Home

User Guide Only - passer

1. 85 Grade of Approach Percentile Uphi Level Downhill a 4 3 2 1 1 2 3 4 25 2 63 2 68 2 73 2 78 2 84 2 90 2 96 3 03 3 11 35 3 28 3 35 3 42 3 49 3 57 3 56 3 75 3 85 3 95 45 3 93 4 02 4 11 4 20 4 31 4 42 4 54 4 66 4 80 55 4 58 4 69 4 80 4 92 5 04 5 18 5 32 5 47 5 64 65 5 23 5 35 5 49 5 63 5 78 5 94 6 11 6 29 6 48 Red Clearance Interval Lengths Using Equation 2 in seconds 85 Width of Approach Percentile Feet Speed 90 30 40 50 60 70 80 90 100 110 120 mph 25 1 09 1 36 1 63 1 90 2 18 2 45 2 72 2 99 3 27 3 54 3 81 35 0 78 0 97 1 17 1 36 1 55 1 75 1 94 2 14 2 33 2 53 2 72 45 0 60 0 76 0 91 1 06 1 21 1 35 1 51 1 66 1 81 1 97 2 12 55 0 49 0 62 0 74 0 87 0 99 1 11 1 24 1 36 1 48 1 61 1 73 65 0 42 0 52 0 63 0 73 0 84 0 94 1 05 1 15 1 26 1 36 1 47 14 Key Point Combining the equation for yellow change interval and the appropriate equation for red clearance time see discussion ensures that motorists are not trapped in a dilemma zone The dilemma zone is defined as a point where a driver cannot stop at a reasonable rate of deceleration and where the same driver cannot pass through the intersection within the yellow time allotted The use of the red clearance interval is primarily a tool to avoid
2. First we will examine the peak hour factor Note from our discussions in Chapter 3 that PHF is a ratio of the total hourly volume or flow rate to the hourly flow rate within the peak 15 minutes of the peak hour If this ratio is less than 0 85 then volumes within the peak hour fluctuate significantly enough that you definitely need to incorporate the PHF into your analysis By entering the PHF you essentially bump up the hourly flow rate to account for heavy volume during the peak 15 minute period within the peak hour For our example problem we do not have PHF information i e we have hourly counts and would need 15 minute counts to compute the PHF so enter 0 90 so that at least some variability is accounted for in your analysis The next item to check is the growth factor which is a convenient way for you to examine the effects of future growth on your signal and intersection Entering a factor of 1 20 means that intersection volumes have grown by 20 percent Since our volumes are current and we are developing timings for present operations we will leave this value as 1 00 The next setting that we can change for each movement is the default or ideal saturation flow rate Since 1900 pephgpl is the currently adopted value for the industry 6 you will only want to change the ideal value if some local factor sight distance problems usual curb radius etc that we have not already included on our analysis affects traf
3. 400 ft depending on volume and where interior turn volume intensity is high With proper splits and offsets it allows almost all traffic movements to progress through the interchange with the exception of the fixed interval transition portion of interior left turn phases for frontage road U turns Four phase operation is a lead lead timing plan and commonly features two fixed interval transitions also known as travel time intervals internal intervals or fixed time intervals and in outdated terminology as internal overlaps fixed overlaps or travel time overlaps These fixed interval transitions are related to the travel time between the two intersections The fixed interval transitions occur when the external arterial movement entering one side of the interchange occurs simultaneously with the frontage road movement of the other side The spacing of the interchange allows these movements to be timed together for a duration no greater than the travel time between the two intersections of the interchange Four phase operation with two fixed interval transitions is common for interchanges with an intersection spacing of less than 200 ft and for interchanges with a spacing of between 200 and 400 ft that experience heavy and unbalanced ramp traffic The phasing sequence is fixed by the strategy Fixed no Interval Transition Left Hand Right Hand Side Exterior Side Exterior Served Served Fixed Inte
4. 27 Popular Signal Timing Analysis and Optimization Programs TSIS CORSIM CORSIM 13 is a microscopic stochastic simulation program It has two modules FRESIM for evaluating freeway traffic conditions and NETSIM for evaluating the quality of a selected signal timing plan TRAFVU is an accompanying graphic animation program NETSIM can be used to analyze the operation of pretimed and actuated signals For a given scenario CORSIM randomly generates traffic keeps track of individual vehicles as long as they are in the system and computes various measures of effectiveness delay stops travel times fuel consumption etc Making a simulation run using CORSIM is similar to one time data collection in the field for instance the duration of the AM peak period on Monday Thus it is necessaty to make several runs using different random number seeds and averaging the results from those runs before drawing any conclusions CORSIM was developed using Federal Highway Administration FHWA support over a period of several decades and is accepted by transportation professionals as a valid analysis tool CORSIM does not provide an optimization routine Therefore it is difficult if not impossible to use CORSIM for developing optimal signal timing plans TRANSYT 7F TRANSYT 7F 14 is a mesoscopic deterministic model for analyzing and optimizing signal timings on arterials and networks Like CORSIM TRANSYT 7F has been developed and tested over a period
5. 53 Impedance Effects Using field data from different locations Kyte et al 31 verified that a higher priority movement has additional effects on the conflicting movement with lower priority besides being part of the conflicting flow This effect is referred to as impedance effect and is due to congestion of the higher priority movements Since Rank 1 movements have the highest priority they are not impeded by any other movements Also it is assumed that Rank 1 traffic does not incur delay Since Rank 2 movements only yield to the Rank 1 movements which do not incur any delay there are no additional impedance related adjustments for these movements Thus the movement capacity of Rank 2 movement 7 Cin i is equal to its potential capacity i e c ae Rank 3 movements must yield to both Rank 1 and 2 movements In this case not all the headways of acceptable length for Rank 3 movements will be utilized because some of these headways will be used by main street left turn movements In other words Rank 3 movements can pass through the intersection only when there is no main street left turn traffic The probability of main street left turn movement operating in a queue free state is Po 1 i Cm j where py probability that conflicting Rank 2 movement will operate in a queue free state v flow rate of Rank 2 movement in vph C movement capacity of Rank 2 movement in vph m j Thus the movement capacity
6. mt td O wi Note that the thresholds for v c ratio are the same as those for signalized intersections However thresholds for determining delay LOS are different The two data access tabs we have not discussed up to this point are Capacity Data and Headway Data As shown below the Capacity Data tab is used to group all data that are used by the program to compute capacity In addition to the lane and volume data these data include peak hour factor growth factor and percent of heavy vehicles The peak hour and growth factors are used to adjust entered volumes Heavy vehicles is used to adjust critical headway and follow up time data Lastly ideal saturation flow rate is used to derive capacity of Rank 1 movements Note that these are the through and right vehicles on the main street approaches If you wish you can supply your own capacity data and check the Lock Capacity option identified below to prevent the program from overriding your data with its own calculations Node Data ol x Export Intersections Controller Type Artery 1 at Artery 2 UnsignaiedTWSC Controller Id 3 AreaType Other v Intersection Data Capacity Data MOEs Headway Data ji gt la h gt a fi 2 ka a 6 oa as ee aso na s as 3 gt lt 1 676 147 73 18 s 75 z3 a w7 is a s4 oso os os oso oso oso oso oso os os oso 100 100 ioo 100 100 fioo
7. Green Splits 22 5 22 5 22 5 35 83 35 83 15 Green Splits sec 27 27 27 43 43 16 17 Sat Flow Data 18 Artery George Bush Dr George Bush Dr 19 Movement EBL EBT EBR WBL WBT 20 Lane Assignment 1 2 11 gt 21 Volume vph 100 100 100 100 100 22 Adjusted Flow vph 100 100 100 100 100 23 Peak Hour Factor 1 1 1 1 1 24 Growth Factor 1 1 1 1 a 25 Heavy Vehicles 2 2 2 2 2 26 Ideal Sat Flow pcphgpl 1900 1900 1900 1900 1900 27 Sat Flow pcphg 1769 61 3725 49 1583 33 907 49 907 49 28 Prot Sat Flow pcphg 1769 61 3725 49 1583 33 907 49 907 49 29 Perm Sat Flow pcphg 1225 16 764 31 30 31 Signal MOEs 32 Artery George Bush Dr George Bush Dr 33 Movement EBL EBT EBR WBL WBT 103 Per WBR 1 100 Prot rFrwana 35 83 43 WBR 1 100 100 1 1 2 1900 1583 33 1583 33 WBR H Texas Avenue SBL SBT 1 3 100 100 Prot Prot Lag Yes 7 4 6 6 3 3 1 1 4 4 20 275 24 33 Texas Avenue SBL SBT A 3 100 100 100 100 1 ud 1 1 2 2 1900 1900 1769 61 5588 24 1769 61 5588 24 1225 16 Texas Avenue SBL SBT SBR Prot SBR 15 15 SBR Chapter Analysis of Isolated TWSC Intersections Using PASSER V to analyze a tvo way stop controlled intersection intersection you are familiar with the PASSER V user interface and the locations of most of the data entry points within the program Our next exercise will analyze an isolated TWSC intersection To speed up the learni
8. Y volume divided by saturation flow for the critical approach in phase 7 n subscript for each phase The figure below reproduced from the Trafic Engineering Handbook 7 highlights an essential result that is derived from the use of Webster s equation 100 S B 80L 2 Flow z Entering 2 60 Intersection a a 3 000 vph 2 407 3 2 800 vph T T gt z t 2 400 vph g y 3 4 Co Co 1 2 Co 1 600 vph 20 3 4 Co Co 1Co z 4Co Co 1 Co 0 20 40 60 80 100 120 140 160 180 Cycle Time sec Note 2 phase 4 leg intersection w equal flow leg equal sat flows equal green times and total lost time of 10 seconds Key Point A range of cycle lengths will produce good intersection operations the flexibility provided by the range can be used to provide extra green time to left turns or critical through movements Minimum Green Time In either pretimed semi actuated or actuated mode of operation each phase at an intersection or a diamond interchange must be programmed with a minimum green time The minimum time is determined based on a number of considerations including the mode of controller and phase operation the presence and location of detectors on the approach served by the phase and the responsiveness of motorists using the facility In pretimed operation in its most simple form the minimum green time and maximum green time for each phase can be set to the same
9. gt SH 6 North Link Length 720 feet Intersection Width Storage Length 684 feet Link Speed 40 mph Travel Time WB SH 6 North gt SH 6 South Link Length 720 feet Intersection Width Storage Length 696 feet Link Speed 40 mph Travel Time Diamond using One Controller Update OK Cancel Notice that even though we changed the link distance the nodes and links did not reposition themselves in our network plan view in PASSER V This is an important point with respect to the PASSER V network editing window changes in the graphic editor will result in changing link lengths in the data set however changing details in the data set which PASSER V does retain and use for analytical purposes by typing in new values in the Link Data dialog box does not result in a repositioning of network features in the user interface The next step in coding a diamond interchange is to tell PASSER V that the link between the two intersections is the interior link of an interchange Click on the Select button and click on the interior link At the bottom of the Link Data dialog box window you will see a check box labeled Diamond using One Controller By placing a check mark in this box by clicking on the box you let PASSER V know to treat the two intersections joined by this link as a diamond interchange After you have clicked on the check box click on the button labeled Show Diamond Data A Node Data window appears see next page 1
10. 100 100 froo 100 100 1 00 1 00 ja 00 1 00 1 00 1 00 1 00 1 00 1 00 1 00 1 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 4488 40 976 03 809 44 5547 50 183 46 0 00 95 49 21065 0 00 94 00 57 58 4488 40 976 03 889 44 5547 50 183 46 0 00 95 49 210 65 0 00 94 00 57 58 Settings Platoon Dispersion Model Lock Capacity I Lock Critical Headway I HCM Lock Follow up Time I C Manar and Baass Update OK Cancel As illustrated below headway data includes critical headway and follow up time Node Data o x Export Intersections Controller Type Artery 1 at Artery 2 Unsignalized TWSC Controller Id 3 AreaType Other v Capacity Data Headway Data l4 Intersection Data moes ja 3 gt a i 2 gt aft 2 gt lt 1 e6 17 44 635 za 19 68 150 13 80 43 4 10 750 650 690 750 650 690 412 752 652 692 752 652 692 220 221 Settings 5 a Platoon Dispersion Model Lock Capacity M Lock Critical Headway I HCM Lock Follow up Time I Manar and Baass Update OK Cancel aso 400 330 a50 400 330 jas 40 aan fasi 40 331 If you have more accurate data from field observations you can enter that data and check the Lock Critical Headway and or Lock Follow up Time options to prevent the program from overriding your data As discussed earlier follow up time can be easily mea
11. 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 1816 Optimization Analysis Tools Select SubNetwo Artery List Harvey Show All Sub Arts Hide All Sub Arts Artery List Harvey PASSER III GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis Redraw Summary Report Detailed Report Cycle 45 sec Print WB Band 12 0sec WB Attain 100 0 Timing Source PASSER Ill Time Efficiency 34 4 Attainability 100 0 10 20 30 40 1 1 1 Harvey EB EB Band 19 0sec EB Attain 100 0 1 SH 6 South Ref Phase 2 No vlo a 13 720 00 ft SH 6 North 127 Now that we know how PASSER III interprets optimal operations at our interchange let s look at using the GA Optimizer on our interchange Click on the GA Optimizer tab and adjust the cycle length range down to 40 seconds to 90 seconds Under fitness routine select Delay based rather than Bandwidth based since this is an isolated diamond Le we are not trying to create progression along the arterial through this interchange Under GA Parameters leave the default values Finally under Diamond Phase Sequence and Offsets choose Optimize phase sequences for all diamonds and Optimize offsets for all signals Note that if we had chosen to not optimize phase sequence or offset the GA Optimizer would have used the settings we chose for
12. 487 62 30 18 38 89 202 24 31 70 32 35 232 22 438 43 193 81 36 17 42 22 79 71 38 00 20 40 135 56 499 35 92 95 29 07 176 67 391 00 141 33 31 00 115 56 255 75 92 44 v wary 000000 0 00 0 145 Chapter Combined Arterial and Diamond Analysis Optimize arterial operation nduding progression through a diamond interchange e can maximize our use of PASSER V s functionality and tools by analyzing complex arterial operations that include diamond interchanges PASSER V s GA Optimizer tool gives you the flexibility to optimize arterial progression while retaining the diamond interchange settings you already developed for a single controller using the PASSER III tool on the interchange This feature combines the utility of the previously separate tools PASSER II and PASSER III An Example Problem For illustration we will use the data set named SH195am p51 in which a small arterial system includes a signalized diamond interchange When you open the file your screen should look similar to the figure on the following page The arterial is geographically oriented north to south in the field and was drawn that way in PASSER V The diamond interchange with one way frontage roads junctions are the third and fourth intersections from the top Similar to large arterials with numerous intersections you can make use of the subsystem analysis features of PASSER V to organize your analysis In this case you can make
13. Stop As illustrated below you can achieve this result by clicking on any one data field for northbound or southbound approach and selecting the appropriate option from the drop down list 3 gt hi 3 gt lt 1 lt 1 fi 2 4 149 676 147 44 635 21 113 80 49 Free Free Free Free Free Free Free Free Free No No No Now you have completed the entry of minimum data needed to perform the analysis Click the Update button You will see the following screen 108 Node Data Export Intersections Controller Type Controller Id 3 Artery 1 at Artery 2 Unsignalized TWSC x Area Type Other v Intersection Data Capacity Data Headway Data MOEs 3 gt e a 2 gt 150 n3 s0 Free Free Free Free Stop Stop Stop Stop No No No o o 400 Settings Platoon Dispersion Model Lock Capacity M Lock Critical Headway IM HCM Lock Follow up Time I C Manar and Baass Update OK Cancel Had you clicked the update button before specifying which two approaches have stop control the program would have given you the following message PASSER Message x Minor street is undefined Now click the OK button You will notice that the node fill has been changed as illustrated below from solid to hatched This demarcation allows easy identification of signalized and TWSC intersections on the map Graphical Display of TWSC Intersection Click on the node again to c
14. The following figure combines all observations The left side of the figure larger screen capture shows the results using the HCM dispersion model and the right side the smaller screen capture shows the results using the alternate dispersion model In this illustration the blue rectangle identifies the results assuming an isolated intersection Notice that these numbers are the same as in the previous case The orange rectangles show the differences between the two platoon dispersion models which ate significant now The last thing to note here is that the capacity for the With Platooning has reduced as compared to Case 1 This reduction is due to the fact that platoon dispersion is more over a longer travel distance Intersection Data Capacity Data Headway Data MOEs Eet jeer wet wer Neu nen gt BE 3 t gt ki 1128 50 5o 1238 4 58 5417 49 240 14 537 88 5700 005265 76 34 151 03 F joio o2 fo 0 83 a A D 0 34 i 410 0 23 0 23 4 5417 49 240 14 537 88 5700 0084 7393 12295 93 128 95 1244 52 61 67 25 57 08 e F F 023 023 ow o2 fo 051 0 49 A a A A A A 0 34 2 39 2 25 Settings gPlatoor exgpersion Model Platoon Dispersion Model Lock Capacity M Lock Critical Headway M Gi C Heh Lock Follow up Time D ffar and Baass 115 Now open Case 3 and repeat the same exercise noticing that the distance between the two intersections has been further
15. Update OK Cancel 90 I Bay is 91 long AM PM L 13 J19 paai T 52 68 1 Ba AM PM R 74 fisoj iia i L 24 44 Truck 2 Ley T 386 635 ZA Wa KX Truck 3 1 a a een l ae 14 S W Military 9 7 Se ee a lt ie Bay is sone 13 Bay is 153 long See eee E E E S T DA EE ar e a a aa AM MRA L 88 ARR AM PM T 397 vaai L 113 113 R 86 Mi d T 85 80 Truck 3 i i i R 45 49 I 1114 Truck 8 1 I I Bay is 126 long i I l S Presa Note that though the data shown are adequate for purposes of generating output using PASSER V Le it shows volumes number of lanes lane usage lane widths it is incomplete in meeting all requirements for a complete signalized intersection analysis Additional data that would make it possible to do a thorough analysis include shoulder presence and width f any pedestrian signal locations presence and location of pedestrian push buttons signal displays for each approach and left turn bay driveway locations close to the intersection etc It is essential that thorough data are at your disposal when conducting any signalized intersection analysis the quality of your intersection recommendations is dependent on yout ability to incorporate all pertinent factors affecting intersection safety and operation To begin your input data entry review the fields at the top of the Timing Data tab folder within the Node Data dialog box Note that this intersecti
16. an interchange respectively l Lane Width 12 common i l Cross St Turn Radii 35 common l No pedestrian features l l ends be Vacant eneth 2 N Church RI Length 150 th P Stop bar Left turn treatment P Protected p Permitted Vacant P p Prot Perm Length 80 Church Vacant Bingham Rd Fast Food Auto Sales Intersection Spacing within Interchange Left turn treatment P Protected p Permitted P p Prot Perm Turn Radii 35 common No pedestrian features 66 An important roadway measute statistic not shown in the previous figure is the spacing between signalized intersections along a given roadway to be analyzed This information is necessaty if coordination is intended or possible between two or more intersections The distance is measured as the length along the roadway between the stop bars of successive intersections The figure below highlights the required reference points for this distance Intersection Spacing in feet Stop bar Intersection 1 Intersection 2 Stop bar An additional consideration in examining diamond interchange geometry is thorough documentation of interior geometry and how the interior is fed by the arterial approaches to the interchange The interchange s ability to process left turns is influenced as much by the arterial ap
17. as SW Military and then click OK Follow the same procedure to rename Artery 2 and Artery 3 as New Laredo and Somerset respectively Our next task is to correct the length and other properties of the link joining our two intersections Click on the Select button and then click on the link Within the Link Data window that appears edit the length so that it is 3425 ft in each direction Also edit the speed to make it 40 mph in each direction as this is our assumed speed along the primary artery S W Military Click on OK when you are finished editing Link Data Next click on the Control button and then click on the left intersection Enter the data from the New Latedo Highway S W Military intersection for this junction Note that our rules for the right turning volume that we used for an isolated intersection Chapter 5 only apply to right turns onto non coordinated arterials For instance at the New Laredo Highway intersection you would look at the right turn volumes and conclude that both the eastbound and northbound right turn volumes could be zeroed out since they could easily be handled as right turns on red or as right turns from their own turn lane However if you were to zero out the northbound right turn volume PASSER V would not be able to account for the flow profile and queuing that would result from those right turning vehicles at the downstream arterial intersection Thus when you ha
18. away from a TWSC intersection When more than one TWSC intersection share the same upstream signal as shown below the platoon originating from the upstream signal may be impeded by the first TWSC intersection IWSC 1 before arriving at the second intersection TWSC 2 HCM does not account for such effects In such cases PASSER V would treat the adjustment factor for TWSC 2 as if TWSC 1 did not exist Upstream Signal TWSC 1 TWSC 2 Shared Lane Capacity HCM 2000 6 uses the following expression to calculate the shared lane capacity CoH DV X b JEn where c shared lane capacity in vph v flow rate of movement in the subject shared lane in vph C movement capacity of movement j in the subject shared lane in vph 57 In PASSER V the capacity of each movement is calculated using the previous formula in an iterative method similar to the one it uses for saturation flow calculations see Chapter 1 As a result the capacity of a shared lane may be different from HCM calculations Flared Minor Street Approaches When a flared approach is present the capacity of a shared right turn lane will increase because the extra storage space allows some of the right turn vehicles to queue at the stop line and complete the movement without obstructing or being obstructed by other movements in the shared lane The increase in capacity depends on storage spaces in terms of passenger vehicles and the average queue length for each
19. currently displayed time space diagram is manually adjusted see next figure If you wish to view the report of the current solution being displayed in the time space diagram viewing window click on the Summary or Detailed Reports tab Note that any manual adjustments to the time space diagram will produce changes in the report s summary statistics and in the MOEs and controller settings output for the manually adjusted intersections 85 Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis Cc Diagram Summary Report Detailed Report Artery List Timing Source Manually Adjusted ae Time ean a ipa Show All Sub Arts Redraw Cycle 85 sec Hide All Sub Arts Efficiency 28 8 Attainability 55 1 Artery List EB EB Band 21 0sec EB Attain 47 7 WB Band 28 0sec WB Attain 62 2 Atep 1 2 40 80 100 120 140 160 180 200 220 Artery 2 Ref Phase 6 No vjj 5 j 49 2500 00 ft Artery 3 Ref Phase 6 No 31 76 2700 00 ft Artery 4 Ref Phase 6 No vj 78 35 1390 00 ft Artery 5 Ref Phase The final tool available for an artery is the Delay Cycle Analysis This tool creates a delay y axis versus cycle length x axis curve for your artery and provides information about the range of cycle lengths that are likely to provide you with optimal or near optimal arterial operation System wide average vehicle d
20. delay These pre calculated splits are then input to the bandwidth optimization algorithm For bandwidth optimization PASSER II starts by selecting a cycle and calculating perfect one way progression in the A arbitrarily selected direction Then it minimizes band interference in the B opposite direction by adjusting phasing sequences and offsets The maximum total band calculated by the program is as follows Total Band G G I where Ga least green in A direction in seconds Gg least green in B direction in seconds I minimum possible band interference in seconds After achieving the best band minimum interference in the B direction the program adjusts the two bands according to user desired options for directional priority The reader should note here that the interference minimization algorithm intelligently searches a very small subset of all possible combinations of signal timings Finally the program calculates delays bandwidth efficiency and attatnability Delay calculation for each interior through movement is based on a macroscopic traffic model whereas delay for all other movements is calculated using the HCM delay formula Efficiency and attainability measure how good a bandwidth solution is Efficiency for a direction is the percent of cycle used for progression Attainability is the percent of bandwidth in a direction in relation to the minimum green split in that direction Theoretically the maximum bandwidth in a di
21. discussed in the following section Left Intersection Right Intersection Lead Lead 4 sa gt gt eT e perertth A PEE Ih One form of three phase timing known as Basic three phase is defined as a lag lag plan that has frontage road phases that are restricted to beginning and ending together Extended three phase operation is a form of Basic three phase lag lag operation wherein one frontage road movement is provided more time than the other The next figure illustrates Basic and Extended three phase diamond interchange operation A v4 Basic Three Phase Extended Three Phase Extended Three Phase Favor Left Side Favor Right Side Jt as iy cS gt or or or GF 28 AEE Tj 2 Four Phase In four phase control the two intersections of the interchange are operationally treated like one large intersection The four phases that give this plan its name are the two exterior arterial phases and the two exterior ramp frontage road phases Four phase plans are known colloquially as either TTI Lead or TTI four phase Protected left turns for the interior movements ate provided The duration of each interior phase is determined by subtracting the sum of the two exterior phase times at that intersection from the cycle length This phasing pattern has become the preferred phasing plan for most diamond interchanges with close spacing 200
22. due to the presence of the channel Recall that left turns on S W Military are protected only while those on New Laredo Highway are protected permitted Assume pedestrian buttons min green 6 sec for all approaches and use a yellow of 4 seconds and ted clearance of 2 seconds Under the Sat Flow Data tab use a PHF of 0 90 and use the truck percentages shown When you are finished with data entry for this intersection click on Update and PASSER V will calculate the green splits fill in the Phase ID of each movement and choose Phase 2 which is EBT and EBR in this case as the coordinate phase Compare your input data to the screen shown After resolving any differences click on OK and move on to the intersection at Somerset Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 SW Miltary at New Lared Pretimed Signal v JEBTEBR BeginofGreen Cycle Length 30 AreaType Other v NTCIP Offset Referencing V yoii a Offset 0 gt Timing Data Sat Flow Data Optimization Data Perfomance Anas Controller Signal MOEs 1 2 gt ja 2 7161 19 1128 Prot Per Prot ProtePen Prot Lead Yes Yes 4 6 4 2 4 4 4 17 78 20 00 1333 15 56 16 18 12 14 Optimization Settings Lock Sat Flows Lock Green Splits Update Dl Cancel 134 At Somerset assume all lanes are 12 ft wide and that left turn bays on each approach are 150 ft long For right turn
23. error regarding the assumption that the opposing through phase is ending can be minimized Note that the 2003 MUTCD prohibits the display of signal indications that create the yellow trap situation Driver Expectancy at Diamond Interchanges When developing diamond interchange timing plans it is essential to consider the environment in which the interchange is located If you are developing a new timing plan for an interchange that is located along a freeway and all other interchanges along that freeway are operating using a TTI four phase strategy driver expectancy develops Essentially drivers will expect this interchange to operate similarly to the other interchanges along this freeway and maybe even in this entire section of the city if most interchanges operate in a TTI four phase mode In this instance there is the expectation that vehicles departing the arterial approach to go through to the other side of the interchange will receive a green through and left turn arrow when they reach the other intersection Also motorists turning left from the frontage road on green with the exception of some U turning vehicles expect to be able to travel through the interchange without stopping again Where the controllers at two or more separate intersections are coordinated for traffic progression coordination may get out of step or fall out of synchronization during cycle by cycle resynchronization during a transition from one timing p
24. in a single stage nor will all of them choose to cross the intersection in two stages HCM 2000 6 introduces the following adjustment factors to account for the joint effects of single and two stage gap acceptance eat 0208 for m gt 0 and y Cri m j Crj V Cn where m number of storage space in the median C phase I movement capacity of movement in vph Cy phase II movement capacity of movement in vph vz flow rate of the conflicting main street left movement in vph Cmj movement capacity of movement j assuming single stage process in vph Once factors a and y are determined the total movement capacity of movement J cp 18 obtained as follows A SO Neu v O y l Cry y lnler v c y l Upstream Signals Capacity adjustment to account for an upstream signal requires the estimation of the proportion of time when the TWSC intersection is not blocked by platoons from that signal and the conflicting main stream flow from the same direction during the unblocked period To this end a platoon dispersion model is used to estimate the proportion of block time at the TWSC intersection If only one major approach has an upstream signal minor movements will encounter two distinct flow profiles namely flow when there is no platoon unblocked period and flow when a platoon is passing through the intersection blocked period However when upstream signals exist on both sides of the TWSC intersectio
