bicycle movements may also be recorded as part of the TMC. In addition, at an already signalized intersection, right-turns-on-red may be recorded separately from right-turns-ongreen if this is considered important information to have. Valuable data that is obtainable from the TMC's includes the percentage of trucks by movement and the number of pedestrians using each crosswalk. Truck percentages are used in certain intersection capacity and signal timing calculations while the pedestrian volumes are important in locating crosswalks and timing pedestrian signals. Copyright 2010 Jeffrey W. Buckholz Page 14 of 24

15 ATR COUNTS Automatic Traffic Recorder (ATR) counts are typically conducted on each approach to the intersection for a period ranging from 24 hours to 7 days. As with the manual TMC's, these machine counts are usually collected using 15 minute recording intervals. However, unlike the TMC's, machine counts do not require someone to stay at the site; the counter is installed on one day and the results retrieved on another. Since the ATR counts are continuous, they permit the design engineer to identify the variation in traffic volume that occurs over the course of the day, and from one day to the next. These time-ofday and day-of-week variations in traffic flow are typically graphed, and then used to determine the periods during which certain timing parameters are to be implemented. V A R I A T I O N I N T R A F F I C For example, the signal timings for the weekday PM peak period might be activated between 3:30 PM and 6:15 PM based on ATR information. The ATR counts also permit the engineer to identify low volume periods during which flashing of the traffic signal may be in order. Some agencies prefer to conduct the ATR counts before the TMC's. Doing this allows the 8 busiest hours to be identified before the TMC is conducted. Other agencies prefer to conduct the ATR and TMC counts simultaneously so that one set of counts can be checked against the other. Copyright 2010 Jeffrey W. Buckholz Page 15 of 24

16 Either TMC's or ATR's can be used to check the volume-based warrants (1, 2, 3, 4, and 7). In practice, TMC's are usually preferred over ATR's because of their greater accuracy. ACCIDENT REPORTS Reviewing reports of accidents that have occurred at an intersection is a good way to identify problem areas that might be correctable with signalization. For example, an intersection with many sideswipe accidents might benefit from the addition of signing and pavement markings that more clearly designate lane usage. Accident information is also required for evaluating signal warrant 7. Unless a well-designed and properly maintained computerized accident record system exists, securing accident records for a given location can be a tedious, time-consuming process. Since the police typically file accident records by date instead of location, it is usually necessary to search through file cabinets full of accident records to obtain those pertaining to the intersection of interest. In collecting and compiling accident records it is standard practice to go back 3-years. Accidents occurring more than 3 years ago are usually not considered relevant. It is important to identify any intersection improvements or modifications that might have occurred within the past 3 years, changes that might have affected the accident pattern. The accident data is summarized in the form of a collision diagram. This diagram provides a graphic representation of past accidents and is a good visual tool for identifying accident patterns: Computerized systems for recording and storing accident records are the wave of the future. Some of these systems even allow automated preparation of the collision diagrams. DELAY STUDIES Copyright 2010 Jeffrey W. Buckholz Page 16 of 24

17 Vehicular delay information on a STOP controlled side street is usually collected using a special computerized board. A human observer pushes one button every time a vehicle arrives at the side street and another button every time a vehicle leaves the side street. The computer keeps track of how many vehicles are queued on the side street and the associated vehicular delay. This delay information is primarily used to evaluate warrant 3. GAP STUDIES Collecting data on gaps in the major street traffic stream is important for evaluating warrant 4 (the pedestrian volume warrant) and warrant 5 (the school crossing warrant). Gap data is also collected using a special computerized board. SPEED STUDIES Speed data is required for a number of purposes: 1. To determine whether or not the 70% reduction should be applied during the warrant analysis. 2. To evaluate whether there is sufficient sight distance in which to see the signal indications. 3. For establishing the proper detector configuration, especially on the major street. 4. For calculating certain signal timing intervals, such as the yellow interval and the all-red interval. When conducting a speed study it is important to collect speed data only on free-flowing vehicles. For example, vehicles that are trapped behind a slow truck should not have their speed recorded (although the truck itself should). Although it is not necessary to record the speed of all free-flow vehicles, enough speed readings should be recorded to ensure that a representative sample has been taken. There are three commonly used methods for collecting speed data: 1. The Radar Gun Method 2. The ATR Method, or 3. The Manual Stopwatch Method Copyright 2010 Jeffrey W. Buckholz Page 17 of 24