25. increased to 4470 feet 0 85 miles The following figure illustrates the results Notice that with increases in link distance the results of both dispersion models are converging towards the Isolated case However the Manar and Baass model is still predicting higher capacity than the HCM model Intersection Data Capacity Data Headway Data MOEs cet lear wer wer ne wer gt BEML 3 e ki 128 50 5o 1233 4 s 023 023 010 A a fa A 0 34 0o23 023 oo 025 A A A 0 34 Settings atoon euspersion Model Platoon Dispersion Model Lock Capacity D Lock Critical Headway M Gy C HEM Lock Follow up Time J afar and Baass 116 Now open Case 4 which illustrated the impact of link volume levels on the prediction of capacity by the two dispersion models Repeat the same exercise and make observations noting that westbound volume arriving from the upstream signal had reduced by 50 percent while the geometry has remained unchanged from Case 3 The following figure illustrates the results Notice that both models are now predicting the same capacity as the isolated case These results also illustrate the volume dependency of the Manar and Baass model Intersection Data Capacity Data Headway Data moes Reduced Westbound Volume 103 10 115 09 83 84 F F 0 23 O61 061 A B 297 103 10 103 10 5417 49 240 14 537 88 5700 00 12 4
26. is detected over the loop In pulse mode a short detection is sent to the controller following a vehicle arrival at the loop At virtually all signalized intersections detectors are operated in presence mode Another feature of detector operation is that detection can be set in locking and non locking memory modes Under locking memory a detection call is remembered by the controller until the phase called by that detection is serviced Under non locking memory the controller only registers a call when a vehicle is over the sensor An example of the usefulness of this mode is right turn on red RTOR situations where the vehicle is effectively forgotten if it is able to make a safe RTOR maneuver Detector layout for common signalized intersections is covered in the Traffic Control Systems Handbook 10 and the Traffic Detector Handbook 11 However there are few existing references to detector layout for signalized diamond interchanges Thus the system employed by TxDOT is presented here for common detector placements for the three and four phase control patterns Three Phase Control The operational practices for use with three phase control are presented below A long i e 6 ft X 40 ft rectangular shaped inductive loop detector is used in the interior of the interchange for left turns The through lanes are equipped with 6 ft X 6 ft loop detectors spaced 200 ft in advance of the stop bar The placement of the detectors alo
27. left turn movements for the interior approaches are provided In general three phase operation tends to produce less overall delay compared to four phase operation when there is adequate space within the interchange interior to store queued vehicles Three phase operation is generally recommended for interchanges with moderate to high traffic volumes wide spacing between the two intersections and high through volumes on either the arterial or frontage road phases As discussed in the previous section the three phase strategy allows for varying left turn sequences including lead lead lead lag lag lead and lag lag In the lead lag and lag lead variations of three phase operation heavy left turn traffic from the right or left frontage roads respectively is allowed to progress through the interchange Variations of the three phase timing patterns are shown in the next figure All of the three phase variations shown have no restrictions on when phases can begin and end with respect to one another any of the three phases for the left intersection can occur in part or whole with any of the three phases for the right intersection Three phase operation should generally be used when the diamond s intersections are spaced greater than 400 ft apart or where the interior left turn volumes of the interchange are low With intersections spaced between 200 and 400 ft and balanced ramp traffic three phase or four phase timing may be appropriate
28. link For calculating the available storage shock wave theory is applied to find the actual available storage at each second c If the next link is blocked flows are stored in the current link itself The routine is capable of performing Step 4 for a specified number of cycles however in the current version of PASSER V this number is fixed to two cycles The routine calculates and reports several MOEs at the end of simulation DAR uses the following assumptions when performing its calculations e fractional flow e no intersection blocking e the only effect of queue spillback will be a decrease in flow from the upstream movements into this link e lane blockages are only considered when all storage space of a lane is used and e no right turns on red Lastly DAR is limited to linear arterial systems It is not capable of simulating networks and it is applicable to pretimed signals only A more detailed description of DAR along with its shock wave model is provided by Kovvali 22 Optimization and Analysis Tools PASSER V provides a number of analysis and optimization features for arterials and for diamond interchanges using a single controller With the aid of this program the user can develop timing resulting in maximum progression efficiency or minimum system delay These features are provided under different tools All of these tools calculate equal saturation 43 green splits using Webster s method Furthermore these
29. may be beneficial to change at least some data values here before beginning to create a new data set Examples of data you may change often are the link speed and peak hour factor For instance if most links in your new data have speeds of 45 mph and you ate planning on requesting the program to adjust entered volume data by a peak hour factor of 0 95 it will be worthwhile to enter these values on this screen before beginning to create the network Note that it is necessary to click Update or OK to register the information you enter on PASSER V s input screens including this screen Also note that the data you enter here become part of the data file when you save and can be different for each file The Signal tab allows you to enter default timing data for signalized intersections As illustrated in the next figure these data items include minimum green time yellow time red clearance all red time lost time for left turn and through plus right movements and cycle length information When you create a new signalized intersection the program will use all values except cycle length range as default values for that intersection Cycle length range is used as default by optimization tools Note that any of these values can be changed later 76 System Parameters Project Info Defaults General Signal Twsc a Cycle Lengrh sec 90 Cycle Length Range sec From To Increment 120 S 5 Data on the TWSC ta
30. movement in the shared lane In general longer usable storage spaces increase capacity of the shared lane Similarly the longer the queue length the smaller the increase in resulting capacity Control Delay For TWSC intersections control delay for each minor movement provides a measute of its level of service LOS However HCM does not define LOS for the TWSC intersection as a whole The following expression is used to estimate control delay for a minor movement 2 3600 c Lv c d 2 900 0 25x mI a B600 cn y es lems J Cm j C m j m j 450 x 0 25 where d control delay of movement j in seconds vehicle v flow rate for movement J in vph Cm j Z Capacity of movement in vph Queue Length Queue length is another important indicator of the operational quality of a TWSC intersection HCM 2000 6 provides the following formula to estimate the 95 percentile queue length 2 On 900 0 25 x E J aa lead ea 95 j x T t J J c c 150x0 25 3600 m j m j where Qs 95 percentile queue in vehicles v flow rate for movement in vph 5 capacity of movement in vph m j This value is an estimated measure that the actual queue length will be shorter than this value 95 percent of the time 58 Impact of TWSC Modeling on Various Tools The previous chapter provided detailed descriptions of different tools in PASSER V This section describes how these tools
31. of several decades and has gained acceptance from the user community as a sound model TRANSYT 7F uses a combination of exhaustive hill climbing and GA based optimization methods TRANSYT 7F uses a delay based traffic model In other wotds it is primarily designed to select signal timings that produce minimum system delay In addition it provides a capability to select several secondary objectives including minimization of stops and maximization of progression opportunities During its optimization process TRANSYT 7F generates second by second flow profiles of vehicles on all links in the network Then it analyzes these profiles to determine MOEs TRANSYT 7F has two delay based traffic models The first model original model performs the optimization in a link wise fashion by optimizing timings for one link at a time This model does not accurately account for queue buildup because it treats a queue of vehicles as an upward stack at the stop bar However it works well for undersaturated traffic conditions Users all over the world have extensively validated this model The second model was recently added to remove the limitations of the first model This model takes into consideration the formation and dissipation of queues in space In addition it accounts for flow interactions on adjacent links through a step by step analysis of all links in the system Conceptually this model is better suited for the analysis and optimization of congested ove
32. of your network will remain the same i e it is representational rather than to scale even if you edit and change a link s length Also if you have manually edited a link to change its length and then you go back and move one of the nodes for that link PASSER V will automatically recalculate a scaled length for the link You will have to go back and re edit the link length to re enter the value you manually entered the first time Once PASSER V has automatically created nodes at link junction points you can select the Control button from the PASSER V function tool bar and click on a node see figure below It is within the Node Data window that appears that you enter the majority of the intersection volume geometry and signal control information for each intersection of your network For each node you specify the cycle length usually left as the default value and PASSER V will perform a cycle length analysis over the range of cycle lengths you specified earlier offset usually left blank for optimization and area type where the signal is located Next you select the lane use permitted movements from each lane specify your input movement traffic flows and select desired controller settings from tabs provided within the Node Data window Finally you can choose to view a delay versus cycle length analysis for this intersection after you have entered the required input data or check on output MOEs for this intersection Node D
33. or facilities where sufficient internal storage space exists Bandwidth Analysis Routine For a given timing plan cycle length splits offsets and phase sequences a bandwidth analysis routine BAR calculates the progression bands in both directions of an arterial In its calculations the routine calculates bands a geometric quantity between all signal pairs This routine was developed for use in generating time space diagrams and for use by bandwidth based optimization using the genetic algorithm After calculating the bands this routine calculates bandwidth efficiency and attainability using equations provided in the section describing the interference minimization algorithm This routine is extremely efficient in its calculations Delay Analysis Routine PASSER V s delay analysis routine DAR employs mesoscopic simulation strategy In other words it simulates fractional flows and updates them every second It performs the analysis of traffic conditions using a two step process described below 1 initialization and 2 simulation and recording of MOEs 41 For these steps the model uses two subroutines the undersaturated routine and the oversaturated routine The program conducts the initialization step for two signal cycles The first cycle uses the undersaturated routine to get a preliminary estimate of queues and the second cycle uses the oversaturated routine to ensure that the queue estimate is realistic After the ini
34. pp 49 53 152
35. speed of vehicle through the intersection in feet second 13 Red clearance equations depend on the type of application where the Traffic Engineering Handbook T states that it is recommended to use Equation 2 where there are no pedestrians Equation 3 or 4 whichever is longer is used where there is the probability of pedestrians crossing and Equation 4 where there is significant pedestrian traffic or pedestrian signals that protect the crosswalk Appropriate detector placement combined with appropriate yellow and red clearance time ensures that motorists are not trapped in a dilemma zone The dilemma zone is a point where a driver cannot stop at a reasonable rate of deceleration and where the same driver cannot pass through the intersection within the yellow time allotted The red clearance interval is primarily a tool to avoid displaying unusually long yellow times For more information see Traffic Engineering 8 The following tables reproduced from an Institute of Transportation Engineers informational report 9 provide yellow change intervals and red clearance intervals for various combinations of speed grade and intersection approach width Note that a red clearance interval may be used to meet the required time shown in the first table when the maximum length of the yellow change interval is set at 5 0 seconds Yellow Change Interval Lengths in seconds
36. the 170 that add more memory Le more timing plans more functions and newer processors with the simple exchange of a circuit board plug in module e NEMA TS1 and TS2 The NEMA TS1 standard came about roughly in the same time frame as the original Type 170 specification Unlike the 170 specification the TS1 standard defined the functionality e what the controller device was supposed to do and what features it was required to have of the controller device rather than the equipment The TS1 also standardized cabinet wiring and harnesses added a conflict monitor i e a cabinet watchdog device and developed a uniform phase reference NEMA TS1 compatible controllers have evolved over time because manufacturers were able to use new microprocessors and expanded memory to fulfill TS1 functional requirements and each manufacturer was able to add additional functionality 1e closed loop system to make their products more marketable Unfortunately each manufacturer pursued functionality outside of the TS1 standard differently and this higher tier of functionality is not compatible across manufacturers The TS2 standard 3 is a major leap in the modernization of the TS1 for current electronics technology Cabinet communications no longer take place using discrete electronic signals over hundreds of wires but over a communications bus The conflict monitor of the TS1 has been replaced by a much more powerful programmable malfunction m
37. the external to external movements for which delay calculations use the HCM equation for calculating uniform and incremental delay 6 Oversaturated Routine As described above the undersaturated routine treats a queue as an upward stack and thus is unable to model queue spillback and effects of any upstream blocking resulting from it In addition it does not account for flow interactions between adjacent links The oversaturated routine overcomes these limitations It uses shock wave theory to more accurately assess delays in congested conditions The program applies this routine during the second cycle of initialization and for all full simulation cycles The oversaturated routine conducts a second by second stepwise analysis of incoming flow at the stop line the available queue storage in the downstream link and the outflow from the link It updates conditions on all links of the arterial each second The incoming flow at the stop line for the internal movements is obtained by applying the TRANSYT 7F platoon dispersion model The routine uses shock wave theory to keep track of the back of the moving queue at each link on a second by second basis In addition it keeps track of the available link storage If the back of queue reaches the upstream intersection the available storage becomes 0 and movement blockage occurs until some storage becomes available This routine performs calculations using the following steps 42 1 For the
38. the diamond during data entry Or if we had opted to let the GA Optimizer optimize according to individual interchange settings it would have optimized either or both depending on whether we locked phase sequence or offset during data entry You can review these settings by clicking on the Control button from the button bar clicking on a node of the diamond and selecting the Optimization Data tab Options for locking the diamond phase sequence and offset are found at the bottom of the window Return to the GA Optimizer under the Tools button if you left to check your diamond interchange input data and check your settings If you wish you can click on the Adv Options tab to change settings for the GA based optimization though this is not recommended When you ate ready return to the Input tab see next figure under the GA Optimizer and click on Run Optimization Analysis Tools Select PASSER III GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis a Input Adv Options Summary Report Detailed Report Artery List Cycle Lenath Range Harvey From To Increment C Show All Sub Atts jao joas p amp Hide All Sub Arts 4 Fitness Routine Artery List Delay Based Bandwidth Based Harvey GA Parameters Population Size 20 Num of Generations 150 Diamond Phase Sequence C Do not optimize phase sequence for any diamonds Optimize phase sequences for all diamonds Optimize according to
39. times we will interpret the change and clearance intervals as 5 seconds of yellow change plus 1 second of red clearance for S W Military and 4 seconds of yellow change plus 2 seconds of red clearance for S Presa Note that though we changed the allocation of time between the yellow change interval and red clearance interval slightly for each roadway we made sure that the sum of the two for each roadway was at least as large as the sum called for by the tables in Chapter 1 When you are finished your Node Data window should look like the screenshot on the next page Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal mail x Begin of Green Cycle Length 90 Area Type Other v NTCIP Offset Referencing fv C BegnofYellow offset OS Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs 2 gt 19 68 113 ProtePert Prot ProtePent Prot Lead Lead Optimization Settings Lock Sat Flows Lock Green Splits Update Cancel The final detail we need to check under the Timing Data tab is the Lost Time entry in seconds for each phase Lost time represents the loss in signal efficiency that results from hopefully no vehicles moving through the intersection at the end of a phase say during the red clearance interval and the fact that no vehicles are moving through the intersec
40. to coordinate the intersections The PM peak hours tend to have slightly higher volumes than the AM peak so we will start with a PM peak analysis It is likely that separate timing plans will be necessary for the AM and PM peak periods or that vehicle actuated control for these intersections will be used Since we are coordinating two or more intersections we will have to have control hardware that is capable of operating in a pretimed coordinated or semi actuated coordinated mode Such control can be implemented using a time based coordinated system or a closed loop system with wireline or radio communications between intersections If the system is vehicle actuated the loop detectors or other sensors must be calibrated and maintained To render the network of two intersections in PASSER V begin by selecting the Two way link button and drawing a line 4500 ft long from left to right approximately halfway down your screen Then draw a north south crossing roadway such that it intersects the first artery about 2 5 grid squares 500 ft from its left side Repeat this procedure on the other 132 side of the network for the crossing roadway to the east When you are finished with these steps your screen should resemble the following figure Next we will enter the actual artery name Click on the Select button from the menu bar Then click on the link labeled Artery 1 and the Link Data window will pop up Rename Artery 1
41. turns have the added advantage of avoiding the yellow trap issue for our southbound and northbound protected permitted left turns The next phasing option is whether or not to allow compatible phases on one side of a barrier to Overlap The following examples illustrate this phasing option Phasing with Overlap Phasing without Overlap aS A a Ovalap A ES a verlap Split Phasing At our intersection we will allow the phases to overlap by keeping this selection to the default value of Yes Generally allowing overlap operation is more efficient than not allowing the overlap to occur It should be noted that this definition of an overlap is different from that in traffic controllers In the latter case overlap is defined as the combination of two or more basic phases 96 Minimum green times in PASSER V are either the minimum green times you would enter into a controller for vehicular phasing based on driver requirements setback detector location etc or where there are no pedestrian buttons the combination of the pedestrian walk and clearance times If there are no pedestrian buttons which would activate pre programmed pedestrian phases with their own minimum WALK and flashing DON T WALK clearance times the sum of the pedestrian walk and clearance times will likely be greater than the vehicular minimum green time and should be entered In other words if the intersection doe
42. value to guarantee the display of a phase of uniform length from cycle to cycle In these cases the minimum green is based on the timing requirements for the maximum green time i e the traffic demand using the phase 11 pedestrian times and phasing requirements for all phases However it must be pointed out that setting the minimum and maximum green times to the same values eliminates phasing flexibility A preferred method for establishing pretimed control is to set the minimum green time accordingly to normal agency practices and set the maximum time Max 1 equal to the desired duration of the green time portion for each phase Then the phase is set to maximum recall recall to max green time and along with all other phases set the same way constitutes the pretimed cycle length This operation allows flexibility since the maximum can be changed easily as the peak off peak timing plan changes and the alternate maximum time Max 2 can be called by time of day for phase time adjustments if necessary For actuated control and this should be checked for pretimed operation as well the minimum green time will always have a lower bound It must be at least sufficiently long to allow motorists to recognize that the signal has gone to green and begin responding to the green signal indication i e remove brake and begin accelerating Minimum green times are governed by the practices of the responsible agency but usually vary from 5 to 7 se
43. variables have complex interacting dependencies and a direct optimization algorithm is unknown Selection and mutation alone cannot solve such problems when the solution space is large Crossover is the real power behind evolutionary algorithms and it improves performance by many orders of magnitude in most problems Mutation Probability Mutation probability is the probability with which a given chromosome changes its state between generations A high mutation probability will essentially lead to a random search of the solution space Replacement Probability Replacement probability specifies the amount of overlap between generations It only applies to steady state GAs Analytical and Simulation Models Several simulation or evaluation models have been included in PASSER V These models are used by optimization algorithms and by other analysis tools In this section we describe these models A significant portion of this section is devoted to the new traffic model included in the program Analytical Models for Basic Calculations PASSER V uses HCM and Webster s methods for calculating saturation flow rates and green splits respectively 6 24 The saturation flow rates for movements sharing a lane are calculated using an iterative procedure that prorates saturation flow rate of the shared lane using volumes of movements served by the shared lane Furthermore PASSER V uses the first two terms of the HCM model for calculating delay
44. we should start by accessing the Delay Cycle Analysis tab Viewing its graph of cycle length x axis versus delay y axis we find that the optimal cycle length for the interchange is in the vicinity of 45 to 50 seconds We can use this information to constrain the range of cycle lengths we analyze using other tools in PASSER V a good range is probably 40 to 90 seconds with a 5 second increment 125 Next let s look at the Volume Analysis tool First click on the Input Information button beneath Volume Analysis and select a cycle length range of 40 to 90 seconds Then click on Chart You will see that for this example problem all three phase Extended and Basic three phase overlap each other on this graph timing plans provide adequate capacity or throughput for this interchange Le current volumes shown in red ate beneath the throughput capacity limit shown in yellow For four phase timing plans these cycle lengths do not provide sufficient capacity If you wish you can review the Critical Movement i e movement having the high v c ratio for each phasing sequence at each cycle length and Throughput table of throughput capacities shown in the graph tabs for details on each cycle length and sequence analyzed Now that we know what cycle length range to analyze and we know in general that three phase sequences have higher capacity than four phase operation within our analysis range we are ready to use PASSER III to
45. will notice that these two groups of MOEs are different now Notice that platoon arrivals from the upstream signal located 836 ft away provide more capacity for the northbound movements at the TWSC intersection than random arrivals The resulting delay and v c ratios are also better Intersection Data Capacity Data Headway Data MOEs 76 34 151 03 F 0 83 D 4 10 023 023 o1 025 A A a A 5417 49 24014 537 88 5700 008 91 96 133 34 1244 19 5414 e F 023 023 o10 02 fo 0 47 A A a A A 034 j 216 Settings Platoon Dispersion Model Lock Capacity M Lock Critical Headway M HCM Lock Follow up Time M C Manar and Baass Now change the platoon dispersion model from the default value to Manar and Baass and click the Update button You will observe no changes for the Isolated calculations as expected and minor changes for the With Platooning case The reason for this similarity between the results of HCM and this model is the close proximity of the adjacent signal which provides little distance to the platoon to disperse Now open Case 2 and repeat the same exercise The first thing you will notice is that the distance between the two intersections has been increased to 2270 feet 0 42 miles which is 2 7 times the distance in Case 1 Next look at the MOEs for the TWSC intersection and compare these MOHs to those obtained using the Manar and Baass model
46. you choose the PASSER II tool you will need to specify a cycle length range and an increment for successive runs You will also need to specify the type of bandwidth progression you wish to establish indicate whether or not you want PASSER II to try and fine tune offsets to minimize delay and indicate whether you want PASSER II to output MOEs for each cycle length analyzed or just for the optimal cycle length After specifying all of these details clicking on Run will execute the PASSER II optimization engine and produce your output A Summary Report is available to give general summary details bandwidth efficiency attainability delay stops etc for each cycle length analyzed and a detailed report includes intersection specific optimal signal settings and MOEs The Plot option lets you compare a pair of selected MOEs against the cycle length range analyzed Clicking on the T Sp Diagram tab at this point will produce a time space diagram for your PASSER I best efficiency solution see below Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis c Diagram Summary Report Detailed Report Artery List Timing Source Original ED Time eae 4 C Show All Sub Arts A A Redraw Hide All Sub Arts Efficiency 28 1 Attainability 54 2 S Artery List EB EB Band 19 0sec EB Attain 46 3 Print WB Band 26 0sec WB Attain 61 9 iter 20 40 60 8 amp 0 100 120 140 160 160 2
47. 00 220 Artery 2 Ref Phase 2 No yj 0 j 40 2500 00 ft Artery 3 Ref Phase 2 No 42 3 0 2700 00 ft Artery 4 Ref Phase 2 No vj 3 38 1390 00 ft Artery 5 Ref Phase 2 82 If you would rather use the GA Optimizer to produce a genetic algorithm derived solution for your arterial click on the GA Optimizer tab The GA Optimizer computes an optimal solution for delay or bandwidth based on which you choose by having seed or estimated solutions compete against one another to filter or evolve until a preferred solution emerges Options for the GA Optimizer include specifying a cycle length range and increment and telling the optimizer whether or not you want to optimize phase sequence and offset for each intersection along the arterial see below Note that the reason the GA Optimizer gives you these options is that in certain situations such as when analyzing an artery that includes a diamond interchange you may want to optimize the diamond first fix its signal settings and then optimize the rest of the artery around the diamond Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis g Input Adv Options Summary Report Detailed Report Artery List Artery 1 Increment rx o Show All Sub Arts jso hoas p E Hide All SubArts Cycle Length Range From i Fitness Routine Artery List C Delay Bas
48. 22 Node Data Export Intersections Controller Type Coord Phase Offset Reference Paint Controller Id 3 Harvey atSH 6 South Pretimed Signal x Begin of Green Cycle Length 90 Harvey at SH 6 North Area Type Other v NTCIP Offset Referencing V C Begin of Yelow Offset 0 lt Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs 0 o 0 o o o fo 0 o 0 o 0 100 100 fioo 100 fioo 100 j1 fioo 100 fioo 100 fioo 100 Optimization Settings Diamond Phasing Lock Sat Flows i Four Phase oe ic 3 4 Lock Green Splits a pa a Update OK Notice from the figure above that the node data required by both intersections of the interchange are displayed simultaneously A vertical red line separates the input data for one intersection from the other intersection Another feature that makes an interchange different from a regular intersection is that a Diamond Phasing select box is available near the bottom center of the Node Data window Using this selection box you can tell PASSER V which phasing sequence you wish to use at this interchange Note that only one selection is possible here but when you optimize the interchange you can have PASSER V analyze all three options to find the most optimal signal timing solution Also note that if you select the Four Phase option a second editing box appears to the right This box allows you to adjust the four phase transition overla
49. 3 1912 24 00 1300 11 00 2400 26 00 85 71 9286 78 57 21 73 3291 44 0 00 0 00 1853 1843 57 00 2800 2900 4071 40 00 100 00 100 00 100 00 21 97 2607 86 0 00 0 00 1853 1906 64 00 3300 31 00 4000 41 25 94 12 100 00 88 57 22 71 2461 88 0 00 0 00 1853 1870 60 00 3000 3000 4000 40 00 96 77 100 00 9375 22 81 2614 82 0 00 0 00 1853 1888 72 00 3500 37 00 4235 41 18 100 00 100 00 100 00 23 86 2526 88 0 00 0 00 1853 1869 76 00 37 00 3900 4222 41 11 100 00 100 00 100 00 25 18 2564 43 0 00 0 00 1853 1869 Notice that in this case as in most cases longer cycle lengths are associated with higher overall average delay To get a better idea about the trade offs between different performance measures with respect to different timing plan it is helpful to view of plot comparing different performance measures Click on the tab named Plot to view plots of any two selected performance measures versus cycle length The performance measures you ate allowed to plot include bands efficiency attainability delays and stops Optimization Analysis Tools Select PASSER II Ga Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis C SubNetwork List Input Output Summary Report Detailed Report Artery List SW Military MOE s Total Efficiency v Average Delay X C Show All Sub Arts Hide All Sub Arts Artery List SW Military SW Military Efficiency and Average Delay N w gt ES 2 T 2 2 5 p N N o
50. 3 Harvey at SH 6 South Pretimed Signal X a BeginofGreen Cycle Length po Harvey at SH 6 North AteaType Other v NTCIP Offset Referencing V Penak Offset 0 gt Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs aft lt a f gt 180 280 37 2 Prot Prot Prot Spit Lead 4 3 1 4 0 0 Optimization Settings Diamond Phasing Lock Sat Flows C Four Phase ce ic 34 Lock Green Splits a pe ee Update ol Cancel Next we need to enter the signal setting details for the interchange We will assume that this interchange has pedestrian push buttons so we do not need to manually calculate walk and clearance times to come up with our minimum phase times Right turns on red are allowed on all approaches but there are shared lanes for our eastbound northbound and westbound right turn maneuvers with no turn bays or channels Also since this is a diamond interchange the frontage road through and left turn phases are intimately tied together and begin and end at the same time resembling split phase operation which is automatically selected by the program The program automatically selects lag lag or lead lead phasing based on the Diamond Phasing option at the bottom of this screen Because this interchange has rather wide spacing greater than 400 ft we will select a three phase timing plan which has a lag lag operation Finally our speed is 40 mph on all appro
51. 4 64 sec veh Performance Analysis Cycle Length Range Analysis Min Delay Cycle 95 sec Min Delay 33 16 sec veh Delay sec veh 100 110 Cycle Length Demand Capacity HCM Dalay j OK Cancel To change the cycle length for this intersection simply click on the Cycle Length text box in the upper tight corner and type in 95 then click on the Update button at the bottom of the screen If you review the splits under the Timing Data folder you will see that they have changed for the new cycle length If you would like to see a phase indication color coded phasing diagram for your intersection click on the Controller tab folder Phasing splits are shown for each phase in NEMA ting order and yellow change and red clearance intervals are shown in proportion to their duration within each green split A screenshot of this diagram is shown below Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal x EBTEBR x BeginotGreen Cycle Length 95 Area Type Other v NTCIP Offset Referencing p C BesnofYetow set OS Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs OK Cancel 101 Next you will want to review the measures of effectiveness for your timing solution MOEs ate found under the Signal MOEs tab folder and include delay seconds per vehi
52. 4 83 84 83 84 e F F 023 023 o0 013 061 061 A a a A B B 0 34 297 297 Settings Platoon Dispersion Model P Diapersion Model Lock Capacity M Lock Critical Headway M Cc Lock Follow up Time M Manar and Baass Chapter Analysis of Signalized Diamond Interchanges Using PASSER V to optimize timing for a signaled diamond interhange result of working on this problem you will know how to code links that form an arterial roadway as well as how to analyze the special operational issues associated with diamond interchanges O ur next exercise will analyze a diamond interchange with a single controller As a The diamond interchange we will use as our example is the diamond located at the junction of the East Bypass SH6 and Harvey Road FM 30 in College Station Texas The following two figures provide the geometric and turning movement details for the interchange SH 6 East Bypass Protected Permitted e Speed 40 mph e Bay Length 300 ft All lanes 12 ft Protected Only 119 211 gt gt 764 180 265 a i Va Harvey Rd s N e 2 Trucks on all Entering Data Start PASSER V and create a new file To make sure we ate ready to save our work at any time go ahead and click on File and then Save As and give this file the name Diamond Next click on the System function button and enter in the project name organization and your name Then click on the Tw
53. 6 B gt 10 20 0 6 lt 0 7 C gt 20 35 0 7 lt 0 8 D gt 35 55 0 8 lt 0 85 E gt 55 80 0 85 1 0 F gt 80 gt 1 0 Finally you may want to save various sets of information about the intersection to a file for later reference By clicking the Export menu located on the left top corner of the Node Data window the program gives an option to export selected data to a file The four options available are illustrated in the figure on the next page In the process it asks the user to select a file name 102 Node Data Timing Data Sat Flow Data Signal MOEs Sat Flow Data Timing Data For example by clicking AIl followed by a file name selection will prompt the program to save all data under Timing Data Sat Flow Data and Signal MOEs tabs to a comma separated value CSV file with the selected name The following illustrates a portion of such a file when viewed using a spreadsheet Intersections Artery 1 at Artery 2 Optimization Data Al A 7 gt D E F 1 Timing Data 2 Artery George Bush Dr George Bush Dr 3 Movement EBL EBT EBR WBL WBT 4 Lane Assignment 1 2 11 gt 5 Volume vph 100 100 100 100 100 6 Movement Type Split Prot Prot Split Prot 7 Left turn Phasing Lead Lag 8 Overlap No 9 Phase ID 5 2 2 1 6 10 Minimum Green sec 6 6 6 6 6 11 Yellow sec 3 3 3 3 3 12 Red Clearance sec 1 1 1 1 1 13 Lost Time sec 4 4 4 4 4 14
54. 65 70 75 80 85 90 Cycle Length Efficiency Avg Delay sec veh 138 After reviewing the summary report and the plots of efficiency and average delay it seems that a good compromise might be to select a 70 second cycle length to get the best of both worlds Right click your mouse over the output table and select the 70 second cycle timing plan To review the details regarding this timing plan click on Detailed Report The Summary tab under Detailed Report tab displays the overall performance of the selected timing plan including efficiency attainability band width delay stops throughput blockage and starvation time Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis A Input Output Summary Report Detailed Report Plot Artery List SW Military Swale Length sec 70 xl HTML File Print C Show All Sub Arts Hide All Sub Arts iming Pian MOES Artery List aS en Optimization Tool PASSER II SW Military Summary Report SW Military Cycle s 70 Efficiency 40 71 GREAT PROGRESSION Attainability 100 00 INCREASE MIN THRU PHASE EB Band s 28 00 WB Band s 29 00 Avg Delay s v 21 97 Total Stops v h 2607 86 Entry Vol v h 1853 00 Exit Vol v h 1906 00 Veh Blocked v c 0 00 Starve Time s h 0 00 NOTE Arterial Progression Evaluation Criteria EFFICIE
55. Arts Cycle eee Redraw i Hide All Sub Arts Efficiency 40 7 Attainability 100 0 Aley List EB EB Band 280sec EB Attain 100 0 Print WB Band 29 0sec WB Attain 100 0 aa ee pip o p o o New Laredo Ref Phase 2 No 0 28 2465 00 ft Artery 4 TWSC 960 00 ft Somerset Ref Phase 2 No 58 j 10 Optimization Analysis Tools PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis ob etwork List Diagram Summary Report Detailed Report Artery List SW Military Summary Timing Plan MOEs HTML File Print C Show All Sub Arts Hide All Sub Arts TWSC9 SW Military and Artery 4 Artery List SW Military Avg Avg Vehicles Delay Delay Thruput Capacity Blocked Phase s v LOS v h v h vic EBL 9 11 100 01 975 42 0 00 EBT 0 00 877 89 5409 96 0 00 EBR 0 00 40 01 246 53 0 00 WBL 9 93 26 17 762 21 0 00 WBT 0 00 811 38 5511 74 0 00 WBR 0 00 23 53 160 02 0 00 SBL 64 19 55 56 113 19 0 00 SBT 74 78 11 11 62 41 SBR 39 99 22 22 124 82 0 00 NBL 60 16 33 33 97 31 0 00 NBT 108 88 11 11 45 20 NBR 22 55 66 67 271 19 0 00 gt ommommrr rey Signal6 SW Military and Somerset Ref Phase 2 Begin of Green Phase Offset 58 sec Avg Avg Vehicles Delay Delay Thruput Capacity Blocked Stops sv Los v h v h vic v h 36 50 56 95 82 75 0 00 45 76 17 24 857 59 1730 78 0 00 583 28 17 24 85 98 173 53 0 00 58 48 35 39 112 22 232 07 93 85 21 37 683 33 2035 71 0 00
56. Cycle Lengt HTML File Print Show All Sub Arts Hide All Sub Arts Timing Pen OES Artery List SW Military Optimization Tool PASSER II Timing Plan SW Military Cycle 70 sec Signal 3 SW Military and New Laredo Ref Phase 6 Begin of Green Phase Offset 0 sec Signal Phase Phase Ring Barrier Position Split sec Movement 1 1 1 1 12 WBU WBL 1 34 EBT EBR A 12 NBU NBL 1 12 SBT SBR SBU SBL 2 30 WBT WBR 2 16 EBU EBL 2 12 SBU SBL 2 12 NBT NBR NBU NBL Signal 8 SW Military and Somerset Ref Phase 2 Begin of Green Phase Offset 0 sec Signal Phase Phase Ring Barrier Position Split sec Movement 2 1 1 1 28 EBT EBR 1 13 WBU WBL 1 15 NBU NBL 1 14 SBT SBR SBU SBL 2 12 EBU EBL 2 WBT WBR 2 2 SBU SBL NBT NBR NBU NBL Finally click on MOEs tab and you will see a detailed report on performance measures with respect to each phase as illustrated in the figure on the next page These performance measures include delay delay level of service throughput capacity blockage stops and starvation 140 Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis c L Input Output Summary Report Detailed Report Pict Artery List SW Military Cycle Length sec HTML File Print Show All Sub Arts Hide All Sub Arts Summary Timing Plaj Artery List SW Military Optimization Tool PASSER II Performance Measures SW Milita
57. DAR this tool indirectly accounts for TWSC intersections Program Limitations When a TWSC intersection is located very close to a downstream signal queues from that signal may partially or fully block it severely reducing its capacity This blockage effect is not considered in the isolated intersection analysis since the HCM and literature review provide no guidelines or methodology to deal with this situation On the other hand this effect on movement delays of a TWSC intersection is partially considered in DAR in which the corresponding movement flow will be zero when a TWSC intersection is blocked Note that capacities of the affected TWSC movements should vary as the blockage clears but they are assumed to remain constant in DAR 60 Chapter Data Requirements All the things you need to know about the arterial and diamond interchange before you can start to analyze it essential to the integrity of any analysis that the input data are up to date accurate and representative of general conditions This maxim is especially important with traffic engineering information in that signal timing a primary output of the process has a direct bearing on safety and efficiency for the motoring public R the maxim for analytical procedures Garbage in garbage out It is In Chapter 3 we will talk about the types and quantities of data that you will need to perform an analysis and optimization of an arterial roadway These data in
58. Data Entry Our next step is to enter the volumes in vehicles per hour for each approach s turning movements Note that the volumes you enter here do not need to be adjusted for peak hour factor see Chapter 3 effects we will enter peak hour factors for each movement later in the data entry process Since we are using the PM peak volumes enter the volumes 149 676 and 147 for the left through and right volume fields respectively for the eastbound approach Then enter the volumes for the movements on the other three approaches When you ate finished your Node Data window should look like this Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller ld 3 Artery 1 at Artery 2 Pretimed Signal x Begin of Green Cycle Length 30 AveaType Other x NTCIP Offset Referencing fw BesnofYelow Oteo Timing Data Sat Flow Data Optimization Data Performance Arasi Controller Signal MOEs m Optimization Settings Lock Sat Flows Lock Green Splits Update OK Cancel Our next task is to select the appropriate Movement Type for each of our left turn movements To review the available options click in the Movement Type field for the eastbound left EBL A pull down box with the options shown below will appear Perfo Timing Data Sat Flow Data Optimization Data 3 gt lt 1 149 676 147 Prot Y Prot Prot Prot Fay Sp
59. HAND SIDE INTERSECTION 4 ARTERIAL QAAR RIGHT HAND SIDE INTERSECTION Time Signal Change Phase a Last vehicle hits detector on left frontage road 64 b Gap Time 4 gaps out A 5 yellow begins c Yellow A 5 red clearance begins d Red Clear A 5 red clearance ends 06 begins e Start Up Y 4 yellow begins Green Time vehicle from A reaches left intersection Measured REIRE E f Yellow 64 red clearance begins g Red Clear Y 4 red clearance ends p1 begins h Buffer Vehicle from D 6 reaches left intersection where calculated in the given order a 0 for this example consider a at time zero b a gap time for 4 b 65 yellow time d c 05 red clearance time h d measured travel time from 6 right to next left intersection g h buffer time between frontage road phase and interior green usually 2 4 sec g red clearance for left frontage road e f yellow time for left frontage road Following these calculations the travel time from the setback detector to the ramp stop bar is calculated as e minus a The detector setback distance is computed as this travel time multiplied by the speed on the frontage road ramp in feet per second If desired 2 seconds can be added to the travel time Le frontage road vehicle will be 2 seconds behind the stop bar at the onset of yellow to i
60. NCY 0 00 0 12 POOR PROGRESSION 0 13 0 24 FAIR PROGRESSION 0 25 0 36 GOOD PROGRESSION 0 37 1 00 GREAT PROGRESSION ATTAINABILITY 1 00 0 99 INCREASE MIN THRU PHASE 0 99 0 70 FINE TUNING NEEDED 0 69 0 00 MAJOR CHANGES NEEDED Besides the performance summary the program generates two detailed reports providing the details of the selected timing plan Timing Plan tab and associated performance measures MOE s tab These details can be printed or exported to CSV files Click on Timing Plan tab and it will display the offset offset reference phase ring barrier structure and phase time of each phase of each intersection see figure below Recall that the offset reference phase for New Laredo intersection was Phase 2 and that for Somerset intersection was Phase 6 when we created these signals However the reference phase for New Laredo intersection has switched to compatible Phase 6 WBT WBR and that for Somerset intersection has changed to compatible Phase 2 EBT EBR in this timing plan These changes are in accord with the NTCIP offset referencing scheme which requires that offset be referenced to the phase that begins first 139 Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis SubNetwork L Input Output Summary Report Detailed Report Plot Artery List SW Military
61. PTER 7 ANALYSIS OF SIGNALIZED DIAMOND INTERCHANGES Entering Data Optimization CHAPTER 8 ARTERIAL ANALYSIS Entering Data Optimization 120 125 131 136 Signalized Arterial with TWSC Intersections143 CHAPTER 9 COMBINED ARTERIAL AND DIAMOND ANALYSIS An Example Problem REFERENCES 146 150 vi Chapter Traffic Signal Theory The gears that turn behind the scenes in your traffic optimization Software behind the tools that are used for the analysis and optimization of signals along arterial roadways Chapter 1 presents this information for signalized intersections along with a general discussion of the PASSER V 09 optimization tool The chapter concludes with a presentation of the different types of traffic analysis software and how these tools are classified based on their features The next chapter describes theory of two way stop controlled TWSC intersections N course on signal optimization should not begin without a discussion of the theory The materials for this course describe the procedural steps for using the PASSER V program to analyze intersections that e have met the Manual on Uniform Traffic Control Devices MUTCD 1 traffic signal warrants and are being signalized for the first time or e have previously been signalized and are currently operating in such a manner as to require signal timing modification or a combination of timing and geometric modifications Background Th
62. Parameters Project Info Defaults Project Name Projet Agency Name jj City Name College Station State OO Analyst JohnDoe Run No 1 gt The Project Info tab provides for entering general identification information including project name agency city name state your name as the analyst and a run number The Defaults tab provides for changing the default values for various parameters As illustrated on next page the default values are further divided into three categories identified by tabs labeled General Signal and TWSC 75 System Parameters Project Info Defaults General Signal TWSC Peak Hour Factor 1 00 Link Speed mph 30 Growth Factor 1 00 Vehicle Length ft 23 Heavy Vehicles 2 00 Scale ft pixel 5 Ideal Sat Fl Pedestrian tpephgpl 1900 Walking Speed fps 1499 Input Units English ss Output Units English od Most data items on the General tab shown above except vehicle length and scale are used to specify data values you want the program to use when creating a new intersection or link The value of vehicle length is used by PASSER V s mesoscopic simulation routine to estimate performance measures In most cases you will not need to change the default value of this variable Scale is used to specify the size of the drawing canvas Increasing this value to a maximum value of 10 ft pixel will allow you to draw a network in a 14x11 miles area In most cases it
63. SSER V and the volume you will enter later for the right turning movement Essentially we want to have our analysis account for right turning on red vehicles and ensure that we enter only the right turning volume in PASSER V that actually uses a portion of the approach phase time For each of the possible right turn geometries shown in the chart on the next page guidance is given to help you determine what right turning volume to enter in PASSER V In some cases especially where a full right turn bay or lane is present you should not enter any right turn volume into PASSER V at all and code the approach such that no right turn movement exists In other words if all right turning vehicles are handled by their own lane or as right turns on red i e they do not use signal green time to get through the intersection code PASSER V data so that there is no right turn movement for that approach Note that this rule of thumb is for isolated signalized intersections only In cases where right turn bays or channels exist along a coordinated arterial you will want to enter at least some right turning volume so that the flow or profile of right turning traffic reaches the next intersection in your network for proper delay calculations Repeat this procedure for each of the other three approaches to the intersection entering in the correct number of lanes appropriate lane movements lane widths and left turn bay presence and length When you ha
64. Split Phasing A technique known as split phasing can be used where left turn paths from opposing directions on a roadway usually on the cross street overlap within the signalized intersection Split phasing allows both movements from one cross street approach say 23 phase 3 for the left turn and phase 8 for the through movement to be fully serviced and terminate together before beginning the phases for the opposing cross street approach say phases 4 and 7 In field implementation this objective is most readily accomplished by witing one cross street approach s left turn and through signal heads to the phase 4 output and the other cross street approach s left and through signal heads to the phase 8 output Then the controller is programmed so that phase 4 and phase 8 are exclusive Le cannot be timed together Alternatively one cross street approach s signal heads left and through can be driven by controller phase 3 and the opposing approach s signal heads left and through can be driven by phase 4 In this operation the internal phase order within the controller will prevent phase 3 and phase 4 from timing simultaneously In either case i e using phases 3 and 8 or phases 3 and 4 the conflict monitor malfunction management unit is programmed so that phases 3 and 8 or phases 3 and 4 cannot be timed simultaneously Other Safety Issues A variety of other influences or specialized treatments may impact signalized operati
65. TEXAS TRANSPORTATION INSTITUTE Transportation Operations Group PASSER V W TRANSPORTATION OPERATIONS GROUP PASSER V 09 Copyright 2009 by the Texas Transportation Institute Transportation Operations Group 147 Gilchrist Bldg College Station Texas 77843 3135 Phone 979 845 9890 e Fax 979 845 9873 http ttisoftware tamu edu Disclaimer The PASSER V program was developed under contract to the Texas Department of Transportation by the Texas Transportation Institute TTI of the Texas A amp M University System It was designed for use by traffic engineers and other transportation professionals Care should be taken to ensure the program package which includes the user s documentation remains intact If the package elements become separated program effectiveness may be impaired Be advised that no warranty is made by the Texas Department of Transportation the Federal Highway Administration the Texas Transportation Institute or the Texas A amp M University System as to the accuracy completeness reliability usability or suitability of the computer program discussed herein this training guide and or their associated data and documentation The guide references to computer programs and references to other literature are provided for purposes of training only No responsibility is assumed by the above parties for the incorrect results or damages resulting from their use All signal timing parameters entered into field devices mu
66. Travel Time As with the PASSER II tool PASSER III produces both a summary report and a detailed report The summary report gives interchange level details delay phasing sequence type for each cycle length analyzed stops presence of spillback etc while the detailed report gives movement specific MOEs and signal settings for the combination of cycle length and phasing type you select from the pull down boxes at the top of the window When you analyze any subsystem including a diamond interchange note that you still have access to most of the analysis tools available from PASSER V Diamond interchanges like other subsystems of signals along the arterial can be analyzed using the GA Optimizer In this case the GA Optimizer picks what it believes is the optimal diamond phasing sequence and cycle length for the interchange This sequence and cycle length may be different than the optimal solution identified by PASSER HI Regardless of the source of the diamond solution its arterial progression bands can be seen in T Sp Diagram see next figure The 87 Volume Analysis and Delay Cycle Analysis tools remain at your disposal as well If you select the Volume Analysis Tool for an interchange you will get a throughput analysis for each diamond interchange phasing type you specify see second figure below Optimization Analysis Tools Select PASSER III GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis F SubNetw
67. Window thb BSC AEM KE All tools described in this chapter can also be used for the analysis and optimization arterials with TWSC intersections In this exercise we will only illustrate key differences between the results for an arterial with and without TWSC intersections To begin the exercise load the data set and apply the PASSER II tool to optimize timings Once the optimization process has completed select the 70 second timing plan as before and click on the T SP Diagram tool The following figure illustrates what you will see Notice that this time space diagram is essentially the same as before page 135 The only difference is the addition of a horizontal green line at the location of the TWSC intersection This green line implies that the through movements at this intersection do not stop If you compare optimization runs from the PASSER II tool for these data with the one you created earlier in this chapter you will notice that the optimal signal timings are the same but the MOE reports are different Furthermore the detailed MOE report for this modified data set has additional MOEs as illustrated on the next page for the TWSC intersection 144 Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis SubNetwork List fi Artery List j Summary Report Detailed Report Timing Source PASSER Il SW Military Time SW Military Show All Sub
68. accommodate TWSC intersections on the arterial The user should note that the following tools are only available if a system or subsystem contains at least two signalized intersections Impact on Delay Analysis Routine Platoon arrival patterns at an unsignalized intersection may change as a result of changes in signal timings at an upstream traffic signal To accommodate such changes in platoon characteristics DAR recalculates the capacities for movements at TWSC intersections in the system whenever it is invoked to assess performance measures or measures of effectiveness for a signalized arterial While HCM methodology takes into account the platoon effect of the upstream signals it fails to consider the blocking effect due to queues at the downstream signals However this limitation has been partially removed in a calculation performed by DAR which uses the following steps to accommodate TWSC intersections 1 It calculates the capacities of all TWSC intersections using the HCM methodology In the process it applies appropriate adjustments to account for upstream signals 2 It performs mesoscopic simulation of the entire system In this process it treats movements at TWSC intersections as if they are being served by permitted phases whose lengths are equal to the system cycle length In this process DAR assumes that the outflow of each TWSC intersection movement is uniformly distributed with respect to its volume and capacity During sim
69. accomplished using a transition period that is composed of the yellow change interval plus the red clearance interval Different agencies have different rules governing how these periods are computed and in some cases all red clearance times are set to zero NEMA also influences phasing behavior within controller devices that meet its standard by requiring that all phases have a yellow change interval of at least 3 seconds 2 The following equations from the Traffic Engineering Handbook T present means for computing the yellow change and red clearance times respectively y t 2a 2Gg 1 where y length of the yellow interval to the nearest 0 1 second t driver perception reaction time recommended as 1 0 second v velocity of approaching vehicle in feet second a deceleration rate recommended as 10 feet second G acceleration due to gravity 32 feet second g grade of approach decimal format 0 02 for 2 percent downhill is negative W L 2 r 7 P L 3 r y P L 4 r 7 where r length of red clearance interval to the nearest 0 1 second W width of intersection in feet measured from the near side stop line to the far edge of the conflicting traffic lane along the vehicle path P width of intersection in feet measured from the near side stop line to the far side of the farthest conflicting pedestrian crosswalk along the actual vehicle path L length of vehicle recommended as 20 feet V
70. aches resulting in yellow change intervals of 4 seconds and red clearance times that we will round up to 2 seconds The final details you need to check are the peak hour factor and heavy vehicle percentage which are found behind the Sat Flow Data tab For this example we have not yet included the effects of peak hour factor on our volumes so enter a value of 0 90 to account for some demand variability during the peak hour Also since our heavy vehicle percentage is 2 percent on all approaches which is also the PASSER V default we do not need to make any changes to truck percentage Data entry for the interchange is complete Click back to the Timing Data tab and click on Update at the bottom of the window Splits for the movements at the interchange will be calculated and displayed Keep in mind that these are splits based on one cycle length shown in the upper right corner and for only one of the three possible diamond interchange phasing sequences we have not yet optimized 124 signal settings for the interchange You can check under the Delay Cycle Analysis tab to note that the default cycle length of 90 seconds actually experiences higher delay than cycle lengths around 55 seconds The Controller tab will show you the splits for the interchange and the Signal MOHs tab will give you delay and queue lengths for all of the movements in the interchange Optimization The next step in our analysis of an isolated diamond interchan
71. am Delay Cycle Analysis Peons Input Summary Report Detailed Report Plot Artery List F B Direct US 190 Diamond Seoliees lilacs tection To Increment 489 C Show All Sub Arts eo 120 5 Link Length feet Hide All Sub Arts a 2 8 Link Speed 35 mph Artery List r Special Phasing Options Storage Lenath 441 feet SH 195 IV Basic 3 Phase Equal Frontage Ramp Phases 3 48 Intersection Width feet US 190 Diamo IV Ext 3 Phase Unequal Frontage Ramp Phases Kakanta n 10 IV TTI 4 Phase with Two External Overlaps Travel Time sec TNB Direction Link Length Link Speed j5 mph Storage Length 441 feet Intersection Width je feet Travel Time 10 sec 149 References 10 11 12 Manual on Uniform Traffic Control Devices Federal Highway Administration United States Department of Transportation 2003 Traffic Control Systems Standards Publication TS 1 1989 National Electrical Manufacturers Association 1989 revised 1994 Traffic Controller Assemblies Standards Publication TS 2 1992 National Electrical Manufacturers Association 1992 Texas Diamond Controller Specification Traffic Operations Division Texas Department of Transportation 1998 Messer C J and M S Chang Traffic Operations of Basic Actuated Traffic Control Systems at Diamond Interchanges Texas Transportation Institute Research Report 344 2F 1985 Highway Capacity Manual 2000 Transporta
72. anagement unit and detection coordination and preemption capabilities have been enhanced e Texas Diamond Controller The State of Texas has continuous frontage roads along most of its interstate and urban freeway mileage Because of this roadway feature the diamond interchange is a popular interchange treatment for junctions of grade separated facilities with major and minor arterials To cope with operating the many interchange geometries and signal orientations at these crossings the Texas Department of Transportation TxDOT developed a specification 4 for a signal controller device that was capable of operating in two of the most versatile phasing sequences common at diamond interchanges Controllers that meet the TxDOT Diamond Specification are programmed with settings for operation in the Basic three phase lag lag pattern and the TTI four phase strategy e Advanced Traffic Controller ATC Type 2070 Like the Type 170 the 2070 is a specification for a piece of electronic equipment Unlike the 170 the 2070 is an open architecture device that has expansion bays for adding processing power and memory for device functionality that can pass far beyond simple traffic control Third party software must still be purchased to run on the 2070 in order to provide traffic control functions However additional cards can be added to the 2070 to accomplish any number of objectives including ramp metering control video camera control and detectio