18 Although radar guns are fairly accurate, they are not a good means for collecting speed data. Drivers are accustomed to seeing radar guns used for speed enforcement and routinely slow down whenever they encounter a radar gun. The use of radar detectors contributes to this phenomenon. Using two road tubes it is possible to configure most modern ATR's to collect speed data and to store this data into pre-defined speed "bins". (For example: 0-30 mph, mph, mph, and greater than 50 mph.) The main advantage in using ATR's is that data can be collected during all hours of the day and night. ATR's are also attractive in that a lot of speed data can be collected for relatively little cost. The main disadvantage of ATR's is that the speeds of following (non freeflowing) vehicles are recorded. Another disadvantage is that some cautious drivers slow down when they encounter a set of tubes stretched across the road. Vehicle speeds can be obtained by drawing two lines on the road a fixed distance apart and then recording with a stopwatch the time it takes for a vehicle to pass from one line to the next. This is an easy and inexpensive method that yields reasonable data and does not "spook" drivers into slowing down. CONDITION DIAGRAMS A condition diagram is a field drawing that contains items of importance in the planning and design of a traffic signal installation. These items include: Existing sidewalk locations and widths Existing crosswalk locations and lengths The lane configuration for each approach Lane widths Posted speed limits Turn prohibitions or restrictions On-street parking areas Approach grades Nearby railroad crossings, draw bridges, or fire stations Designated school crossings Sight distance restrictions Driveway Aprons Utility features (manholes, storm drains, etc.) Copyright 2010 Jeffrey W. Buckholz Page 18 of 24

19 INTERSECTION LAYOUT As a general rule-of-thumb, a condition diagram should show all relevant items located within 300 feet of the intersection. A "distance wheel" is frequently used to obtain distance measurements at an intersection. These wheels are available in both English and metric units. If a required measurement is long, then the use of a vehicle-mounted DMI (Distance Measuring Instrument) will prove more convenient. Most DMI's also allow measurements to be made in either English or metric units. A comprehensive set of color photographs, or a video tape, often come in very handy during traffic signal planning and design. These items should also be included as part of the data collection effort. PHASING SELECTION One of the most important decisions in planning a traffic signal installation is the signal phasing that will be used. The use of phasing that is too complex will produce unnecessary delay whereas the selection of phasing that is too simple might be unsafe, or might unduly penalize a given movement. Copyright 2010 Jeffrey W. Buckholz Page 19 of 24

20 As a general rule, an attempt should be made to hold the number of phases to a minimum, without compromising safety or creating awkward traffic operations. PHASING SELECTION WHICH? ONE The key to good signal phasing revolves around the efficient handling of left turn movements. The primary decision is whether or not an exclusive left turn phase will be required. This decision must be made for both the major street approaches and the minor street approaches. Both protected-only phasing, which requires all left turns to made on a green arrow, and protected/permissive phasing, which allows the motorist to turn left on either the green arrow or the green ball, have a separate left turn phase. These phasing types are appropriate for locations with a significant number of left turns or where the potential for left turn accidents is high - such as locations with sight distance problems. Copyright 2010 Jeffrey W. Buckholz Page 20 of 24

21 LEFT TURN PHASING WHICH ONE? When left turns are prohibited, when permissive phasing is used, or when opposing approaches are split-phased, no separate left turn phase is provided. With permissive phasing the motorist must turn left on the green ball or during the yellow ball change interval. Under split-phasing, opposing approaches are timed separately. All traffic from one direction moves first, and then traffic from the opposite direction moves. When split phasing is in use the left turns from each approach proceed unopposed. In some cities, such as Detroit, left turns are prohibited at major intersections and motorists must turn right and then make a U-turn in order to go left. "DETROIT LEFT" Copyright 2010 Jeffrey W. Buckholz Page 21 of 24