73. analyze the interchange Click on the PASSER III tab and underneath the Input tab change the cycle length range to 40 to 90 seconds Next review your phasing sequence options for the interchange Since we know that three phase sequences ate more appropriate for this interchange we could reasonably turn off the four phase analysis However for our example problem we will leave all phasing sequence options active Also check the details for the interchange in both the eastbound and westbound directions along the arterial Note that you can edit speed of travel if you wish that link length is broken down into its storage length between intersections and intersection spacing intersection width components and that you can edit the travel time for the interior of the interchange if desired Note that especially for shorter interchange spacing you do not want to increase the travel time as this would result in for a four phase sequence the platoon from an external approach on the arterial not having a green indication to receive them in the interior of the interchange a potential violation of driver expectancy However you can decrease the travel time if you want the interior arterial approach of the interchange to start earlier with respect to the arrival of the platoon from the external arterial approach After you review all of your input information click on Run to activate the PASSER III optimization engine A message will indicat
74. and install the program in Windows The final installation screen you see will be a screen indicating the install is complete and asking if you would like to launch PASSER V as soon as you close the installation program 73 Running PASSER V When you are ready to begin working with PASSER V click on Start Programs Passer V 09 PASSER V You will see a screenshot like the one below This is the primary PASSER V work area and it contains a primary file menu bar a file access toolbar and a PASSER V function toolbar The buttons on these toolbars provide all of the functions you need to create edit enter data for and analyze signalized isolated intersections intersections along an arterial roadway and diamond interchanges File Menu Bar File Access Toolbar PASSER V Function Toolbar File Menu Bar The PASSER V file menu bar contains the headings File View Window and Help From the File menu you can start a new PASSER V file open an existing file close the current file save the current file or save the current file under a new name With an active analysis case open in PASSER V you can also print a hardcopy of the current window or create a report on your analysis The final function accessible is the option to exit the program The View menu allows you to show snap to the grid set the grid point spacing adjust drawing speed toggle on off the view of node and link identification numbers and zoom in and out of th
75. anes and we have no other data to indicate that permitted left turns should not be allowed so select Prot Perm for these approaches In general it may be necessary to reduce the amount of right turning volume you enter into PASSER V due to right turns on red the presence of a right turn channelized island and or a right turn bay If there is a right turn channelized island or bay it does not necessarily mean that the right turn volume should not be included in PASSER V If a queue for the through movement on the approach blocks access to the right turn channel or bay then the amount of right turning traffic number of vehicles per hour blocked by the queue should be entered into PASSER V as right turning volume The low right turning and through volume on the northbound approach means that there will be ample opportunity for all right turners to clear on red The very low right turning volume on the westbound approach can easily be served as RTOR even though the through volume is relatively high you would have to observe any intersection you study to verify the number of possible right turns on red Thus these volumes were ignored by the way we assigned movements to lanes However on the southbound approach the right turn volume was sufficiently high even though the through volume was low that we did not feel that all vehicles could make RTOR so we reduced the 150 vph turning right by an estimated 75 RTOR to come up with 75 vph
76. anual steps cycle length optimization followed by offset and phase sequence optimization in a specific order It optimizes cycle length by analyzing all cycles in the defined range Synchro optimizes offsets using a multi stage process At each stage it uses a different step size depending on the optimization level selected by the user For instance if the user requests extensive offset optimization Synchro first simulates all offsets in 4 second increments followed by a search using 2 second increments Finally it performs another search using 1 second increments in the vicinity of the best offset from the second stage Unlike TRANSYT 7F Synchro s traffic model does not consider platoon dispersion As an alternate it recommends when to coordinate two adjacent signals by calculating a coordinatability factor using link distance travel time and traffic volumes as input Also unlike other programs Synchro generates optimal signal timings for each signal by averaging the analysis results of ftve volume scenarios for that signal For this purpose it assumes that a volume entered by the user is the mean and variance of the real traffic volume Poisson distribution Then it applies factors from a Normal distribution to generate four additional volume scenarios representing minus 2 10 percentile minus 1 30 percentile 1 70 percentile and 2 90 percentile standard deviations from the mean In this scheme user supplied volumes ate treat
77. are not displayed simultaneously A phase is initiated by the detection of a vehicle over the approach sensor This initial detection provides a minimum green for the movement As the detector continues to collect demand vehicles activate the detector and calls are placed to increase the green time or phase time by a given amount of time known as the passage time This process is continued until there is a sufficient gap in the demand to warrant ending the phase or the maximum green time is reached At this time this phase will terminate through yellow change and the all red clearance if there are vehicles waiting on conflicting approaches If there are no vehicles waiting on conflicting phases this phase will remain green until the time that a vehicle does pass over a detector on a conflicting phase This type of a system is heavily dependent on the detectors for operation If detectors fail it will be necessary to adjust the controller to always cause a phase to display 1 e be set to recall for at least the minimum time for each phase experiencing detection failure Speed and Travel Time When analyzing multiple signalized intersections and or interchanges it is necessary to have information not only about the length and features of roadways that join the intersection s interchange s but also about the speed of travel between the signalized junctions In the case of PASSER V such information is required when analyz
78. art from the Windows taskbar select Programs Passer V 09 folder and the PASSER V 09 program Start a new analysis file project by clicking on File New from the file menu bar or by clicking on the blank sheet icon on the PASSER V file access toolbar Then click on the System icon from the PASSER V function toolbar and enter your project agency and analyst name information Activate the background grid by selecting View from the file menu bar and clicking on Show Grid Geometry Data Entry You actually begin using PASSER V when you draw links in the main PASSER V editing window In this case you will draw an east west roadway two way about 2000 ft long PASSER V displays the length of the link you are actively creating in the lower left corner of your screen next to the axis coordinate display and then draw a north south roadway two way also 2000 ft long that intersects the east west roadway near the center of both roadways When you are finished your screen will look like the figure on the next page Notice from your computer screen that PASSER V automatically numbers the nodes of the network as you create them and that the intersection between the two roadways you drew has been automatically created as a signalized intersection along both arteries The circle or node representing this intersection is red in color to indicate that input data entry for the node i e geometric details turning movements and signal setting in
79. ase is the result of the NTCIP Offset Referencing scheme Under the NTCIP Offset Referencing scheme the through phase that starts first on the coordinated artery will be selected as the coordinate phase When a new signalized control is created the artery that contains Phase 2 and Phase 6 movement is regarded as the coordinated artery by default If NCTIP Offset Referencing option is checked then either Phase 2 or Phase 6 will be the coordinate phase depending on which phase starts see the figure on next page On the other hand if NTCIP Offset Referencing scheme is not chosen Phase 2 will always be the coordinate phase by default 135 Coordinate Phase If you wish you can check Delay vs Cycle Analysis for each of the two intersections to note that the default cycle length of 90 seconds actually experiences higher delay than cycle lengths around 55 or 60 seconds The Controller tab will show you the splits for each intersection and the Signal MOEs tab will give you delay and queue lengths for all of the movements at each intersection Optimization When you are ready to optimize the arterial close any open data entry or editing windows and click on the Tools button from the PASSER V function bar Click on SW Military from the list of arteries on the left side of the Optimization Analysis Tools window that appears Notice that you now have access to the PASSER II arterial optimization tool and the GA Op
80. asing under phasing operations where left turns are in a lead lag configuration Le either NEMA phase 1 or 5 leads while the other lags In a lead lag situation with protected permissive phasing the normal sequence of indications for the primary arterial begins with the display of the leading left protected turn and its corresponding throughtright green ball indication The protected turn then terminates through the appropriate clearance indication and becomes a permissive left turn where the leading left turners must now find gaps in the opposing through traffic stream which has just received its green ball indication 20 The yellow trap emerges when the leading left in permissive mode and its corresponding through movement are being terminated through their clearance intervals so that the opposing direction s lagging left turn can receive its protected arrow Drivers on the leading left turn approach will see that the through movement in their direction of travel is being terminated through its clearance interval at the same time the permissive green for their left turn is being terminated and may think that the through movement in the opposing direction is being terminated as well If such drivers decide to try and sneak through the intersection on the yellow they are directly in the path of opposing through vehicles that still have a green signal indication In fact the opposing through movement remains green and
81. ask is to use PASSER V to optimize traffic operations along an Entering Data Begin your analysis by starting a new file and immediately saving the blank file as Arterial PASSER V will automatically add a p51 extension when it saves the input file Then click on the PASSER V System button and enter the name of the project company analyst etc Click on OK when you are finished The next figure shows the details of two intersections along S W Military Drive in San Antonio Texas Our task is to develop a signal timing plan for both intersections that will provide for minimum delay and progressive flow between the two intersections All data currently available are shown below Before actual timings are developed for these two sites detailed drawings should be made based on actual lane width measurements and current lane usage and the intersection spacing should be measured in the field The location and type of control of the intersections between our study intersections if any should be noted If they are signalized full turning movement counts and all other data must be obtained for these intersections as well so that they can be included in the PASSER V analysis Note that information about left turn treatments type protected permissive both is not indicated in the diagrams You will have to make assumptions about the type of left turn treatments that information is not available to us for the time being but you must ha
82. ata Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal lf x BeginofGreen Cycle Length 30 AreaType Other v NICIPOlfsetReferencng 7 Bean of Yelow Offset 0 o Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs lo 100 100 100 fioo 100100 Optimization Settings Lock Sat Flows Ez Lock Green Splits a Update 79 If you are entering data for more than one artery at a time your network may resemble the one shown in the following figure Keep in mind that even though you can enter data for all of the intersections in such a network PASSER V can only analyze them on an artery by artery basis If you do choose to code multiple arteries at once you may find it helpful to click on the Subsystem button from the function bar A window dialog box similar to the one below will appear By clicking on an artery name within this artery listing you can see how PASSER V is organizing your network into arterials The selected arterial will be highlighted in red within the PASSER V main view screen if you click on Show within the Subsystem window SUBSYSTEM C SubNetwork List Artery Name Artery List Artery List Artery 1 Artery 2 Artery 3 Artery 5 Artery 6 Cancel 80 You can also use the Subsystem button menu to create groups of
83. ate Population Condition Satisfied Create Next Generation of Population Using Genetic Operators Crossover Mutation etc Genetic algorithms provide the capability of optimizing all the signal timing parameters in parallel unlike the hill climbing method which optimizes one timing parameter at a time Consequently GAs may also require more time Many studies conducted to date have shown that GA based optimization performs better than the hill climbing method A GA software or driver must be employed for applying this optimization technique PASSER V uses the GA library GAlib developed by Matthew Wall 23 because of its flexibility and availability without cost and copyright regulations Understanding the following terminology will be beneficial to the users of PASSER V Types of Genetic Algorithms There are several types of GAs The most common types ate e simple genetic algorithm and e steady state genetic algorithm A simple genetic algorithm creates an initial population by cloning the individual or population passed when it is created For each generation the algorithm creates an entirely new population of individuals by selecting pairs of individuals from the previous population 33 and mating them to produce two new offspring for the new population This process continues until the stopping criteria are met determined by the terminator A steady state genetic algorithm applies overlapping popu
84. atterns that will serve traffic demands A consistent cycle length and a continuous repetition of the same sequence of signal indications characterize pretimed operation The cycle times and phase splits are easily measured and recorded using a stopwatch If an interchange is currently operating in pretimed mode it may not have vehicle detectors that are required for actuated operation Pretimed solutions are effective where volumes follow repeatable patterns Semi actuated control is characterized by a background cycle length that as with pretimed mode can be measured using a stopwatch The difference in comparison to pretimed mode is that some phases may be skipped shortened or lengthened depending on how many vehicles are queued over the loop detectors for each phase What ate consistent from cycle to cycle are the cycle length and the fact that the main street through phases i e phases 2 and 6 will always appear in the phase sequence A uniform reference point for measuring the cycle length is the beginning of the phase 2 or phase 6 pedestrian DON T WALK indication A stopwatch measuring the time from the beginning of the phase 2 DON T WALK of one cycle to the beginning of the phase 2 DON T WALK indication of the 68 next cycle has recorded the cycle length of a semi actuated controlled intersection The beginning of the phase 2 DON T WALK is also the offset reference point The offset between the coordinat
85. b is the default HCM data used in the analysis of two way stop controlled intersections The current HCM procedure for the analysis of TWSC intersections does not explicitly account for U turns because the current state of the art technology lacks procedures for addressing U turns PASSER V provides separate fields for all applicable U turn data to allow better calibration of U turns if additional data were to become available At present the default values for U and left turns are assumed to be the same Furthermore HCM methodology is based on data collected for arterials with four or fewer lanes As such it discourages the use of its procedure to analyze unsignalized intersections on arterials with more than four lanes PASSER V provides fields where users can enter headway and follow up time data for six lane arterials if different and better data were to become available At present PASSER V assumes the default data for stx lane roads to be the same as that for four lane roads Users are encouraged to use caution when using the program for such facilities A screenshot of the TWSC tab is shown on the top of the next page 77 System Parameters Project Info Defaults General Signal TWSC Base Critcal Headway and Follow up Time for TWSC Intersections Adjustment Factors for Heavy Vehicles 1 00 2 00 2 00 0 90 1 00 1 00 Adjustment Factors for Grade 0 20 0 20 The Two way and One way buttons are us
86. better than the current best Hill climbing methods guarantee optimal solutions only when the function to be optimized is unimodal has one peak or valley For multi modal functions the hill climbing method may terminate with a sub optimal solution depending on how good the base scenario is Most implementations of hill climbing algorithms use sophisticated techniques such as a variable step size to speed up the search process Mathematical Programming Techniques Mathematical programming techniques such as linear and integer programming require a complete specification of the objective fitness function along with all the applicable constraints of the traffic model in mathematical form equations and or inequalities These techniques are based on systematic procedures programs that are designed to search a small subset of all possible scenarios in an intelligent manner Mathematical programming techniques are applicable only when a closed form mathematical model exists When applicable these techniques also guarantee the best solution Further discussion of these techniques is beyond the scope of this report Genetic Algorithms Genetic algorithms GAs belong to a class of algorithms known as evolutionary algorithms which have been developed fairly recently A GA starts with a subset of scenarios some members of a population and applies principles of natural selection mating gene mutation etc to generate a new or revised set of sce
87. cle level of service 6 based on delay v c volume to capacity ratio LOS based on v c ratio stops per vehicle average and maximum queue length vehicles and fuel consumption gallons per hour for each of the movements at your intersection Make sure that you always check the MOEs to ensure that no movement or phase is experiencing a disproportionate amount of average delay or queue length A screenshot of the Signal MOEs folder is shown on the next page for your reference Node Data Export Coord Phase gt EBT EBR Offset Reference Point gt Begin of Green C Begin of Yellow Intersections Artery 1 at Artery 2 Controller Type Pretimed Signal Cycle Length 95 Offset a Controller Id 3 Other v Sat Flow Data Area Type NTCIP Offset Referencing IV Optimization Data Performance Analysis Controller Signal MOEs Timing Data n3 4774 81 685 14 1846 70 5706 27 1846 70 1943 89 1652 31 1727 56 l h2 a 2 js s 12 277 3144 4614 3049 ai 28 2331 880 E c pd F D c cE o47 os fos 047 014 012 015 oss A aA a A A A A E oan on loss o76 oss 0o63 ose 125 1344 189 118 1372 o5 134 3 03 1560 224 121 1565 a52 140 156 32 1273 195 107 1283 044 120 4 09 In the isolated signalized intersection output illustrated above PASSER V uses the following thresholds to assign LOS classifications Level of Service Delay sec veh v c Ratio A 0 10 0 lt 0
88. clude traffic volumes roadway geometrics and any available information about current signal timing and operation Traffic Volume Information A peak hour turning movement count IMC for all intersection approaches is often the most useful intersection data for purposes of developing a signal timing plan TMC data are often supported by average daily traffic ADT counts which are 24 hour counts along the intersection approach roadways ADT counts may cover a single direction or both directions and usually cover all traffic lanes in a given direction Pedestrian counts may be performed simultaneously with TMCs or may be collected as a separate study It may also be desirable to collect truck data Turning Movement Counts TMC data are collected using a variety of techniques The most common method is to dispatch a technician to visit the site and conduct the count while in the field A variation on this method would be to have the technician videotape the intersection including portions of each approach roadway return to the office with the videotape and perform TMCs from the video The video creates a permanent record of intersection operations and can also be used to determine the current signal timing at the intersection 61 TMC data are most useful for peak periods of the day with data collected in two hour blocks that bracket the peak hour For instance a common PM peak hour would occur from 5 00 PM to 6 00 PM so a good data collec
89. conds Controllers operating in either semi actuated or fully actuated mode make more direct use of the minimum green time programmed into the controller for each phase In either of these actuated modes the minimum green time is the minimum length of time that a green indication will be displayed for each phase The duration of the minimum green is usually based on the location of the detectors that service the phase where the minimum green is adequate to serve all vehicles located between the stop bar and the detector location which is usually set back from the intersection stop bar Some controller devices also offer variable initial e variable minimum green which bases the duration of the minimum green on the amount of green time required to serve the number of vehicles that have crossed the detector before that phase becomes green When variable initial is used there is an absolute minimum green that must remain present but the minimum green time can be extended up to the maximum initial Le longest minimum green time Minimum green times are an especially important consideration at diamond interchanges because of phasing complexity and controller programming required to ensure proper and appropriate operation Some diamond interchange phasing sequences especially TTI four phase operation require that multiple phases be used to serve some or perhaps all depending on controller configuration interchange traffic movements This ob
90. ctor for area type Shared Lane Case 1 Use the iterative procedure described above to calculate saturation flow rate for each shared lane Sra as well as the saturation flow rates for each shared movement for instance the saturation flow rate for left turn movement s in that lane 2 Calculate saturation flow rate for each permitted left turn movement S permitted 28 if the movement was served by an exclusive lane 3 Calculate the adjusted saturation flow rate for the shared lane as follows S MN L adjusted S permitted A So z s Total S 4 Allocate Sagjustea using the original movement saturation flow to Sfory ratios For instance the calculation for left turn movement is S adjusted X se 1 Stora 5 Apply the left and right turn factors when appropriate Split Calculations Once the saturation flow rates have been obtained PASSER V calculates equal saturation splits for the given cycle length as follows 1 Calculates effective cycle length by subtracting the total lost time from the cycle length 38 2 Calculates volume to saturation flow ratios for each movement and determines the critical movements in each barrier 3 Allocates the effective cycle length to each critical movement using the flow ratio for the subject movement and the sum of flow ratios for all critical movement 4 Adds lost time back to each movement 5 Calculates splits for non critical phases Estimation of Delay Th
91. ctors Detectors such as inductive loops or video imaging systems communicate the status of the detector to the controller and logic within the controller determines whether to continue the phase or reduce the time allotted to that particular phase Detection systems and detector locations for the interchange can vary based on the type of phasing 5 Theory Traffic engineering theory supports the methods and procedures for all traffic engineering analysis software The Highway Capacity Software HCS for instance is a software encoded version of the major analysis procedures described in the Highway Capacity Manual HCM 6 PASSER V incorporates a range of traffic engineering optimization and queuing theories Basic elements and definitions taken from the Traffic Engineering Handbook 7 and Traffic Engineering 8 are presented here to assist in understanding the PASSER V program Saturation Flow Rate One of the most fundamental aspects of traffic engineering is the response to the question How much traffic can this road accommodate The saturation flow rate defines the amount of traffic flow that can travel past a point in vehicles per hour green vphg The ideal rate is up to the analyst and depends on local driver behavior but a rate of 1900 pcphgpl passenger cars per hour green per lane is common The ideal rate is reduced based on local conditions which are accounted for by the use of factors The overall equation i
92. d observation 5 Using revised critical headway recalculate the factor relating it to follow up time 49 Additional Factors Affecting Capacity Several geometric characteristics may significantly affect the capacity of minor movements These factors include number of legs grade median width the presence of flared approaches right turn channelization and the presence of upstream signals T intersections have higher capacity than four legged intersections because cross street drivers do not have to worry about opposing traffic Grade may also have a significant impact In general down grade increases capacity and up grade decreases capacity The presence of a median wide enough to store one or more vehicles permits cross street drivers to cross one major stream at a time This process is referred to as two stage gap acceptance The capacity of this two stage process depends on the number of vehicles that can store in the median The following illustration shows a facility with a storage space of two vehicles in the median In such a case vehicles on the higher priority movements i e eastbound left turn use the space first Any available space is used by the cross street vehicles to complete the first stage of the two stage gap acceptance It should be noted that a two way left turn lane TWLTL may provide storage space for more vehicles sal 1 as E Stage po is As shown in the next figure a flared approach increases t
93. del as PASSER II PASSER V PASSER V 09 is the latest in the PASSER series of programs developed by TTI 21 for timing arterials and signalized diamond interchanges It has a graphic user interface that is integrated with the best optimization technologies currently available Although the program focuses on the coordination of two or more signals on a linear arterial it also provides basic features to analyze and time isolated signals PASSER V can develop signal timings to maximize progression of minimize systemwide delay Its traffic simulation can analyze undersaturated and oversaturated traffic conditions along signalized arterials This section describes models used by PASSER V descriptions of tools available in the program and its limitations Optimization Algorithms in PASSER V PASSER V uses several optimization algorithms These algorithms include exhaustive search interference minimization and genetic algorithms This section describes these algorithms 30 Interference Minimization Algorithm This algorithm is a revised version of the optimization algorithm used by PASSER II Here we first describe the PASSER II implementation of the interference minimization algorithm and then describe modifications for implementation in PASSER V Like most programs PASSER II calculates preliminary splits for each signal based on Websters method Then PASSER II applies an optimization method to adjust these splits to minimize intersection
94. detection zones of traffic detectors A minimum and maximum time are set for each phase The first vehicle in the queue at the stop bar guarantees that the minimum time will be given to the phase Subsequent detections extend the phase for a given amount of time up until the maximum where green will go to the next conflicting phase that has a detector call If the maximum time is reached and no vehicles are waiting on conflicting phases green remains on the first phase Le past the maximum time until a detection on another phase occurs One critical aspect of fully actuated operation is the maintenance of detectors if the detectors do not work green time does not show on phases where vehicles are waiting and drivers become frustrated This mode is appropriate where traffic volumes and patterns are reasonably to highly vatiable where intersections are isolated ie far away from other signalized intersections or where volumes are light and quick response to a vehicle detection is desired Diamond interchanges operating in fully actuated mode also have no background cycle length In fact one of the advantages of a single controller for diamond interchanges is the ability to operate in a fully actuated mode without the need for a background cycle length Fully actuated traffic control is more adaptable to the traffic conditions that exist Actuated controllers are able to adjust phase lengths based on the traffic demand that is sensed by dete
95. displaying unusually long yellow times For more information see Traffic Engineering 8 Pedestrian Treatment The MUTCD 1 states that under normal conditions the WALK interval should be at least 7 seconds In addition the MUTCD indicates that the minimum pedestrian walking distance to be used in computing pedestrian green requirements is the curb to curb distance beyond the farthest traveled lane see distance D in the figure below The distance pedestrians must travel to cross the intersection is the main criteria for selection of a minimum pedestrian time at the intersection The figure shows the various points from which distances for pedestrian walking distance have historically been computed Bi I The 2003 MUTCD requires you to use the full walking distance across the street Once an appropriate distance is selected Equation 5 is used to compute pedestrian time Distance Gp Ped nin w 5 where Gp pedestrian time in second Pedmin minimum pedestrian WALK display varies by agency in second Distance distance measured in feet using appropriate distance D D from above figure with D being the minimum W walking speed in feet second 3 5 and 4 0 are commonly used when pedestrian speeds are lower school age elderly or handicapped pedestrians speeds should be reduced 15 The next figure relates how the pedestrian time minimums may influence signal timing where the minimum time r
96. e PASSER V network editing window The Window menu comes in handy when you ate analyzing multiple projects simultaneously in PASSER V With functions under this menu you can minimize or arrange the windows for each analysis problem or select the desired problem from the list of open files Finally the Help menu allows you to access the contents or index of the online help system access the developer s homepage which contains up to date information about PASSER V and view information about PASSER V 74 File Access Toolbar The PASSER V file access toolbar is designed to make it easy for you to close open and save files Additional buttons allow you ease of control over the way multiple projects files are arranged in the PASSER V view window By button click you can choose to have multiple open files arranged in a cascade horizontally or vertically The last item accessible under the file access toolbar is a zoom control where you control the percentage of zoom from 25 percent zoom out to 400 percent zoom in PASSER V Function Toolbar The final menu button bar is the PASSER V function toolbar When starting a new file the only accessible buttons are the Select Two way One way and System buttons The first data elements you should enter are available under the System button As illustrated below there are two tabs Project Info and Defaults within the pop up window that appears after you click on the System button System
97. e from the minimum green time for pedestrian requirements to ensure that the sum of the pedestrian walk and clearance times less the yellow change interval remains greater than the minimum green for vehicular requirements Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal x BeginotGreen Cycle Length 30 Area Type Other x NTCIP Offset Referencing p BesinofYelow offset fo Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs hi 2 gt 19 gs 113 ProtePen Prot ProtePen Prot Lead Lead Yes 6 23 3 3 1 4 0 0 0 Optimization Settings Lock Sat Flows Lock Green Splits Cancel 97 The last elements of data entry necessary for an isolated intersection are the Yellow change interval and Red Clearance interval times Recall from Chapter 1 that yellow change and all red clearance times are dependent on the speed and grade on each approach and on the intersection s width in that approach s direction of flow Assuming a speed of 45 mph on S W Military our east west street and 35 mph on S Presa our north south street we come up with yellow change and red clearance times of 4 31 and 1 35 seconds see tables in Chapter 1 for S W Military and 3 57 and 2 14 seconds on S Presa Since PASSER V cannot accept non integer inputs for the phase
98. e optimization process If the optimization objective fitness function is to maximize progression this tool treats TWSC intersections similar to the PASSER II tool That is TWSC intersections are assumed to have no effect on the progression bands However as mentioned previously use of DAR to generate MOEs does accommodate the analysis of TWSC intersections If the selected objective is to minimize delay the GA tool employs DAR to obtain delay estimates during the optimization process Volume Analysis Tool This tool assumes that demands of all TWSC intersection movements in the system can be served and none of these movements will be a bottleneck Thus only the signalized intersections are considered in the volume analysis routine To determine the maximum potential throughput of the system the throughput of the TWSC movements are added to the resulting throughput obtained from the volume analysis routine T Sp Diagram Tool This tool displays progression bands on a time space diagram T Sp diagram for the currently loaded timing plan It ignores TWSC intersections in calculating progression bands but identifies these intersections in its display by showing a horizontal green line at the location of the TWSC intersection The green line signifies the fact that the through movements have continuous greens Delay Cycle Analysis Tool This tool displays a plot of system wide delays versus cycle length Because delays are calculated using
99. e program uses the following equation for estimating control delay for all approaches where random artivals are assumed Control Delay d d d d 2 _8 E 05xCx 1 i i minh VA x lt 2 d 900x0 25x t i k 1 palie c c 0 25xc where d Be AACA 7 Ss uniform control delay in seconds vehicle incremental delay in seconds vehicle effective green in seconds cycle length in seconds capacity in vph volume in vph Estimation of Queues and Stops Queues and stops are estimated by the program using the following models Average Stops per Vehicle h oa l u me where u l y qC green split ratio g C flow ratio q s flow in vehicles per second cycle length average overflow queue in vehicles 39 Average Overflow Queue Ny a C 1 4 e 1 4 2 5 f where Q capacity in vehicles per hour n x degree of saturation q Q 0 67 sg 600 where sand g are saturation flow rate and flow period in hours assumed 0 25 Xo effective green time respectively Average vehicles in queue N qr No In the above equation ris the effective red time in seconds The maximum queue length N m is calculated as follows Estimation of Fuel Consumption The program calculates estimates of fuel consumption using the following procedure used by PASSER II and PASSER IV Aj AV AV TT F 4A AV A3V D 4z AV ARV S where F fuel consumpt
100. e that PASSER III is running and it will disappeat when the analysis is complete Click on the Summary Report button to look at yout output which is organized from lowest delay to highest delay with cycle length and phasing sequence details given As it turns out a 45 second cycle length Extended three phase sequence has the lowest delay see figure at the top of the next page To get more detailed information about the least delay solution click on Detailed Report and view the details under the Art Summary Timing Plan and MOEs tabs Note under Art Summary that the solution provides good progression and that attainability is 100 percent Also note under MOEs that there is a wide discrepancy in the delays LOS A to LOS E experienced by different intersection movements It is likely that a solution with a longer cycle length can help rectify these issues even though average intersection delay will increase 126 Optimization Analysis Tools Select is Je Artery List Harvey Show All Sub Arts Hide All Sub Arts Artery List Harvey Click on the T Sp Diagram tab to view a time space diagram for the arterial within the interchange see figure below Note the T Sp diagram shown is for the least delay solution which for our case is the cycle length of 45 seconds and an Extended three phase sequence If you want to look at the time space diagram for a different cycle length and or phase sequence you need to go back to
101. each column proportional to the values for each movement volume in that lane For instance the calculation for left turn volume in lane 2 will be 52 63 52 63 125 Movement Movement Type Left gu 38 4 Volume 562 95 1337 05 1214 27 685 73 1900 562 95 T0 Step 6 through N Repeat Steps 4 and 5 until saturation flow rates converge that is stop changing At this point add the saturation flow rates for each row to calculate the final adjusted saturation flow rate for each movement Then apply adjustment factors for turns and adjustment factors for heavy vehicles to the adjusted saturation flow rate to obtain the final saturation flow rate for each movement The reader can verify that the final matrix is as follows Movement Movement Volume Type Left on a 47 1194 85 705 15 TO The final saturation flow rates for left turn through and right turn movements are 1805 1900x0 95 4995 and 599 705 15x0 85 respectively 37 For permitted left turn phases the program applies additional adjustments based on whether the permitted movement is from a shared or exclusive lane These adjustments are described below Exclusive Lane Case 4 5v pp 13600 Vopp S permitted 1 2 5v pp 3600 x x x x x a e where v Opposite volume in vph f factor for lane width f factor for approach grade f factor for parking f factor for bus stops fa fa
102. ec EB Attain 100 0 pam WB Band 29 0 sec WB Attain 100 0 SW Mitoy 10 20 3 40 50 6 70 8 W 100 110 120 130 New Laredo Ref Phase 6 No vj 0 24 3425 00 ft Somerset Ref Phase 2 No vlo 22 Now we will use the GA Optimizer to view its performance for this same PM peak condition To start this analysis click on the GA Optimizer tab Enter the selected cycle length range of 40 to 90 seconds Also select Bandwidth based rather than Delay based as the fitness routine so the results will be more directly comparable to the results from the PASSER II tool From phase sequence options opt to optimize both phase sequence and offsets at all intersections Then click on the Run button When the optimization process has completed click on the Summary Report tab Notice that the GA Optimizer selected a 70 second cycle length in comparison to PASSER I which selected the 85 second cycle length In large measure the GA Optimizer s decision was based on a thought process similar to our own wherein we sought a timing plan that provides good but not necessarily the best progression As with the PASSER II tool you can view and or print the timing plan and MOE output details for the preferred solution and you can go to the T Sp Diagram tool to view the time space diagram for the GA based solution For compatison purposes let s go back to the GA Optimizer Input tab and this time select Delay based as our fit
103. ed Bandwidth Based l rer 1 GA Parameters Population Size 20 Num of Generations 150 Phase Sequence Do not optimize phase sequence for any signal C Optimize phase sequences for all signals Optimize according to individual signal settings Offsets C Do not optimize offset for any signal C Optimize offsets for all signals Optimize according to individual signal settings The GA Optimizer can be adjusted by changing the GA Parameters underneath the Input tab or by altering the settings for the optimizer beneath the Adv Options tab However it is recommended that you do not adjust these settings unless the GA Optimizer is having difficulty deriving an optimal solution for the artery you are analyzing If this is the case first try increasing the Num of Generations setting under the Input tab Note that it usually takes longer to run the GA Optimizer than it takes to run a PASSER II analysis The progress bar along the bottom of the Tools dialog will give you an idea of how quickly the optimizer is running your analysis As with the PASSER II tool the output for your analysis can be seen under the Summary Report and Detailed Report tabs In this case the GA Optimizer evolves to a single optimal solution so only this solution s summary statistics are viewable in the Summary Report To view a time space diagram of the GA Optimizer based solution click on the T Sp Diagram tab As illustrated in the following figure note
104. ed as 50 percentile volumes In Synchro terminology delay calculation based on this averaging method is referred to as the percentile delay method Using this method Synchro incorporates a method to model phase gapping and skipping behavior for actuated and actuated coordinated signals Synchro has by far the best user interface of all signal timing tools currently available to traffic engineering professionals It provides features to easily fine tune a timing plan Furthermore it provides for data conversion to other popular software PASSER II PASSER II 16 is a bandwidth based program for optimizing signal timings for signalized arterials Originally developed for TxDOT about 30 years ago it has been one of the most popular programs in its class The heuristic signal timing optimization model of PASSER II is based on a graphical technique and is simple efficient and powerful 17 PASSER II has passed the test of time and is known to produce good signal timing plans PASSER II can determine all four signal timing variables described earlier It selects the plan that maximizes 29 progression efficiency a unitless quantity obtained by dividing the progression band by the cycle length Because of its simplicity it is also the most computationally efficient program in its class PASSER II performs exhaustive searches over the range of cycle length provided by the user It starts by calculating splits using Webster s method Then it app
105. ed intersections is the difference in time between the start of phase 2 DON T WALER at one intersection to the start of phase 2 DON T WALER at the next intersection If DON T WALK indications are not present reference the beginning of the phase 2 yellow interval Actuated control is used at locations where traffic is less predictable and where demand can vary significantly Actuated control utilizes input from detectors and logic within the controller to adjust green times to serve demand inputs The standard eight phase controller with an actuated control strategy allows the use of phases in any sequence provided opposing movements are separated The signal controller can also omit phases if detectors indicate no demand for a particular movement This capability can benefit the competing movements and the entire intersection by reducing the time required for servicing the movements with demand The main advantage of actuated control is that the cycle length is allowed to vary to meet traffic demands Reduced cycle lengths are desirable attributes for isolated interchange control 5 Fully actuated mode operates without the constraint of a fixed cycle and can only be implemented within a single controller Phase start and duration are determined by the presence of vehicles over loop detector sensors in the pavement Internal controller logic maintains a background phase pattern called a ring structure so conflicting movements
106. ed timing plan A TSD is a scaled pictorial representation of an arterial roadway and the progress of time in relation to signal timing cycles It is usually presented in the form of an X Y graph In PASSER V the distance along the arterial is displayed on the Y axis in scaled consistent units and time on the X axis in scaled consistent units TSDs give the analyst the big picture of traffic operations and signal timing at each intersection along an arterial The slope of each line represents the speed of travel necessary to achieve the green bandwidth shown TSDs can be formulated for interchanges as well either to show the progression along frontage roads ot to show the progression that exists along the arterial roadway through the interchange The T Sp Diagram tool identifies the source of timing plan being displayed Furthermore it also provides access to the detailed report This feature is especially useful when the timing plan has been manually adjusted The user should note that the Redraw button must be clicked for the tool to display changes made to any offset and for the tool to generate the report Delay Cycle Analysis Tool This tool calculates and plots delays estimated by DAR PASSER V s mesoscopic simulation and HCM models for all cycle lengths in the user defined range For using both models it assumes that all signals operate under a common cycle length For each cycle length it first calculates green splits for all
107. ed to create a representation of your network Wherever two way or one way segments intersect PASSER V will automatically create an intersection Thus the normal procedure is to draw the primary artery in the PASSER V window and then follow by creating the appropriate number of cross streets Use the PASSER V grid to approximate the location of your cross roadways when you first draw them Notice that as you add roadway elements additional buttons become accessible on the PASSER V function toolbar see screenshot below The Select and Move buttons can be used to update correct link lengths and or intersection locations PASSER V 09 Untitled0 p5i 78 Notice by clicking on Select and then clicking the mouse pointer over a node or link that you ate presented with node or link information In the case of a node you can change the node s identification number and or its x and y coordinate information Nodes that are displayed in red indicate that the node is either missing required input data volumes signal settings etc or that there is a logical problem with the data entered for the node Le sum of minimum green times for all phases is greater than the lower cycle length bound For links you can edit link length from stop bar to stop bar the travel speed along the link intersection width and link queue storage length link length less the intersection width for each travel direction Keep in mind that the drawing
108. ee Free Free Free Free Free Free Free No No None None None lo Platoon Dispersion Modet Lock Capacity IM Lock Critical Headway I HCM Lock Follow up Time I Manar and Baass Update OK Cancel 107 We are working with the PM data for the S W Military and S Presa intersection shown in the figure reproduced below Recall that we had made some adjustments to lane assignments and right turn volumes to ensure that vehicles turning right on red are excluded from determining the splits for corresponding through phases This type of adjustment should not be made for TWSC intersections At this point change the lane assignments and volumes back to the configuration values shown below I i Bay is 91 long AM PM L 13 19 1313 7 T 52 68 l if AM PM R 74 150 13 L 24 44 Truck 2 1 T 386 635 y R 16 21 Ayy Truck 3 1 i 7 aaraa eea E S 14 S W Military 22222222221 a o eer A 12 Bay is 148 long lt 10 4 gt 13 Bay is 153 long 12 zz T E rae a ee tee 12 cl AM MS YX L 88 149 l 4 AM PM T 397 676 aff L 113 113 R 86 147 i T 85 80 Truck 3 1 i R 45 49 11 14 Truck 8 1 Bay is 126 long i l S Presa We are not done yet because we have not specified which approaches are controlled by stop signs To do this change the sign one line below the volume data line for northbound and southbound approaches from Free to
109. elay is provided both from PASSER V and for comparative purposes from the Highway Capacity Manual 6 The figure below is a screenshot of a delay versus cycle length curve from PASSER V This tool is useful for analyzing the effects of cycle length variation on delay Offsets can be kept fixed or adjusted proportionally to account for any cycle length changes you wish to analyze Note that if you adjust cycle length you may no longer have optimum offsets for the new cycle length Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis c Cycle Length Range Offsets Option Artery List From To Increment Fixed Offsets Artery 1 40 2 hos b a C ShowAll SubAtts x v C Offsets Proportional to Cycle Hide All Sub Arts Artery List Timing Source GA Based Model Artery 1 Delay vs Cycle Analysis PASSER Model HCM Isolated Delay sec veh Min Delay Cycle P5 80sec HCM 90 sec Min Delay P5 20 43 sec HCM 23 95 sec 60 60 70 80 90 100 110 120 Cycle Length 86 If your analysis includes a diamond interchange which you specify by clicking on the link representing the interior of the interchange and checking the box at the bottom of the link edit dialog box for a diamond interchange you also have access to the PASSER II optimization tool PASSER III optimizes signal settings for the two intersections of the diamond using one o
110. entof minor street at T intersection 0 0 otherwise According to Kyte et al 31 the follow up time of a movement 7 is defined as the time span between the departure of one vehicle from the minor stream and the departure of the next under a condition of continuous queuing As recommended by HCM 2000 it is estimated as follows De i OP base te uv Pav where t follow up time of minor movement 4 in seconds tr pase base follow up time from Exhibit 17 5 of HCM 2000 tr uy adjustment factor for heavy vehicles in seconds _ 0 9 for two lane major street 1 0 for four lane major street Py proportion of heavy movements for the subject movement Once conflicting flow rate critical headway and follow up time of movement 7 have been obtained potential capacity is calculated as follows 6 vette 13600 pi ci over 73600 where c potential capacity of movement 7 in vph conflicting flow rate for movement in vph lt c i Capacity Adjustment As described above potential capacity calculation is based on several assumptions and accounts for heavy vehicles grade and number of approaches or legs Estimation of actual capacity using HCM methodology requires adjustments for additional applicable factors including impedance two stage gap acceptance process upstream signals shared lane and flared minor street approaches Adjustments for these additional factors are described in this section
111. equited for vehicles is shorter than the minimum time required for pedestrians Location of yellow WALK Flashing DON T WALK p all red depends on Pedestrians 4 07 Distance W policy as to allowing Minimum Pedestrian Time gt pedestrian flashing DON T WALE to Yellow occur simultaneously All Red with vehicular Min Green Clearance clearance Vehicles Pea Minimum Vehicle Time Yelow Yellow All Red All Red Vehicular Green Clearance gt Clearance gt Signal Timing aS Flashing DON T WALK gt Minimum Pedestrian Time Controls It is important to consider that if pedestrian push buttons are not present and pedestrian activity is probable the minimum green yellow all red displayed for the through phase must be at least as long as the minimum pedestrian time G of the parallel pedestrian movement When push buttons are present the pedestrian WALK and flashing DON T WALK times entered into the controller are subject to the same minimum requirements presented and calculated in this section If computed pedestrian minimums are longer than vehicular minimums the longer of the two minimums will control and should be entered Some jurisdictions allow timing the pedestrian flashing DON T WALK interval to time concurrently with vehicular clearance times others do not Key Point If pedestrian push buttons are not present and pedestrian activity is probable t
112. ersection see screenshot below These are the optimal splits for a 90 second cycle Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal EBTEBR BeginofGreen Cycle Length 30 AreaType Other NTCIP Offset Referencing I7 ea Begin of Yellow Offset 0 Timing Data Sat Flow Data Optimization Data Performance Anapsis Controler Signal moes a lear h 7 11380 Prot Per Prot Prot ProtePert Prot Lead Yes if 4 6 29 4 4 2 2 4 4 12 M Optimization Settings Lock Sat Flows Lock Green Splits Now we want to go to the Performance Analysis tab folder to see if a 90 second cycle is really the optimal i e least delay cycle length for the PM peak at this intersection You will see a graph just like the one below Notice that minimum delay actually occurs around the 95 second cycle length rather than 90 seconds 100 Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal EBTEBR BeginofGreen Cycle Length 30 AteaType Other gt NTCIP Offset Referencing fv Begin of Yellow Offset 0S Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs Cycle Length Range From To Increment 40 s 120 5 5 J z Coded Cycle Length Cycle Length 90 sec Delay HCM 3
113. ery S W Military Note that if we were specifically creating an AM or PM peak signal timing plan and we wanted to favor either eastbound or westbound traffic we would select that direction underneath Options Next under MOE options we want PASSER V to be able to make slight adjustments in offset to try and minimize intersection delay rather than forcing the program to use only the exact offset for progression i e delay savings can be realized without affecting progression quality in most instances Finally we want to look at MOEs for each cycle length rather than just the output for the one optimal cycle length When your PASSER II tool input data entry is complete click on the Run button PASSER V will indicate that the PASSER II tool is running when it is complete output results will appear to the right in the Input Output tab Notice that the PASSER II tool selects an 85 second cycle length for our artery though the minimum delay cycle length is only around 60 seconds This behavior in PASSER II is explained by the fact that the tool is trying to maximize bandwidth or the amount of green time devoted to progression between intersections on the artery Longer cycle lengths are one means of providing increased bandwidth while only slightly penalizing overall intersection delay Click on the Summary Report tab to view the results for all of the cycle lengths analyzed The report should resemble the following figure Optimization Analysi
114. expected throughout the year Since signal timings are usually changed only once every few years it is important to account for month to month variations in your analysis If you used only your February values in your analysis and computed signal timings from 64 those volumes alone there may be large amounts of delay at your intersection in July because the signal cycle lengths and splits could not accommodate the increased volume Key Point Use your knowledge of how frequently your signal timings will be updated to frame your analysis If timings are changed infrequently you should consider applying factors for monthly variation to volumes used in your analysis Use maximum likely peak hour volumes to compute peak hour timing If volumes in your area are increasing rapidly it will be necessary to update your count data and signal timing mote regularly Roadway Geometric Information A complete understanding of roadway features is just as critical as accurate traffic volumes when the goal is signal timing generation As we have seen in the saturation flow computations in the theory section of this training guide factors ranging from mixed use lanes lanes where through and turning traffic are both present to driveway spacing from the intersection have an impact on how efficiently a lane approach or intersection can process vehicles The most appropriate way for the signal analyst to determine the presence and extent of these factors
115. f the three signal phasing schemes Basic three phase Extended three phase or four phase used in Texas for these interchanges Note that when a diamond interchange is specified along an artery the PASSER II optimization tool is not available to optimize the overall artery the GA Optimizer must be used Also note that the PASSER HI tool only appears in the tools list when the interchange which exists as a subsystem along the artery is selected as a subsystem When using the PASSER III tool see image below you specify the cycle length range and increment for your analysis as well as the type of diamond interchange phasing you want in any combination If multiple phasing options are selected PASSER II will determine which one produces the least delay for each cycle length analyzed Optimization Analysis Tools Select PASSER III GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis c Input Summary Report Detailed Report Plot Artery List hacks z ee Cycle Lenath Range EB Direction Diamond tom To Increment Link Length C Show All Sub Arts jo s i20 ES g s Hide All Sub Arts Link Speed Storage Length Artery List Special Phasing Options Artery 1 IV Basic 3 Phase Equal Frontage Ramp Phases Diamond IV Ext 3 Phase Unequal Frontage Ramp Phases IV TTI 4 Phase with Two Extemal Overlaps Travel Time Intersection Width WB Direction Link Lenath Link Speed Storage Length Intersection Width
116. fferent from those in PASSER II In addition researchers found that in some cases the interference minimization algorithm of PASSER II ends before finding the best solution This early termination is because of the heuristic nature of the algorithm and the fact that the algorithm only considers a subset of all possible solutions This discrepancy was resolved in the PASSER V implementation of the algorithm by applying the algorithm for both directions thereby increasing the ability of the algorithm to find better solutions Exhaustive Search Method PASSER V uses this search method for cycle length versus delay analysis of isolated signals for cycle length optimization in conjunction with the interference minimization algorithm and for the optimization of isolated diamond interchanges Genetic Algorithm PASSER V uses a genetic algorithm to provide new features to develop signal timings for minimizing delay or for maximizing arterial progression Because this technology is fairly new we provide a more detailed description of genetic algorithms GAs are optimization techniques based on the concepts of natural selection and genetics Genetic algorithms differ from traditional algorithms in that they work with a coding of the parameter set not the parameters themselves search from a population of points not a single point and use probabilistic rules not deterministic rules In the genetic algorithm approach the variables are represented as gene
117. fic in such a way as to reduce the speed and flow rate at which motorists are willing to drive Finally we need to enter the heavy vehicle or truck percentage for each of our approaches From our diagram plan view of the intersection our truck percentages are 1 percent for each approach Once you have entered this value for all movements you have finished data entry for this isolated intersection and your screen will look like the screenshot below 99 Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal x x Begin of Green Cycle Length 90 AreaType Other x NIGPOlectReferencng i ca a ints fo Optimization bata Perfomance Analysis Contr Signal MOE Timing Data Sat Flow Data lt 1 li 635 7 n3 100 10 10 3 3 3 3 3 a s i 1001 00 1 1 00 1001 00 1 00 fioo 1 00 1 1 1 1 1 1 li 1 1900 00 1900 00 1900 00 1900 00 1300 00 1900 00 1900 00 1900 00 1900 00 1300 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1300 00 1900 00 1900 00 1300 00 1900 00 1900 00 1900 00 1900 00 1900 00 1400 00 1400 00 1400 00 Click on the Update button at the bottom of the Node Data window and the saturation flow rates will be adjusted to account for the change in the truck percentage Then go back to the Timing Data tab folder and look at the green splits that PASSER V calculated for the int
118. first cycle second step during the initialization period of the oversaturated calculations it obtains the queues stored at the end of link wise simulation If the queue is greater than the link storage space it is adjusted to be equal to the link storage DAR has the capability to keep track of both movement wise queue storage and lane wise queue storage For the initialization period it uses the movement queue storage only 2 Actual simulation starts using the flow profile from link wise simulation together with the queue storage movement storage and lane storage from the previous step as the initial conditions 3 It updates queue storage movement storage and lane storage for all links on a second by second basis In the process it applies platoon dispersion to the back of the queue and evaluates any link blockages and lane blockages 4 It applies second by second flows to model link and movement blockages using the following steps a For each link it first updates the downstream flows For internal to external movements the available movement storage and lane storage are reduced by the amount of outflow possible For the downstream internal to internal movements the outflow is updated considering the next link s available movement and lane storage b For each direction the internal to internal and external to internal movements for a given link are updated by obtaining the available movement lane storage for the next downstream