22 The Traffic Control Devices Handbook (TCDH) contains a set of suggested warrants for the installation of a left turn phase. The TCDH recommends installation of separate left turn phasing if one or more of the following traffic volume, delay, or safety warrants is met. The left turn volume warrant pertains to peak hour traffic activity. It states that a left turn phase is warranted on a particular intersection approach if both of the following conditions are met for that approach: 1. Multiplying the number of left turning vehicles by the number of opposing thru vehicles produces a value greater than 50,000 for a 2-lane street, or a value greater than 100,000 for a 4-lane street. 2. The left turn volume is greater than 2 vehicles per cycle. The left turn delay warrant also pertains to peak hour traffic activity. It states that a left turn phase is warranted on a particular intersection approach if all three of the following conditions are met for that approach: 1. There is a total left turn vehicle delay of 2 hours or more. 2. The left turn volume is greater than 2 vehicles per cycle. 3. The average delay per left turning vehicle is at least 35 seconds. The left turn accident warrant states that a left turn phase is warranted if either of the following conditions is met: 1. For one approach, the number of left turn accidents is 4 or more in 1 year or 6 or more in 2 years. 2. For both approaches, the number of left turn accidents is 6 or more in 1 year or 10 or more in 2 years. Note that the delay and accident warrants can only be used to decide if a left turn phase should be installed, they cannot be used to decide if it should be removed. For making phase removal decisions the volume warrant must be used. It should be emphasized that agencies vary widely in their view of left turn phasing, with some agencies almost routinely installing left turn phases while others shun them except where absolutely necessary. Knowledge of local criteria and practice is important in selecting left turn phase treatments that are appropriate for the local environment. Copyright 2010 Jeffrey W. Buckholz Page 22 of 24

23 Additional right turn capacity can be achieved by using a right turn overlap arrow in unison with a left turn arrow. RIGHT TURN PHASING Most agencies, but not all, require that a separate right turn lane exist before a right turn arrow can be installed. As strange as it might first seem, the decision on whether or not to install a right turn overlap phase is dependent upon the need to permit U-turns on the approach having the simultaneous left turn. A conflict can occur between right turns and u-turns if a right turn arrow and a left turn arrow are displayed at the same time. In this case it is unclear as to which driver has the right-of-way, the right turn vehicle or the vehicle making the U-turn and such ambiguities can lead to accidents. The U-turns must either be prohibited or a sign must be erected which forces the U-turns to yield to the right turns. If the street with the left turn phase has a divided median and U-turns are necessary to reach driveways located between the intersection and the next median break, prohibiting U-turns to permit the installation of a right turn phase is not an attractive option. FLASHING TRAFFIC SIGNALS Copyright 2010 Jeffrey W. Buckholz Page 23 of 24

24 Traffic signals are flashed under one of the following four conditions: 1. During low volume periods, such as late at night or on weekends (to reduce vehicular delay). 2. When the signal malfunctions (so that incorrect signal displays, such as conflicting green indications, are not dangerously provided to motorists). 3. Before a new signal is turned-on or before an unwarranted signal is removed (to warn motorists that a traffic control change is about to occur). 4. During special events (such as a parade). Flashing a traffic signal during low volume periods can significantly reduce motorist delay while also cutting electrical consumption. In order to safely place a traffic signal into flashing operation, certain criteria must be met: 1. Traffic volumes during the flash period should have fallen to sufficiently low levels on both the major and minor street so that side street queuing and dangerous conflicts do not occur. 2. Adequate sight distance must exist so that minor street motorists can safely enter or cross the major street. 3. The intersection should not be so complex that flashing operation will confuse the right-ofway situation, such as might happen at an intersection with signal controlled frontage roads. It is also a reasonable policy not to flash the intersection of two major routes, except in the case of signal malfunction or during a special event. Flashing a major intersection may violate driver expectations regarding traffic control which could lead to an increase in the number of accidents at the intersection. If many signals in a row are flashed then a "speedway effect" could occur. With a speedway effect, motorists on the major street travel at a high rate of speed because they know that they will not have to stop for any of the signals. Consequently, it is a reasonable policy to leave one of the signals in normal operation. The major decision that must be made with respect to flashing signals during a malfunction is whether a red indication should be shown to all approaches (turning the intersection into a multiway stop), or whether the red indication should only be shown to the minor street approaches while a yellow indication is shown to the major street (resulting in a 2-way stop). Where two major routes intersect, or where sight distance restrictions are present, it is recommended that an all-red flash be used. Otherwise, the more standard yellow/red flash is recommended for efficiency reasons. Copyright 2010 Jeffrey W. Buckholz Page 24 of 24