119. flowing speeds of platoons between stop bars at successive intersections Trial runs during both off peak and peak periods should be made to determine if different average speeds occur Floating car speed studies are safely performed having two persons in the study vehicle one person to concentrate on the driving and the other person to record travel time information Usable speed information can also be obtained from a speed study performed in the middle of the block between the study intersections A variety of devices can be used to collect such data including radar guns traffic counting devices and microwave traffic detectors If no other information is available or can be collected about the average speed between intersections the posted speed limit should be recorded and used in analyses Note that speed and or travel time data are also required between the two intersections that make up any interchanges that you analyze The best methods for collecting travel time information between the stop bars of the two interchange intersections are the obsetvation stopwatch technique and the floating car technique The final speed elements that pertain to intersections and interchanges are the speeds and travel times between arterial intersections and the diamond interchange that are needed for diamond arterial coordination Additional Information In addition to the above classes of information which are primarily designed around the input inf
120. for isolated signal approaches Since Webster s formula for calculating minimum delay cycle length fails for signals near at and over capacity PASSER V calculates splits and delay for each cycle length in the desired range to determine and recommend the minimum delay cycle length for a signal In addition PASSER V uses Akcelik s models for calculating stops per vehicle average queue and maximum queue 25 Finally the program uses the PASSER II program s model for calculating fuel consumption These models are presented in the following subsections Saturation Flow Calculations For calculating saturation flow rates for each movement PASSER V defines each lane as a separate group and begins by assigning ideal saturation flow rate to each movement Then it applies adjustment factors for lane width grade parking stopping buses and area type The program also applies adjustment factors to volumes These include appropriate left and or right turn adjustment factors peak hour factor growth factor and factor for truck percentage Finally it performs iterative calculations to prorate saturation flow rate for movements serviced by shared lanes This procedure assumes that the number of vehicles in 35 each lane remains balanced In the calculation process the procedure also identifies any shared lane that is a de facto left or right turn lane The example given below describes this procedure For this example we assume values of 1 0
121. for most adjustment factors In addition we assume that all left turn phases are protected only Step 1 Create a matrix containing rows and columns identifying types of movements and number of lanes for the current approach Enter ideal saturation flow rate 1900 pephgpl and user supplied volumes 150 500 and 60 for left turn through and right turn movements in appropriate fields Enter a 1 under each lane to identify the movements permitted from that lane Our example has one exclusive left turn lane one shared left through lane one exclusive through lane and one shared through right lane Movement Movement Volume 150 1900 Step 2 Apply adjustment factors for turns that is divide volume by 0 95 for left turn and by 0 85 for right turn and adjustment factors for heavy vehicles 1 in this example to each movement volume Movement 7900 Step 3 Allocate saturation flow rate in each column equally to each movement allowed for that lane Column Operation Movement Movement 4 Volume 7500 Step 4 Allocate movement volume to each lane providing for that movement in proportion to the non zero values For instance the calculation for left turn volume in the left lane will be 36 157 89 x pea 105 26 1900 950 Movement Movement Type Volume Left 105 26 52 63 157 89 Through 125 250 125 500 Right 70597059 Step 5 Allocate saturation flow for
122. for the right turn that we entered into PASSER V Finally on the eastbound approach both the through and right volumes were high Therefore we only reduced the 147 vph right turn volume by an estimated 50 RTOR and entered 97 vph as right turn volume in PASSER V When your adjustments to right turn volume are complete your screen should look like the one below 95 Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal ai x Beginof Green Cycle Length 90 AteaType Other x NICIP Offset Referencing fv C BegmofYelow offset 0S Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs 1 2 gt 13 6 113 Prot Pen Prot ProtePen Prot Lead Lead Yes Optimization Settings Lock Sat Flows Lock Green Splits Update Signal Settings Data Entry We will now focus our attention on the signal settings parameters that we must enter into PASSER V The first selection we must make is whether each left turn phase leads precedes in the signal sequence the opposing through phase Since this is an isolated intersection Le coordination needs do not lead us to prefer one phasing sequence over another we will pick Lead for all of our approaches Leading left turns are the default left turn treatment due to their location in NEMA dual ring operation and leading left
123. formation has either not been started or is not yet complete 89 Since our current analysis is for an isolated intersection we are finished creating links and nodes for the network Next we focus on the internal signalized intersection node and the types of information we need to enter for this node Click on the Control button from the PASSER V function toolbar and then click on the internal node maf JAE ry pe mar Control Button A Node Data dialog box with data entry tabs will appear for the node see figure below and you will use the tabs within this dialog box to enter your input data and get your isolated intersection signal settings from PASSER V The example problem we will use for this analysis is the intersection of S W Military Drive and South Presa in San Antonio Texas An intersection plan view on the next page shows all available facts at our disposal concerning the signalized intersection We will be coding volumes for the PM peak hour Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id B Artery 1 at Artery 2 Pretimed Signal gt x Begin of Green Cycle Length 30 Area Type Other x NTCIP Offset Referencing fv C BesnofYelow set OS Data Entry Timing Data Sat Flow Data Optimization Data Performance Anasi Controler Signal moes Tabs 100 100 mM Optimization Settings Lock Sat Flows Lock Green Splits
124. ge is optimization PASSER V makes a number of tools available to help you analyze a diamond To access these tools click on OK to close the Node Data window and click on the Tools button from the PASSER V button bar A new window entitled Optimization Analysis Tools will appear Within the left side of this window is a listing of the arteries in your network In our case only one artery exists and it is the arterial portion of our interchange Click on Harvey and the PASSER V interchange analysis tools tabs folders will appear see figure below Note that the tools available are PASSER II for analyzing diamond interchanges GA Optimizer Volume Analysis T Sp Diagram and Delay Cycle Analysis Optimization Analysis Tools Select ie c i Input Summary Report Detailed Report Plot Cycle Lenath Range EB Direction From T Artery List Harvey neeme Link Length 720 feet o C Show All Sub Arts 40 120 2 5 Hide All Sub Arts _ Link Speed 40 mph Artery List Special Phasing Options Storage Length 684 feet Harvey IV Basic 3 Phase Equal Frontage Ramp Phases Intersection Width 38 feet IV Ext 3 Phase Unequal Frontage Ramp Phases fiz V TTI 4 Phase with Two External Overlaps Travel Time ser WB Direction Link Length 720 feet Link Speed jo mph Storage Length 598 feet Intersection Width ja feet Travel Time 12 sec Since we do not know what cycle length range is most appropriate for our interchange
125. h When you have finished your system should look similar to the screen below Orv Oy O Harvey O Harvey SH 6 South SH 6 North CX Q We next need to clean up the geometric details of the two way link that joins the two sides of the diamond together Click on the Select button from the function bar and click on the blue interior link of the interchange Again a Link Data dialog box window will appear allowing you to edit properties of the link joining the two intersections You will notice that PASSER V has automatically computed a length of roughly 700 ft for the link based on the scale at which you drew the interchange However our true stopbar to stopbar distance 121 along the arterial is 720 ft Edit the link length in each direction and make it 720 ft Then edit the speed value from the default 30 mph to the 40 mph at our interchange Check your intersection width values and enter 36 ft for the left intersection and 24 ft for the right intersection Based on default frontage road lane geometry and the speed we entered PASSER V has automatically calculated some parameters for the interior of the interchange including the actual queue storage distance in the interchange interior and the travel time from one side of the diamond to the other see figure below When you have finished reviewing these details click OK Link Data Link tink Artery Name Harvey EB SH 6 South
126. hat the 350 vehicles counted from each direction are roughly the same as the TMC numbers shown for about that same time period 5 00 to 6 00 PM This check helps validate that neither the TMC values nor the ADT values are out of scale with representative values The high count for the hour from 5 00 to 6 00 PM also shows that the hourly volume used to compute ADT numbers indicate what we already discovered that the PM peak is somewhere around the 5 00 to 6 00 PM range and it turns out from the TMC that the actual peak is from 4 45 to 5 45 PM 63 Another valuable use for ADT values is to examine the rate of traffic growth over time If we had counted ADT from the northbound NB direction in our example once every year between 1993 and 1997 we would have five data points for checking growth rates over time The table below presents hypothetical traffic counts over this time period Year NB ADT Percent increase over previous year 1993 1700 1994 1800 5 9 1995 1850 2 8 1996 1900 2 7 1997 1950 2 6 As seen in the table the volumes constantly increase though at different rates and volumes over time Between 1993 and 1994 volumes increased 5 9 percent but between 1996 and 1997 volumes increased only 2 6 percent Itis reasonable to calculate an average annual growth rate which in this case is 5 9 2 8 2 7 2 6 4 3 5 percent Thus ADT history has given us a good estimate of how much ADT is likely to increase in the next coup
127. he capacity of a shared right turn lane by allowing right turning vehicles to complete their movements while through and or left turn vehicles are occupying the same lane In this illustration the flared approach has room for one right turn vehicle 50 For a right turn movement to be qualified as channelized the right turn movement must be separated by a triangular island and has to comply with a yield or stop sign As shown below channelization of a right turn movement in this case northbound right increases the capacity of the opposite left turn movement in this case southbound left turn The HCM method accommodates channelization by removing the right turn volume from the analysis J Channelized Right Turn The presence of upstream signals will also have an impact on the operations of TWSC intersections For example if the majority of vehicles arriving from an upstream signal are in a compact platoon longer headways will be available for minor movements after the platoon has crossed the intersection The HCM method takes this phenomenon into consideration by assessing the probabilities of a TWSC intersection being blocked by any platoons from each direction In the case of one traffic signal on each side HCM uses a crude method to derive the joint effect Capacity Calculations Capacity is an essential ingredient in analyzing the performance of TWSC intersections because it is required for estimating other MOEs includ
128. he minimum green yellow all red displayed for the through phase must be at least as long as the minimum pedestrian green time G of the parallel pedestrian movement When push buttons are present the pedestrian WALK and DON T WALK times entered into the controller are subject to the same minimum requirements presented and calculated in this section of the training guide If computed pedestrian minimums are longer than vehicular minimums the longer of the two minimums will control and should be entered 16 Traffic Detection In either semi actuated or actuated modes of operation signal controllers require information about traffic approaching the intersection or interchange Devices known as detectors provide this input to the signal controller A variety of detectors are applicable but the most common is the inductive loop detector or simply loop The loop itself is two or three turns depending on loop length and environment of wire placed in a sawcut in the pavement along the approach to the intersection interchange Wire leaders connect the loop to an amplifier which is then connected to the controller Loops and or loop systems can be designed to cover multiple approach lanes Depending on approach speed single or multiple loops may be used within a lane Detectors can be operated in either presence or pulse mode In presence mode the amplifier sends the controller a call at all times when a vehicle
129. ies are available and include Volume Analysis T Sp Diagram and Delay Cycle Analysis see figure below Optimization Analysis Tools Select GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis SubNetwor Input Adv Options Summary Report Detailed Report Artery List SH 195 Increment From To Show All SubsArts so 120 5 amp Hide All Sub Arts Cycle Length Range M Fitness Routine Artery List C Delay Based Bandwidth Based MGA Parameters Population Size 20 Num of Generations 150 Phase Sequence C Do not optimize phase sequence for any signal Optimize phase sequences for all signals Optimize according to individual signal settings Diamond Phase Sequence C Do not optimize phase sequence for any diamonds Optimize phase sequences for all diamonds Optimize according to individual diamond settings Offsets C Do not optimize offset for any signal Optimize offsets for all signals Optimize according to individual signal settings 148 To analysis the diamond interchange expand the artery list select the subsystem US 190 Diamond and your Optimization Analysis Tools window should look the same as following figure Since this subsystem contains only a single diamond interchange you can perform similar analysis as described previously for diamond interchange Optimization Analysis Tools Select PASSER Ill GA Optimizer Volume Analysis T Sp Diagr
130. in vph of a Rank 3 movement k Cn S Cink Ck I Po j where j Rank 2 movements The impedance effect of Rank 4 movements on the other hand is more complicated Rank 4 movements are impeded by main street left turn and opposite cross street traffic However the probability that cross street through traffic will be queue free depends on the main street left turn movement To account for these interdependencies an adjustment factor p which is the adjustment to the main street left and minor street through impedance factots is introduced as follows p 0 65p 0 6 p p 3 54 where p adjustment to the main street left and minor street through impedance factor P Po j Pox Po j Probability that the conflicting main street left turn movement is in a queue free state Pox probability that the conflicting minor street through movement amp is in a queue free state Then the following equation is employed for estimating the movement capacity in vph of a Rank 4 movement Cn p p Xp where j conflicting Rank 2 minor street right turn movement In addition to vehicle induced impedance pedestrians crossing the streets will also obstruct conflicting traffic streams to and from the minor streets If there are a significant number of pedestrians capacity calculations should account for resulting impedance Achieving this result requires the determination of the probability that a conflicti
131. individual diamond settings Offsets C Do not optimize offset for any signal g Optimize according to individual signal settings A progress bar at the bottom of the window indicates the optimization progress When the optimization is complete go to the Summary Report tab to see which cycle length and phase sequence the GA Optimizer selected In our case the optimal solution was the same 45 second cycle identified by PASSER II Note however that the GA Optimizer and 128 PASSER HI may not always identify the same cycle length and or phase sequence as optimal Under the Detailed Report you can find such details as the timing plan for the 45 second Extended three phase solution and the MOEs including delay for each movement for the interchange If you wish you can choose to print any or all of these details by using the Print button in the upper right of the window within each tab folder Also you can view a time space diagram based on the GA solution by clicking on the T Sp Diagram tab GA Optimizer will be identified as the source of the time space diagram being displayed Congratulations You have now completed your first optimization analysis of a diamond interchange For practice you may want to go back and experiment with some of the features of the PASSER III or GA Optimizer tools and see what kinds of effects your changes make on the optimal solution For instance a 40 second cycle length is rather short for peak pe
132. ing an artery with an adjacent intersection when analyzing an interchange or when analyzing these elements in combination Note that for purposes of coordinating intersections rather than documenting system performance which is mentioned later the speed information 69 collected should be based only on driving that occurs at a driver s average chosen speed in traffic 1 e it should not include delay at signals The simplest technique for collecting speed data along the artery is to simply select vehicles in the traffic stream at the site and using a stopwatch time how long it takes each vehicle to travel from stop bar to stop bar at successive intersections Combining this time information with knowledge of the distance between the intersections allows you to easily compute the speed Of course this technique requires that line of sight exists from a safe vantage point to the same direction stop bars at adjacent intersections The average speed between intersections can also be obtained using the floating car technique though this requires much more data collection planning than the observation method The floating car study is based on the average speed found to exist between two points by traveling within or following platoons of vehicles The average speed is estimated from five to ten trial runs during off peak traffic volume conditions and five to ten trial runs during each peak period condition The speeds obtained should be free
133. ing control delay and queue lengths HCM gap acceptance model used by PASSER V for calculating capacity is summarized below Step 1 Calculate potential capacity of each movement assuming that each movement is serviced by an exclusive lane Step 2 Adjust potential capacity for effects due to impedance two stage gap acceptance process and upstream signals 51 Step 3 Calculate movement capacity using an iterative method This iterative method is similar to the saturation flow calculation method for signalized intersections described in the previous chapter Step 4 Adjust movement capacity for flared minor street approaches Potential Capacity Potential capacity is defined assuming that e the TWSC intersection is not blocked by the major street traffic e each minor stream movement is serviced by an exclusive lane e traffic on major street arrives randomly and e no other movement of Rank 2 3 or 4 impede the subject movement In other words potential capacity defines the potential traffic volume that can depart from the stop line for a minor stream Calculation of potential capacity of a movement requires total conflicting flow rate critical headway and follow up time for the subject movement Traffic on a lower priority movement must yield to all traffic on conflicting movements with higher priority Thus its potential capacity is constrained by all higher priority conflicting volume Generally the impact of each highe
134. ion in gallons per hour TT total travel in vehicles miles per hour D total delay in vehicle hours per hour S total stops per hour V cruise speed A model coefficients given below 0 075283 1 5892E 3 1 50655E 5 A 0 73239 0 0 0 0 0 0 0 0 6 14112E 6 40 Volume Analysis Routine This routine uses the following analytical model developed by Chaudhary et al 26 Maximize V Subject to ee ee VieE C Pi veii mgt Viel C Pi where C cycle length in seconds V hourly flow rate demand for the system in vph gi effective green time of movement per cycle in seconds si hourly saturation flow of movement in vph E set of exterior movements I set of interior movements Pi ratio of volume for approach to sum of exterior volumes This model calculates the maximum number of vehicles per hour that can go through the facility until some movement becomes a bottleneck It assumes that an exterior movement becomes a bottleneck when its service volume becomes equal to its capacity For all interior movements the bottleneck capacity is equal to a user specified percentage or fraction of its capacity The default value for internal movements is 95 percent of capacity Furthermore this model assumes that the origin destination pattern does not change over the analysis period Also it does not consider the affects of blocking Thus it is especially suitable for analyzing four phase diamonds with U turn lanes
135. is by visiting the site and taking the time to observe the operation of the intersection s interchange All significant roadway and intersection details should be noted including but not limited to these common items e lane use by lane for all approaches and departures e lane widths by lane for all approaches e roadway names e number of lanes for each approach and departure e type of intersection approach lane striping for all approaches e lengths of turn bays along each approach and departure e turning radii within the intersection field approximation e presence and location of stop bars e presence location and size of protective islands e presence location and type of signal heads and pedestrian push buttons e presence and location of pedestrian crosswalks e pedestrian walking distances see Chapter 1 e north arrow e adjacent land use e presence and location of roadside angled or parallel parking e distance to nearest driveway upstream from the intersection on each approach and departure and 65 e spacing between intersections measured from the stop bar of the upstream intersection to the stop bar of the downstream intersection along a given roadway link if the analysis is not for an isolated intersection Some factors may not apply to all cases but the above list includes most significant roadway details The figures below ate examples that show most necessary information for an intersection and
136. is section presents uniform terminology that will be used throughout the course Discussion items include intersection geometries signal phasing nomenclature controller types and modes of intersection control operation Intersection Configuration The following figures represent very typical geometric configurations of signalized intersections l I I l l I I l l I I l l i Three leg Intersection Four leg Intersection l Diamond Interchange NEMA Phasing The National Electrical Manufacturers Association NEMA defined a method for organizing phases in a dual ring structure as part of its Traffic Signal 1 TS1 standard 2 The phasing reference scheme and the dual ring concept are extremely versatile and powerful methods for depicting intersection phasing structure Both are shown in the figure below 4 E iji E NEMA Dual Ring Phasing 6 ne Ce 2 gt leading lefts no overlap 7 Main Street A tl Cross Main Street Lead Lag Street Cross Street Split Phased The NEMA phase reference system can be extended for use at a diamond interchange The figure below shows a common representation of the phases at a diamond interchange using a NEMA like phase numbering and reference system 63 4 NJ Crossing Arterial Overlap A 01 62 6 5 J p2 Overlap B 5 06 gt 8 X NEMA Phase Left Side Right Side Frontage Ramp Frontage Ramp The number of intervals and the se
137. istency For example is the sum of the minimum phase times greater than the lower bound of your cycle length range Remove inconsistencies or coding errors until all nodes are ready for analysis Le light blue in color When you ate ready to perform your optimization on yout artery arteries click on the Tools button of the PASSER V function toolbar From the dialog box that appears you can select the first arterial or subsystem from the list on the left side and then choose the type of analytical tool you want to use If you are analyzing an arterial or subsystem where no sections ate specified as a Texas diamond interchange the tools available include PASSER IL GA Optimizer Volume Analysis T Sp Diagram and Delay Cycle Length Analysis see next figure 81 Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis c Input Output Summary Report Detailed Report Plot Artery List Artery 1 Cycle Lenath Range Output Show All SubArts To Increment Cycle Lenght fq Hide All Sub Arts gt 5 L Total Interference s po aoi Options Efficiency 0 ren Full Band in EB Direction Attainability 0 Full Band in WB Direction EB Band WB Band Volume Weighted Band Band s gt MOE Options Efficiency fj fj Fine tune Offsets toMinimized Delay Attainabiiy z P P IV Output MOEs for Each Cycle Avg Delay s v fo Bun If
138. jective is accomplished by using overlaps which allow multiple phases to cause a green indication to be displayed for the traffic movement for which the overlap is programmed The overlap will maintain a green indication for a movement during the green time yellow change interval and red clearance interval of the first serviced phase within the overlap if the following phase s are also programmed to be a part of the overlap However in all diamond phasing sequences that operate in semi actuated and or fully actuated mode it may be possible to skip any or all of the phases programmed to be part of the overlap Thus it is critical that each phase whether it is part of an overlap or not be programmed with adequate minimum green time yellow change interval and red clearance interval for the appropriate approach and movement In practical implementation this requirement may mean that the interchange s cycle length must be lengthened slightly to provide all 12 phases with minimum green and clearance times When implementing output and timing recommendations from the PASSER V 09 program it may be necessary to extend certain phase times and thus the cycle length to ensure that all phases whether they compose an overlap or not are provided with adequate and appropriate minimum green times yellow change intervals and red clearance intervals Yellow Change Red Clearance Phase termination before the start of a conflicting phase is always
139. lan to another during some pedestrian service calls and during preemption As the controllers attempt to regain coordination shorter or longer phase times may be displayed for some phases causing driver expectancy issues This type of driver expectancy is mainly an issue for closely spaced intersections where signal heads may have visibility issues and drivers expect a certain operation timing This effect is intensified and some additional controller limitations may impact operations when the dual controllers managing the diamond are coordinated with other intersection and or interchange controllers All transition and coordination impacts must be thoroughly investigated by the traffic engineer developing the plan If the features of the timing plan that is to be implemented are significantly different than other interchanges in the area of the study interchange serious thought should be given to all driver expectancy issues It is significant to note that the closer the interchange spacing the greater the driver expectation of green in the interchange interior If driver expectation issues cannot be avoided temporary signing should be displayed to indicate that signal operation at the interchange has been altered As with all signal timing plan development issues examination of driver expectation issues and if necessary countermeasures must be studied in depth approved and implemented by licensed civil traffic engineers and staff
140. lations with a user specifiable amount of overlap The algorithm creates a population of individuals by cloning the chromosomes of the previous population For each generation the algorithm creates a temporary population of individuals adds these to the previous population and then removes the worst individuals in order to return the population to its original size The amount of overlap between generations is selected by specifying the Replacement parameter This is the percentage of the population that should be replaced each generation Newly generated offspring are added to the population and then the worst individuals are destroyed so the new offspring may or may not make it into the population depending on whether they are better than the worst in the current population Elitism Elitism applies only to a simple GA Elitism means that the best individual from each generation is always carried over to the next generation Selection Scheme The selection method determines how individuals are chosen for mating If one uses a selection method that picks only the best individual then the population will quickly converge to that individual So the method should be biased toward better individuals but it should also pick some offspring that are not quite as good overall but may have good characteristics Some of the more common selection methods include roulette wheel selection the likelihood of picking an individual is proportional to the indi
141. le of years A useful equation for computing compounded growth rates i e those that grow and build upon each other from year to year and future volumes is Future Volume Present Volume x 1 00 r where r annual growth rate i e 5 percent is expressed as 0 05 n number of years for the traffic projection This equation should only be used when historical traffic volumes have shown a consistent compounding increase over time i e an exponential increase It is important to note that a traffic count TMC or ADT is never an exact count Not only are errors often made in counting and recording but traffic volumes themselves are never consistent from day to day week to week or month to month For instance even in an area where traffic is not growing a count performed in February will be much different than a count performed in July If historical records of traffic volumes and month to month average ADT variations are available this information can be used to calibrate your recent count information to account for month to month variations For instance historic monthly records may show that counts in your area tend to be 1 15 times higher in July than in February and July tends to be the busiest month of the year To cover all cases i e use the maximum reasonable traffic volumes in your signal timing analysis you would multiply yout TMC and ADT counts by 1 15 to produce a reasonable estimate of the highest volumes
142. lies a hill climbing approach and adjusts splits to minimize delay Finally it applies its bandwidth optimization algorithm using the pre calculated splits as input to that model At the optimization stage it can find the cycle length offsets and phase sequences that produce maximum two way progression PASSER IlI PASSER III 18 is a delay based program for optimizing timings at diamond interchanges especially those using a single controller It can also coordinate a series of diamond interchanges along one way frontage roads For a single interchange PASSER HI uses an exhaustive optimization method It evaluates each timing plan cycle split and phasing using delay difference of offset method 19 and selects the plan that produces the least interior delay For a series of interchanges PASSER III saves four least delay timing plans for each interchange Then it applies PASSER IP s interference minimization algorithm and selects plans for interchanges that result in maximum progression along the two frontage roads PASSER III produces accurate results for undersaturated traffic conditions and can be applied to diamonds with or without U turn lanes using three or four phase signal operations PASSER IV PASSER IV 20 is a program for maximizing arterial progression in arterial and multi arterial signal networks This program uses a mathematical programming technique for optimizing signal timings It uses the same macroscopic delay mo
143. links at each time step These models can accurately account for the behavior of queued traffic and traffic flow interactions between adjacent links and they are better suited for all types of traffic conditions in signal systems The down side is that they are also more intense from a computational point of view Also the accuracy of these models may depend on the number of cycles simulated Macroscopic Traffic Models Models in this category simulate the cycle by cycle behavior of platoons of traffic at each link in the system and are deterministic in nature These models may or may not account for platoon dispersion Macroscopic models treat a queue of vehicles at an approach as an upward stack Thus they are accurate only for undersaturated flow conditions Because of 25 their simplistic nature mactoscopic models are the most efficient from a computational point of view Optimization Models and Search Algorithms As mentioned earlier traffic models simulate a given set of traffic and control conditions In other words they are able to tell only how good or bad a given scenario is Optimization and search algorithms are techniques that systematically generate scenarios compare their fitness or objective function value Le delay bandwidth efficiency throughput etc obtained by using a simulation or analytic model and select the best scenario based on a predetermined criterion For instance if delay minimization is the desired objec
144. lit The available left turning movement types have the following meanings e Perm Permitted left turns under a green ball indication left turners must find acceptable gaps in the opposing traffic stream before making their maneuver 94 e Prot Protected or exclusive left turn movement under a green arrow indication Left turning vehicles have right of way to make their maneuver e Prot Perm Combination of permitted and protected left turns a portion of this approach s left turn time is under a protected green arrow and a portion is available for permitted maneuvers If lagging protected left turn phases are used be aware of the possibility of a yellow trap see Chapter 1 emerging e Split Split Phased operation wherein the left and through movements for an approach are programmed into the controller to begin and end simultaneously Usually used for minor street approaches where left turn paths from opposing directions overlap For our example isolated intersection the eastbound and westbound approach s left turning vehicles have to cross three opposing lanes of traffic As it is usual engineering practice to only allow permitted left turns across no more than two opposing lanes the eastbound and westbound approaches are protected only Click in the data entry field for each of these movements and select Prot Our northbound and southbound approaches left turners only face two opposing through l
145. lits and phasing being optimized However the users should also note that larger values for population size and number of generations will in general also cause the optimization process to take longer to complete Lastly the users should note the fact that the bandwidth progression optimization capability provided by the PASSER II tool is more efficient and produces better results than the bandwidth optimization feature of this 44 tool Thus we recommend the GA based tool for progression bandwidth optimization only when coordination of a diamond interchange with adjacent signals on the arterial is desired Volume Analysis Tool This tool uses the volume analysis routine described earlier to provide an analysis of cycle length versus ideal throughput capacity of a facility at the point one or more movements reach maximum capacity It provides results in graphic as well as tabular form and identifies the bottleneck movement s This tool can be used for isolated diamonds arterials or a combination of the two The reader should recall that interior distance travel time of a diamond interchange is explicitly taken into account for calculating the timings for a four phase diamond Hence for isolated diamonds this tool provides a planning option that can be selected to investigate the distance versus capacity issue of the TTI four phase strategy T Sp Diagram Tool This tool displays a time space diagram ISD for the currently load
146. llowing subsections Traffic Signal Analysis Models A traffic model takes traffic volumes geometric information for the facility and a complete description of a traffic control plan as input Then it evaluates or simulates the described 24 scenario and outputs measures of effectiveness MOEs Typical MOHs include average or total delay number of stops fuel consumption bandwidth efficiency average or maximum queues etc Most models provide an estimate of several if not all MOEs One method of model classification is the primary MOE estimated by the model The two main types of models are delay based and bandwidth based Furthermore the level of detail or abstraction used by a model is another classification method Thete are three common types of traffic models based on the latter classification microscopic mesoscopic and macroscopic Microscopic Traffic Models Microscopic traffic models provide the most detailed analysis by simulating the behavior acceleration deceleration car following lane changing etc of individual vehicles in the traffic stream In general these models are also stochastic in nature and rely on a random number generator that uses a seed value to generate values of various parameters during simulation To obtain another sample the user must change the seed value and re run the simulation Running the simulation with different random number seeds is equivalent to collecting random samples of data simila
147. mprove operational efficiency and reduce cycle 19 length The next figure shows a typical detector setup for a diamond interchange operating under four phase control If detector setback is compromised for practical considerations it is important to have good gap settings and maximum times for efficiency See previous Table for details See previous Figure for details L 6 ft x 40 ft Stop bar Ol 6 ft x 6 ft Setback loop s See previous Figure for details See previous Table for details Both the three phase and four phase detector placement procedures must be calibrated based on site specific features An additional feature of TxDOT s detector plans for interchanges is that the 6 ft X 40 ft stop bar loops may be turned off after 1 the approach receives a green indication 2 the detectors experience a gap of usually greater than 0 5 second and 3 the other intersection s arterial phase has a detection This procedure known as detector switching effectively uses the stop bar detectors to call the phase and then turns them off so that the setback detectors can efficiently extend the phase Without detector switching the stop bar detectors unnecessarily extend the green beyond the time required to clear the platoon Safety Considerations Yellow Trap An important safety issue emerges for agencies considering using protected permissive also known as exclusive permissive ph
148. n changeable message sign control etc The 2070 is also envisioned as a device that will make use of the National Transportation Communications for ITS Protocol NTCIP so that devices from multiple manufacturers are 100 percent compatible with one another in terms of device to device communications Modes of Control All modern controllers are capable of operating in one of three modes pretimed semi actuated or fully actuated Choice of mode is dependent on a variety of considerations including availability of communications infrastructure wireline or wireless traffic flow characteristics at the site intersection spacing detector placement and detector maintenance Discussed below are descriptions conditions for application and common examples of each mode of controller operation e Pretimed As the name implies pretimed control is fixed in terms of cycle length and phase split Once programmed the same order and duration of phase indications will occur at the intersection until the controller settings are manually reprogrammed or another set of fixed duration settings is selected by time of day or day of week month year This mode is not traffic responsive i e uses no detectors but it can be used in coordinated systems along arterials or in network systems One common system application of pretimed controllers is in a closed loop system that does not use detectors Pretimed operation tends to be most effective where there is lit
149. n vehicles on cross street movements may face one of the following four conditions 56 e no platoon e platoon from the left side only e platoon from the right side only or e platoon from both sides The joint platoon arrival patterns created by two traffic signals may be extremely complicated depending on a number of factors For simplicity HCM methodology incorporates the effects of each upstream signal separately and then applies additional adjustment to arrive at the total proportion of blocked and unblocked times PASSER V implements the platoon dispersion model suggested by HCM 2000 6 and related corrections and changes 30 In this model dispersion is a function of speed and distance where dispersion continues to increase with distance The dispersion factor in this model is independent of traffic volumes According to Baass and Lefebvre 32 however the amount of dispersion varies with volume They observed that platoon dispersion initially increases with volume and starts to reduce when volume reaches 60 80 percent of link capacity and becomes zero as volume approaches link capacity To account for this phenomenon Manar and Baass 33 proposed a modified platoon dispersion model This additional platoon dispersion model is also included in PASSER V Because platoon dispersion over long distances produces traffic flow patterns similar to random arrivals PASSER V ignores the effect of traffic signals located more than 1 mile
150. narios called the next generation A GA based optimization model uses a specified traffic simulation model to evaluate the fitness of each member i e a signal timing scenario in the current population Then it generates a new population by combining the characteristics of that is by mating selected pairs of scenarios members The principles of natural selection ensure that the characteristics of the fittest members i e those with higher bandwidths or lowest delays depending on the objective of optimization have a high probability of transmission to the next generation A GA terminates when either no more improvements occur or a certain number of user specified generations are complete whichever occurs first GAs are different from all previously described search algorithms in that they utilize codings of variables rather than the values of variables Given a large enough population and sufficient number of generations a GA can provide the global optimum because GAs perform simultaneous optimization of all selected variables Furthermore GAs can be applied to all types of optimization problems even those that cannot be described in closed forms Their effecttveness depends on the scheme used for coding the variables and the details of the natural selection process used Conceptually an exhaustive optimization algorithm is a GA that uses all members of a population and it applies only the initial generation of the optimization algorithm
151. nd Total PM Left Thru Right Left Thru Right Left Thru Right Left Thru Right 4 30 4 45 10 35 10 10 35 10 10 35 10 10 35 10 220 4 45 5 00 15 40 15 15 40 15 15 40 15 15 40 15 280 5 00 5 15 20 45 25 20 45 25 20 45 25 20 45 25 360 5 15 5 30 20 50 25 20 50 25 20 50 25 20 50 25 380 5 30 5 45 20 55 25 20 55 25 20 55 25 20 55 25 400 5 45 6 00 10 40 15 10 40 15 10 40 15 10 40 15 260 6 00 6 15 5 20 5 5 20 5 5 20 5 5 20 5 120 6 15 6 30 25 30 5 25 30 5 25 30 5 25 30 5 240 Peak Hour Total 75 190 90 75 190 90 75 190 90 75 190 90 1420 4 45 5 45 An important item to consider when performing your analysis is how volumes peak within the day and within the peak hour itself Notice that the total intersection volume between 5 30 and 5 45 PM is 400 vehicles However the volume between 4 45 and 5 00 PM is only 280 vehicles Both values are within the peak hour but there is a sizeable difference between them The average 15 minute volume is 280 360 380 400 4 355 vehicles per 15 minutes An indicator known as the peak hour factor PHF is computed as the peak hour counted volume divided by four times the highest 15 minute volume Thus 1420 400 x 4 0 8875 PHF can range from 0 25 to 1 00 A PHF of 0 25 would indicate that all of the hourly traffic occurs within a single 15 minute period very unlikely A PHF of 1 0 would indicate the hourly traffic was evenly distributed among the four 15 minute periods also unlikely The smaller the PHF is the more likel
152. nd the PASSER II and GA Optimizer tools for arterials include the Volume Analysis T Sp Diagram and Delay Cycle Analysis tools The Volume Analysis tool shown below allows you to graph throughput for each cycle length you wish to analyze you specify the range and shows the current volume entered for the artery Optimization Analysis Tools Select PASSER II GA Optimizer V T Sp Diagram Delay Cycle Analysis J i me Artery List Input Information Chart Critical Movment Total Throughput Artery 1 Show All Sub Arts Hide All Sub Atts Throughput Analysis Artery List L i i Existing Volumes i H Max Throughput Artery 1 r 5 70 80 0 Cycle Length 84 In the previous figure the horizontal line displays the total volume that wishes to enter the facility The throughput versus cycle length curve identifies the capacity of the facility at the point where at least one movement becomes a bottleneck The junction of the throughput curve and the horizontal line identifies the minimum cycle length required to service all demand sum of volume at external approaches to the facility Note that the Volume Analysis tool does not incorporate the effects of progression and queuing between intersections Therefore these results should be used with caution The Volume Analysis tool s Critical Movement tab see screenshot below can be used to identify which intersection movement and at which intersecti
153. ness routine After the run is complete click the Summary Report tab Notice that the GA Optimizer selected a 55 second cycle length when the optimization criterion was changed from bandwidth to delay Again this is not surprising since a 55 second cycle length is very close to the minimum delay cycle lengths for each of the two separate intersections Note however by observing the time space diagram that delay based optimization has resulted in much less bandwidth and bandwidth efficiency than when maximizing bandwidth was the criterion Comparing to PASSER II the GA Optimizer is a more robust optimizer as it can be applied to different artery systems including diamond interchange See Chapters 7 and 9 However limitation on the GA tool at this point is that you cannot fix a subset of phasing sequences 142 for optimization as you can with the PASSER II tool by making selections under each intersection s Optimization Data tab Instead you can only ask the program to either search for ALL possible phasing sequences or use the existing phasing sequence To review some of the details of the output for this timing solution rerun the GA Optimizer to restore the bandwidth based solution and click on OW to close the Optimization Analysis Tools window Then go to each intersection and review its timing details Click on the Control button from the function bar and then click on the right intersection New Laredo Highway S W Militar
154. ng names Intersection Data Capacity Data Headway Data and MOEs 106 As you learned in the previous chapter the next step is to enter the lane assignment data Instead of doing that from scratch we will start this exercise with the isolated signalized intersection data we created in the previous chapter sample data file named isolate p5r supplied with the program also contains the same data To begin open this data file and proceed with steps 2 and 3 In Step 3 you will click the node displayed as This action will take you to the following screen Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal v EBTEBR x BeginofGreen Cycle Length 30 Area Type Other gt NTCIP Offset Referencing fo BegnofYelow oie 0 Timing Data Sat Flow Data Optimization Data Perfomance Analysis Controler Signal MOEs hi 2 gt 19 le Prot Perr Prat Yes 4 6 4 2 4 e9 3889 1333 4 23 4 2 4 38 a a h2 Optimization Settings Lock Sat Flows Lock Green Splits ki Now change the controller type to Unsignalized TWSC You will see the following screen Node Data Export Intersections Controller Type Controller Id 3 Artery 1 at Artery J Unsignalzed TWSC v Area Type Other v Intersection Data Capacity Data Headway Data MOEs ast 3 li 2 gt fet ft 2 ao a 6s l3 se na jao Fr
155. ng pedestrian stream does not block the subject movement This probability is expressed as v w s Py 1 P 3600 where Ppx pedestrian impedance factor for pedestrian movement x vi hourly number of pedestrians of movement 7 w lane width in feet Sp pedestrian walking speed in feet per second fps Taking into account pedestrian impedance the adjustment for Rank 3 movements becomes II Po TP x gt where pedestrian movement x is the conflicting pedestrian movement of the J X subject movement Similarly the adjustment for Rank 4 movements becomes pp Ju Pra Two Stage Gap Acceptance The capacity calculation for a cross street movement given above assumes single stage gap acceptance When median storage is available minor street left turn and through movements may cross the TWSC intersection in two distinct stages by crossing one major stream at a time In such a case capacity will be calculated for each stage separately by taking into account conflicting flow for each stage as described below 1 Conflicting flow for Stage 1 is the main street traffic from the left side 55 2 Conflicting flow for Stage 2 of cross street through traffic is the main street traffic from the right side 3 Conflicting flow for Stage 2 of cross street left turns is the main street traffic from the right side and the opposing through and right turn Note that not all vehicles of the subject movement will cross the intersection
156. ng process however we will use the isolated signalized intersection data set that we created and saved in the previous chapter Furthermore we will assume that the northbound and southbound approaches are controlled by stop signs N ow that you have worked through an example problem for an isolated signalized Data Entry and Analysis To create a TWSC intersection and enter all data needed to analyze such an intersection you will follow the same procedure as for a signalized intersection Recall from the previous chapter that this procedure consists of the following three steps 1 drawing an intersection 2 clicking the B Control button and 3 clicking on the internal node E interest In this case however you will change the controller type from its default value of Pretimed Signal to Unsignalized TWSC as illustrated in the next figure It should be noted that you can change the controller type at any time to convert an existing signalized intersection to a TWSC intersection or vice versa 105 Node Data Controller This action will display the following screen which has a similar format to the data screen for a signalized intersection Node Data Note the following differences between the two screens Signal timing related data from the top part has disappeared the bottom part has been changed and the six data access tabs in the middle have been replaced by four new tabs with the followi
157. ng the arterial and frontage road ramp approaches is dependent on speed Stop bar detection is made possible by a 6 ft x 40 ft loop in each lane Stop bar detection is augmented by advance detectors usually measuring 6 ft X 6 ft placed in each lane The advance detection placement is dictated by the speed of approaching traffic The table on the next page shows setback distances that have been computed for use with 6 ft X 40 ft stop bar loops under a three phase control strategy The figure below the table illustrates the resulting detector layout for a diamond interchange under three phase control 17 See Table OO above for See Table g g details above for ini details OO Phase Calling PREN ge aa id O EAE eee eee l al See Table aun s3 C 6 ft x 40 ft Stop bar a E 3 above for above for O 6 ft x 6 ft Setback loop s details details 0 00 Four Phase Control The detector placement is slightly different for diamond interchanges under four phase control Arterial detector placement is based on the same logic used for three phase control advance detectors supplemented with stop bar detection to ensure phase calls for stopped vehicles Frontage road ramp detector placement is dependent on clearance times and travel times within the interchange The following procedure illustrated in the next figure can be used to calculate the setback distances for frontage loops 18 LEFT
158. ntersection which in this case is Somerset is also 0 second As you review this output remember that the reference phases Phase 6 at New Laredo and Phase 2 at Somerset ate different at these two intersections Also notice that large and equal sized bands in the two directions are characteristic of plans with good progression Such results are not always possible to achieve for larger arterial systems Note for future reference that PASSER V does not allow locking of user specified offsets when using the PASSER II optimization tool However under the settings for each individual intersection by selecting Control clicking on an intersection and then selecting the Timing Data tab it allows the user to specify a preference for sequences to consider during phasing sequence optimization These choices ate No no optimization All select from all possible sequences consider only lead lead or lag lag LL or GG phasing sequences or consider only lead lag or lag lead LG or GL phasing sequences In the case of no optimization the program uses the existing sequence 141 Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis 3 ig Ll Summary Report Detailed Repott G j ENEA Timing Source PASSER II SW Military iro m SW Military C Show All Sub Arts Cycle 70 sec Redraw Hide All Sub Arts Efficiency 40 7 Attainability 100 0 ae Artery List EB EB Band 280s
159. o Way button from the function toolbar and draw a 2000 foot long east west roadway for Harvey Road Then click on the One Way button since we will be drawing one way frontage roads and draw a southbound frontage roadway such that it intersects Harvey Road about 3 grid squares each grid square is 200 ft by default to the left of center Finally draw a northbound frontage road such that it intersects Harvey Road 3 5 grid squares to the right of the southbound frontage road When you are finished your screen should look like the image below PASSER V 09 Untitled0 p5i Ea View Window Help Oc o o Omi im O a o 5 5 amp amp oP O 120 Next we will rename the artery for better reference Click on the Select button from the function bar and click on any link labeled Artery 1 A Link Data dialog box window will appear To change the artery name simply replace Artery 1 by Harvey as shown in the figure below To confirm the change click OK Link Data Link Z Linki Artery Name Harvey EB Dummy gt Artery 2 Link Length 595 feet Intersection Width Storage Length 547 feet Link Speed 30 mph Travel Time WB Artery 2 gt Dummy Link Length 595 feet Intersection Width Storage Length 547 feet Link Speed 30 mph Travel Time r L eron Diamend Date i ax Now follow the same procedure to label Artery 2 as SH 6 South and Artery 3 as SH 6 Nort
160. omplete data entry and analysis A perusal of the sketch of the intersection on page 108 reveals that the right turns at this site are channelized In the case of a signalized intersection we had ignored this information because it was not important However this information is important for determining conflicting flow at TWSC intersections As illustrated below change all right turns to channelized and click the Update button Free Free Free Free Free Free Stop Stop Stop Stop Stop Stop Yes Yes l Yes Yes 1 1 1 109 This site does not have median flared approaches or pedestrians so we will leave the remaining data fields unchanged Now click on the MOEs tab The following screen capture illustrates the output of analysis Note that you will have to stretch the screen to replicate this illustration Node Data Export Intersections Controller Type Controller Id 3 Artery 1 at Artery 2 Unsignalized TWSC v Area Type Other v Intersection Data Capacity Data Headway Data MOEs 5547 50 183 46 0 00 210 65 ooo 94 00 66 17 159 29 218 85 F F 013 013 os 095 546 433 76 52 A A E E 5547 50 193 45 0 00 21065 ooo 94 00 57 58 66 17 159 29 218 85 F F o17 017 oo 0 73 0 95 A A a c E 017 5 62 5 46 Settings Platoon Dispersion Model Lock Capacity Lock Critical Headway I HCM Lock Follow up Time P C Manar and Baass As identified by two blue rectangles this screen provides re
161. on Texas June 1991 Messer C J R H Whitson C L Dudek and EJ Romano A Variable Sequence Multiphase Progression Optimization Program In Transportation Research Record 445 TRB National Research Council Washington D C 1973 pp 24 33 Venglar S P Koonce and T Urbanik I PASSER III 98 Application and User s Guide Texas Transportation Institute College Station Texas 1998 Wagner F A D L Gerlough and F C Barnes Improved Criteria for Traffic Signals on Urban Arterials NCHRP Report 73 1969 Chaudhary N and CJ Messer PASSER IV 96 Version 2 1 User Reference Manual TTI Report 1477 1 College Station Texas October 1996 Chaudhary N V Kovvali C Chu J Kim and S Alam Sofware for Timing Signalized Arterials TTI Report 4020 1 College Station Texas September 2002 Kovvali V Development of a Robust Arterial Coordination Software Ph D Dissertation Dept of Civil Engineering Texas A amp M University College Station Texas December 2001 Wall M Galib A C Library of Genetic Algorithm Components lhttp lancet mit edu ga Accessed May 8 2001 Webster F V and B M Cobbe Traffic Signals Road Research Laboratory Technical Paper 56 Her Majesty s Stationary Office London 1966 Akcelik R Traffic Signals Capacity and Timing Analysis ARR 123 Australian Road Research Board Victoria March 1981 Chaudhary N C L Chu K Balke and V Kowvali Coordination of Diamond In
162. on is critical at each cycle length within the cycle length range you specify This information can potentially be used to help identify where geometric improvements right or left turn bay improvements etc might play a role in improving overall corridor mobility along the artery Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis C 2 5 DEER Input Information Chart Critical Movment Total Throughput Artery 1 C Show All Sub Arts Hide All Sub Arts Artery List Artery 1 The T Sp Diagram tool within PASSER V is mentioned as an analysis tool because it can be used for purposes beyond showing the progression resulting from a PASSER II or GA Optimizer analysis Offsets can be manually adjusted in the T Sp Diagram tool so that you can see what effect different offsets at each intersection will have on your arterial roadway In some instances you may want to manually adjust a PASSER II or GA solution to create increasing bandwidth at successive downstream intersections or you may want to ensure that at least some arterial green is available at each intersection to clear out the main approach before the arrival of the progression band Once you have made the desired manual adjustments clicking on Redraw will show you the resulting progression If you do opt to make manual offset adjustments the T Sp Diagram tool will indicate that the source of the
163. on is not located within the CBD central business district or downtown area of San Antonio Also note that because this is an isolated intersection analysis we do not need to worty about primary coord phase offset reference or offset value information for coordination Begin your data entry by clicking in the Lane Assignment field for the eastbound movement A new window appears prompting you for information about all of the lanes of the eastbound approach to the intersection In this case there are a total of four lanes The leftmost lane is a left turn bay the middle two lanes are for through movements only selectable by clicking in the check boxes beneath the lane and the right lane is for through and right turn movements Make sure to enter the 148 foot length for the left turn bay and 91 the appropriate lane widths for each lane Note also that we are assuming level grade for this approach no bus stops in the peak hour and no on street parking maneuvers Median Type and Median Width data fields will be discussed in the next chapter For now ignore these fields When you are finished with geometric data entry for the eastbound approach your approach entry window should look like the one below EB at Node 3 TotalLanes 4 Median Type None Aa v a v Grade fo Cancel Remove All Lanes It is important to note at this point that there is a relationship between the right turning geometry you enter into PA
164. ons The prevalence of signal preemption devices or Intelligent Transportation System ITS technologies may require special consideration Specialized traffic engineers who understand the specific standards and guidelines required for installation install many of these special configurations for dealing with these circumstances These circumstances include railroad preemption if the intersection interchange is adjacent to a railroad grade crossing fire and or emergency medical service EMS priority treatment and bus transit and or rail transit priority treatments ITS technology for instance changeable lane assignment signs controlled by the traffic signal controller is less prevalent and may also require special modifications to standard timing procedures developed within this guide Analysis Tools A number of computer programs are available to assist in the analysis and coordination of traffic signals on an arterial All of these programs are based on the abstraction of reality and have their inherent weaknesses and strengths In this section we provide a description of the most commonly used programs for analyzing and optimizing signal timings We begin by describing key concepts needed to better understand these programs Traffic analysis software may contain one or both of the following modules e a traffic simulation model also called a traffic model and e an optimization model These two types of models are described in the fo
165. ontains a number of new analytical tools that will provide further insight into the traffic engineering operations problems you analyze n this chapter we will discuss the PASSER V program s user interface and the Installing PASSER V Install PASSER V on your computer by inserting the PASSER V CD in your CD drive and allowing the installation process to execute automatically If the installation process does not begin automatically navigate to your CD drive using Windows Explorer and double click on the setup exe file Once the installation process begins a splash screen appears with the Texas Transportation Institute the program s developer logo Click on Next to advance to an installation screen telling you to exit other applications before installing PASSER V After exiting other open applications click Next At the next screen enter your username and company name followed by a click on Next The default folder for PASSER V installation is C Program Files PASSER V 09 It is recommended that you retain this default install folder and click on Next The next screen will tell you that PASSER V icons will be inserted into a program folder that the installation program creates named Passer V 09 Once again retain the default values and click on Next At the next screen confirm your installation settings and click on Next The installation process will copy the PASSER V files onto your computer
166. ork Li Summary Report Detailed Report G Artery List Diamond 7 Timing Source PASSER Mi tae Time Artery 1 Diamond Show All Sub Arts Cycle 40 sec Hide All Sub Arts Efficiency 31 3 Attainability 100 0 Artery List EB EB Band 15 0sec EB Attain 100 0 Sates WB Band 10 0sec WB Attain 100 0 Artery 10 20 Diamond i i Artery 2 Ref Phase 2 No 0 11 400 00 ft Artery 3 Optimization Analysis Tools Aaa PASSER Ill GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis a IE List Input Information Chart Critical Movment Total Throughput Diamond Show All Sub Arts Hide All Sub Arts Throughput Analysis Artery List 3 H Existing Volumes i i F gt g Basic 3 Phase El Artery 1 E en a on a O Ext 3 Phase Diamond X A i TTI 4 Phase 400 ft 60 70 SO 0100 Cycle Length 88 Chapter Analysis of Isolated Signalized Intersections Using PASSER V to analyze an individual or isolated intersedion available in PASSER V is for an isolated intersection But before you start PASSER V make sure that you have all of the input data you need geomettic details traffic volume data and signal settings well organized and within reach T he most simple type of optimization analysis you can perform using the tools When you are ready to start PASSER V double click on the program icon if you created one or click on St
167. ormation required by traffic analysis software there may also be a need to collect information about how well the current system is operating These data may be collected to determine the level of service of the existing arterials or intersections to calibrate and compate against results produced by the analysis tool or to document the overall impacts of mobility ie overall fuel consumption or emissions information Typical studies 70 performed for these reasons include delay studies which involve recording the number of vehicles in queue for each intersection approach at a set time interval travel time studies which usually involve driving a probe vehicle along the study arterial and either manually or electronically recording travel time between intersections or a combination of both of these studies to gain an overall understanding of arterial system performance Detailed information about performing both types of studies can be found in the Manual of Transportation Engineering Studies 29 71 Chapter PASSER V 09 An easy to use Windows graphical user interface alloys you to quickly anabyze arterials andl or diamond interchanges with this sofware optimization features of the software With its broad range of applicability PASSER V can be considered an upgrade and replacement for the PASSER II and PASSER II programs that precede it in that PASSER V contains the analytical capabilities of both tools In addition PASSER V c
168. p time also known as the internal overlap or the travel time overlap from the PASSER V calculated default value if you choose to do so However if you do make adjustments to this value make sure that you do not make the overlap greater than the value automatically calculated by PASSER V since this could result in arterial vehicles from one side of the interchange arriving at the interior of the second intersection before the light is green to recetve them a possible violation of driver expectancy depending on interchange spacing and previous operation mode Similar to the data entry procedure that we used for an isolated intersection we will now code the geometric traffic volume turning movement and signal setting details for the interchange Remember as you enter right turn volumes to remove right turn volumes for right turn lanes and bays of adequate length Le not blocked by the through vehicle queue by telling PASSER V that there is no right turn movement for that approach If some right turns on red are consistently likely i e where a right turn channel is provided without a bay be sure to factor down the right turn volume you enter by the number of right turn on red vehicles you expect in the peak hour After you have finished with the geometric and volume details your screen should look like the one on the next page 123 Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id
169. proach geometry and left turn storage space as it is by the lane assignment and storage space in the interior The figure below illustrates important points to consider Alignment of lanes through all intersections applies throughout the interchange Length of exterior left turn storage bay if applicable Signal Timing Information For a signalized intersection interchange several important details about existing signal timing are essential for analyzing existing conditions An engineer must also consider the capabilities and features of the control hardware in the field prior to strategy development Other details about the intersection interchange include the type of infrastructure at the intersection which includes signal heads controller cabinets and other devices 67 Left Turn Treatment The presence of a designated left turn bay within the interior of the interchange allows special consideration to be given to this movement Observation of the intersection in the field will reveal its current mode of operation The left turn bay may be controlled with either a three or five section signal head A separate three section head limits the type of phasing to either protected or permitted only A five section signal head display will allow a protected permitted left turn which can be used to increase the performance and flexibility of the approach Field observation of left turn operations will verify the type of ope
170. ptimization Models and Search Algorithms Popular Signal Timing Analysis and Optimization Programs PASSER V Optimization Algorithms in PASSER V Analytical and Simulation Models Optimization and Analysis Tools Program Limitations CHAPTER 2 THEORY OF on FPN 10 11 13 15 17 20 24 24 26 28 30 30 35 43 46 TWO WAY STOP CONTROLLED INTERSECTIONS HCM Analysis Approach Critical Headway Follow up Time Calibration of Field Data Additional Factors Affecting Capacity Capacity Calculations Control Delay Queue Length 47 48 49 49 50 51 58 58 Impact of TWSC Modeling on Various Tools Impact on Delay Analysis Routine Impact on Optimization Tools Program Limitations CHAPTER 3 DATA REQUIREMENTS Traffic Volume Information Turning Movement Counts Average Daily Traffic Roadway Geometric Information Signal Timing Information Left Turn Treatment Mode of Operation Speed and Travel Time Additional Information CHAPTER 4 PASSER V 09 Installing PASSER V Running PASSER V File Menu Bar File Access Toolbar PASSER V Function Toolbar CHAPTER 5 ANALYSIS OF ISOLATED SIGNALIZED INTERSECTIONS Geometry Data Entry Volume Data Entry Signal Settings Data Entry CHAPTER 6 ANALYSIS OF ISOLATED TWSC INTERSECTIONS Data Entry and Analysis Effects of an Upstream Signal 59 59 59 60 61 61 63 65 67 68 68 69 70 73 74 74 75 75 89 94 96 105 112 CHA
171. quence of movements at the interchange determine the interchange phase pattern or sequence Phasing sequence names are linked to whether or not the interior left turn precedes or leads the opposing through movement on each side of the interchange The four basic left turn sequences are e Lead lead protected left turn movements from the interior lanes lead the opposing arterial movement at both intersections e Lead lag protected left turn movements from the interior lanes lead the opposing arterial movement at the left intersection and lag the opposing arterial phase at the right intersection e Lag lead the mirror image of the lead lag phasing pattern e Lag lag protected left turn movements from the interior lanes lag the opposing arterial movement at both intersections In addition to alternative phasing sequences left turn treatments at diamond interchanges also vary The interior left turn movements may be protected only protected plus permitted or permitted only Le no left turn phase In the permitted only case these phases would not exist i e their duration would be set to zero and the interchange would operate with only two timed phases This alternative is desirable if a large number of acceptable gaps exist in the opposing traffic stream and sight distance is adequate By allowing permitted left turns it is possible to reduce the overall delay of the interchange by reducing the number of phase changes required Permit
172. r priority conflicting movement on a lower priority movement is different depending upon its movement type i e main street through or right turn and geometrics Therefore HCM suggests that conflicting flow for a minor movement be calculated as a weighted sum of its conflicting higher priority movement flow rate as illustrated in Exhibit 17 4 of HCM 2000 6 According to Kyte et al 31 the critical headway of a movement 7 is defined as the minimum length time interval that allows intersection entry to one minor stream vehicle Though it may be derived from the field data the derivation process is complicated HCM 2000 provides an estimate for this value which is shown as follows ti 5t t uy Pav t ter b or ci c base where t critical headway of movement 4 in seconds ne base critical headway in seconds Exhibit 17 5 of HCM 2000 tuy adjustment factor for heavy vehicles in seconds _ 1 0 _ for two lane major street 7 A for four lane major street Py proportion of heavy vehicles for the subject movement t g adjustment factor for grade in seconds _ 0 1 for right movement of minor street 0 2 for left or through movement of minor street 52 G percent grade divided by 100 adjustment factor for each part of two stage gap acceptance c T process in seconds 1 0 for first or second stage 0 0 for single stage t r adjustment factor for intersection geometry in seconds 0 7 for left movem
173. r to collecting data for a peak period over several consecutive days Due to the level of detail simulated these models require the maximum amount of data and are the most computationally intense Mesoscopic Traffic Models These models simulate traffic flow in specified time steps and they are usually deterministic in nature The time step can be 1 second 2 seconds or higher For each time step these models estimate the flow of traffic entering a link traveling downstream stopping due to a red light and moving again when the light turns green Some of these models also account for platoon dispersion as vehicles travel from one point to a downstream point in space Mesoscopic models can be further classified as link based or time based Link based models simulate traffic flow one link at a time for all time steps in a signal cycle These models treat a queue of vehicles at the signal approach as an upward stack As a result all vehicles arriving during red travel to the stop bar and join a vertical upward stack queue Link based models cannot account for queue spillback because they do not keep track of the back of the queue In addition they may allow more vehicles to stack in a queue than a link s storage capacity Thus these models are not suitable for congested conditions or for short links where sub optimal timing may cause queues to block flow from the upstream signal Step based models on the other hand simulate traffic flow on all
174. ration If no separate indication for left turn vehicles exists e the only signal heads for each approach are the two three section heads called for by the MUTCD 1 permitted operations are virtually always present unless signing indicates that there is a protected turn on green Le no conflicting vehicles are present as in split phase operation If a left turn treatment of a particular type is desired and the hardware and or geometry ate not capable of accommodating the required display either the timing strategy will have to be changed or new hardware or geometry will have to be installed in the field Mode of Operation Current practice for signalized intersection timing calls for the use of one of three methods pretimed semi actuated or fully actuated control The type of controller affects the type of timing plan that can be implemented The timing strategies that may be applied are a function of the type and capability of the controller and the operational requirements of the intersection 28 Most new controllers are actuated controllers that can execute any of these types of control Basic pretimed also known as fixed time strategies can be used when traffic at the intersection is relatively steady day to day These plans utilize a fixed cycle length phase sequence and phase lengths to serve traffic Different timing plans may be programmed to deal with fluctuations in traffic volume throughout the day and to implement p
175. rection can be no more than the smallest through green split in that direction The following formulas are used to calculate combined efficiency and attainability for the two arterial directions Arterial Band a Arterial Band 2 x Cycle Length x 100 Progression Efficiency Arterial Band Arterial Band x100 Min Green Min Green Progression Attainability The reader should note that while bandwidth generally increases with an increase in cycle length efficiency may increase decrease or remain constant Thus it is desirable to select a solution that provides the best efficiency and an attainability of 100 percent In addition the timing plan should not use cycle lengths larger than that necessary to move traffic through all approaches on the arterial 31 It is a well known fact that PASSER I 90 has a tendency to select larger optimal cycle lengths Recently TTI researchers found that this tendency is because of the split optimization feature implemented in PASSER II What happens in PASSER II is that the split optimization gives more and more green time to the through traffic as cycle lengths increase Since PASSER II optimizes bandwidth it tends to select larger cycle lengths because of larger bands and efficiencies TTI researchers decided to not use the split optimization feature in the PASSER V implementation of the algorithm Thus the splits calculated in PASSER V are slightly di
176. reet right turn movements These movements yield to only Rank 1 streams Cross street through and left turn movements are ranked 3 and 4 respectively Cross street left turns have the lowest rank because these movements yield to all movements including opposing through and right turn vehicles on the minor street HCM methodology assumes 47 that these priority rankings are maintained at all times even though observations have shown that these rankings are often violated by drivers Traffic Movement 2 3 5 6 15 16 1 4 13 14 9 12 HCM methodology is based on gap acceptance theory which recognizes that the drivers of minor movements that is Rank 2 3 and 4 at TWSC intersections are not given positive indication about when it is safe to enter the intersection Thus each minor movement driver must determine if e the major stream has large enough gaps that would allow a safe maneuver and e itis his her turn to use the gap Gap acceptance theory formalizes these steps into models which assume homogeneous drivers These models use two key parameters namely critical headway and follow up time for each minor movement at a TWSC intersection The following subsections describe these parameters and other factors Critical Headway Headway measured in seconds is the time between two successive vehicles as they pass a point on the roadway It is measured using the same feature of both vehicles Le f
177. riod operations For practical reasons you may want to use a cycle length no shorter than 60 seconds To make this change simply go back into the PASSER III and or GA Optimizer tools and increase the lower bound on the cycle length range to 60 seconds And because of the wide spacing 720 ft between these two halves of the diamond you might want to just use a Basic three phase or Extended three phase solution in PASSER HMI To remove the four phase sequence from being an optimization option simply go to the Input tab under PASSER III and remove the check mark next to the four phase option When you are finished experimenting with the isolated diamond move on to the next chapter to analyze multiple intersections along an arterial For practical reasons the example is limited to two intersections along an arterial for this class However using PASSER V you can analyze multiple arterials each with many intersections For convenience and or consistency with your physical configuration in the field you can even separate intersections along arterials into groups 129 Chapter Arterial Analysis Using PASSER V to analyze multiple intersections along an arterial roacun arterial roadway As we proceed through the process of data entry analysis and review of results for this multi intersection example we will be using features of PASSER V that we did not use for our isolated intersection or our diamond interchange O ur next analysis t
178. rithms use some level of exhaustive search combined with other search algorithms Hill Climbing Algorithm A hill climbing or valley descent algorithm starts with a base scenario specified by the user selected by the program using a fixed criterion or selected randomly Then it selects a variable to be optimized Le offset cycle length etc and creates two additional scenarios for this variable one by increasing the value of that variable and the other by decreasing the value Initially the value of the selected variable is increased or decreased by a specified amount called the step size Following this the algorithm uses a traffic simulator to calculate the fitness values for each of the two new scenarios and compares them with the base scenario These evaluations identify the two best scenarios and consequently a direction of further search For instance if increasing the value of the selected variable resulted in a better fitness value the search algorithm will mark this new scenario as the current best and continue searching in the direction of increasing values for the variable In the next iteration the search algorithm generates a new scenario by increasing or decreasing the value of the 26 selected variable in the selected search direction calculating the new fitness value and comparing it with the two current best values The algorithm continues in this manner until the fitness value for the new scenario ceases to be
179. ront bumper front axel rear bumper etc Critical headway is the minimum time between successive major stream vehicles in which a minor movement vehicle can make a safe maneuver Critical headway differs from driver to driver and even varies for the same driver depending upon traffic conditions For instance drivers are willing to accept smaller than normal headways as wait times increase during peak traffic hours Other factors that affect critical headway include type of minor movement i e cross street left turn number of lanes on the main street and visibility 48 For use in gap acceptance models a representative value of critical headway must be obtained Such a value cannot be observed directly in the field but must be derived from data collected in the field Literature contains several methods for deriving the critical headway The values recommended in HCM 2000 used as default values in PASSER V were obtained using the maximum likelihood ML method This method uses an iterative procedure to determine critical headway using the distributions of accepted and rejected headways from field observations This approach is complicated and time consuming Instead the following formula can be used to obtain critical headway Critical Headway Follow up Time 0 6 The definition of follow up time and its field measurement are discussed in the next subsection Follow up Time Follow up time applies to vehicles on minor movemen
180. rsaturated facilities Consequently it also requires more computation time Until recently the main deficiency of TRANSYT 7F has been its inability to optimize signal phase sequences In TRANSYT 7F version 9 this deficiency was removed through the addition of a GA based optimization algorithm TRANSYT 7F models actuated signals as equivalent pretimed signals and it has the ability to half double cycle traffic signals 28 TRANSYT 7F performs exhaustive searches for cycle length For each cycle it starts by calculating equal saturation splits and applies a hill climbing method to optimize signal offsets and splits For this reason its final results depend on the base timing plan supplied by the user Although it contains a good delay based traffic model TRANSYT 7F s bandwidth analysis model is not very good Last but not least learning to use TRANSYT 7F requires considerable effort Synchro Synchro 15 is a delay based program for analyzing and optimizing timing plans for arterials and networks Its objective function also minimizes stops and queue size by applying penalties for these measures Synchro s traffic model is similar to the link based model in TRANSYT 7F Synchro uses an exhaustive search technique to optimize signal timings To reduce the number of scenarios analyzed for a coordinated system it relies on the divide and conquer principle To optimize timings for an arterial the program requires the user to apply several m
181. rval Transition Controller Types As electronics and computer technologies have continued to evolve over time these advancements have cartied over into traffic signal controller technology to produce more reliable flexible and functional devices Five general types of controllers are described below e Electromechanical These devices use synchronous motors and cams to open and close electronic circuits that govern the signal indications at an intersection They are pin programmable for such variables as cycle length and phase split they provide the engineer with the capability of changing cycle length split and offset C S O and they can accommodate changing C S O by time of day Though electromechanical controllers can still be found in the field today they should be considered technologically obsolete and replaced with modern solid state controllers e Type 170 170 controllers are based on a hardware equipment specification jointly developed by the states of California and New York Buying a Type 170 controller is like buying a personal computer you get a standard piece of electronics but you have to buy software to make it do something useful Several national vendors provide a range of software for the 170 Type 170 controllers have proven extremely reliable and flexible over time but the technology i e eight bit microprocessor is over 20 years old Some vendors offer updated programmable read only memory PROM cards for
182. ry Cycle 70 sec Signal 3 SW Military and New Laredo Ref Phase 6 Begin of Green Phase Offset 0 sec Avg Avg Vehicles Delay Delay Thruput Capacity Blocked Stops Phase sv Los v h v h vic v h EBL 29 36 127 78 323 59 102 20 EBT 17 55 846 67 2368 35 0 00 556 38 WBL 32 21 49 45 75 04 0 00 39 42 WBT 14 38 803 64 2020 09 0 00 524 77 WBR 14 38 42 18 106 02 0 00 27 54 SBL 24 62 85 56 336 99 0 00 67 22 SBT 35 48 178 89 330 85 151 99 SBR 47 82 56 67 104 80 48 15 NBL 28 97 21 11 191 74 17 04 NBT 30 78 142 22 417 58 117 96 NBR 37 19 60 00 177 47 0 00 54 00 B C B B Cc D D Cc c D Signal 8 SW Military and Somerset Ref Phase 2 Begin of Green Phase Offset 0 sec Avg Avg Vehicles Delay Delay Thruput Capacity Blocked Stops siv LOS v h v h vic v h 33 38 57 37 82 75 0 00 46 32 16 84 867 55 1730 78 0 00 612 66 16 80 86 69 173 53 0 00 60 98 35 39 112 22 232 07 93 85 21 37 683 33 2035 71 0 00 487 62 30 18 38 89 202 24 31 70 32 35 232 22 438 43 193 81 36 17 42 22 79 71 38 00 20 40 135 56 499 35 92 95 29 07 176 67 391 00 141 33 31 00 115 56 255 75 92 44 ooo0Ng0ofc 070m To get an idea of the quality of progression provided by our 70 second cycle length timing plan click on the T Sp Diagram tool Your time space diagram should look like the illustration provided below Notice that the offset for the first reference intersection is 0 seconds and the offset for the second i
183. s S SN fu fuv eSpor fafu fir frr tipo f rpp where fa fiu fir frr fip Rpb actual saturation flow rate for lane group in vphg ideal saturation flow rate per lane usually 2000 2200 pephgpl number of lanes in the lane group adjustment factor for lane width adjustment factor for heavy vehicles adjustment factor for grade adjustment factor for adjacent parking lane and parking activity adjustment factor for local bus stopping ad ad adjustment factor for left turns in lane group J justment factor for area type justment factor for lane utilization j adjustment factor for right turns in lane group pedestrian bicycle adjustment factor for left turns pedestrian bicycle adjustment factor for right turns The HCM 6 is the primary source for information about saturation flow rate adjustment factors Key Point The calculation of appropriate saturation flow rates is essential to performing an analysis that will produce reliable results Saturation flow rates are a key element of most manual and computerized forms of traffic engineering analyses including PASSER V Webster s Equation F V Webster developed an equation that approximates the minimum delay cycle length for an intersection 10 1 5L 5 10 Y Y n Co where Co optimum cycle length in seconds L lost time per cycle generally the sum of the total yellow and all red clearance per cycle in seconds
184. s Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis C Input Output Summary Report Detailed Report Plot G Artery List SW Military Title p Il Summary Report for Sw Military CSV File Print C Show All Sub Arts PUES Cycle Total EB WB Total EB Total EB WB Avg Total Starve Vehicles Enty Exit Artery List sec Band Band Band Eff Eff Att Attain Attain Delay Stops Time Blocked Vol Vol SW Miltay sec sec sec _ sec veh veh hr sec hr veh cyc veh hr veh hr i i 72 00 3500 37 00 4235 41 18 100 00 100 00 100 00 23 86 2526 88 0 00 0 00 1853 1869 76 00 37 00 3900 4222 41 11 100 00 100 00 100 00 25 18 2564 43 0 00 0 00 1853 1869 57 00 2800 2900 40 71 40 00 100 00 100 00 100 00 21 97 2607 86 0 00 0 00 1853 1906 5200 2600 2600 40 00 40 00 100 00 100 00 100 00 21 53 2718 38 0 00 0 00 1853 1912 64 00 3300 31 00 40 00 41 25 9412 100 00 88 57 22 71 2461 88 0 00 0 00 1853 1870 60 00 3000 3000 40 00 40 00 96 77 100 00 93 75 22 81 2614 82 0 00 0 00 1853 1888 43 00 2300 2000 35 83 38 33 93 48 100 00 86 96 21 18 2889 52 0 00 0 00 1853 1922 32 00 1800 1400 29 09 3273 84 21 94 74 73 68 21 35 3101 56 0 00 0 00 1853 1925 24 00 1300 11 00 2400 26 00 85 71 9286 78 57 21 73 3291 44 0 00 0 00 1853 1843 By default the results are organized by assuming total efficiency percent of bandwidth devoted to progression as the desired selection criterion Another cri
185. s at Somerset assume all westbound right turns can be handled as right turns on ted i e do not code a right turn movement for this approach On the other three approaches the fact that the right lane is a shared lane with no channel will lead us to enter the full right turn volume Left turns on S W Military are protected only while those on the cross street are protected permitted Also assume pedestrian buttons min green 6 sec for all approaches and use a yellow of 4 seconds and red clearance of 2 seconds Under the Sat Flow Data tab use a PHF of 0 90 and use the truck percentages shown Check the figure below to confirm your input data for Somerset Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id fe SW Military at Somerset Pretimed Signal z weT WBA v BeginofGreen Cycle Length 90 AreaType Other v NTCiPDleetHeferencing Cee e eo Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs 2 gt 3 209 122 159 Prot Per Prot Prot Pem Prot Lead Yes Yes 2000 21 11 2778 27 78 8 hs a 5 Optimization Settings Lock Sat Flows Lock Green Splits Update At this point you should notice that PASSER V chooses WBT WBR Phase 6 as the coordinated phase for this intersection while it selects EBT EBR Phase 2 for the intersection at New Laredo This difference in selecting the coordinate ph
186. s between adjacent signalized intersections Often it is desirable to analyze the operational performance of such intersections and their impact on the operation of adjacent traffic signals or vice versa Such a need arises especially when evaluating various access management alternatives PASSER V 09 provides for such analyses through the incorporation of HCM 2000 6 procedures for TWSC intersections HCM Chapters 10 and 17 In implementing these procedures all corrections to date 30 have been applied Macroscopic modeling of TWSC intersections is complicated As such no attempt has been made to reproduce HCM methodology here The intent of this chapter is to provide information necessary for the effective use of PASSER V Readers interested in further details are referred to the two references cited above P ASSER V s primary application is the coordination of traffic signals on signalized HCM Analysis Approach Two way stop control regulation implies that the main street through and right turn movements have absolute priority Other movements yield to these streams and use any remaining intersection capacity according to the assigned priority rankings The following figure shows the ranks of movements for a four legged TWSC intersection Rank 1 streams include through and right turn movements on the main street and pedestrian movements on the cross street Rank 2 movements include main street left turn main street pedestrian and cross st
187. s not include pedestrian buttons you need to ensure the minimum phase times you enter into PASSER V are adequate for pedestrians as well as vehicles We will assume there are no pedestrian buttons at our intersection and we will compute pedestrian minimum phase time as 7 seconds of walk plus the result of roadway width divided by an average walking speed of 4 feet per second For the northbound through phase this equation results in a minimum pedestrian phase time of 7 124 114 124 134 12 11414 4 28 25 or 29 seconds After computing and entering the minimum pedestrian phase times for all approaches your screen should resemble those in the screenshot on the next page Note that the MUTCD 1 indicates that a pedestrian walk signal should be active for 7 seconds unless conflicting phases need added green time in such situations 4 seconds can be used as the pedestrian walk time Another consideration that affects the entered minimum green time is whether your agency allows you to simultaneously clear the vehicular and pedestrian portions of a phase Le simultaneous vehicular yellow ball indication and pedestrian flashing DON T WALK If your agency allows this control scenario note the MUTCD 1 does not prohibit you from allowing it subtract the phase s yellow change interval time from the minimum green time and enter the result into PASSER V Remember to check the minimum green time after subtracting the yellow change interval tim
188. s of a chromosome The standard genetic algorithm proceeds as follows 22 1 It randomly or heuristically generates an initial population generation 0 of candidate solutions for a given problem 2 For every evolutionary step known as a generation it evaluates the fitness bandwidth delay etc of each solution 3 It forms a new population the next generation by selecting the individuals with best fitness and applying natural selection schemes genetic operation mutation and recombination to pairs of individuals 4 It removes unwanted members of the population to make room for new members 5 It evaluates new individuals and inserts them into the population pool 6 If termination criterion is met it stops otherwise it goes back to step 3 A single iteration of this loop is referred to as a generation Natural selection guarantees that individuals with the best fitness will propagate into future populations Using the recombination operator the GA combines genes from two parents to form two new 32 offspring that have a high probability of having better fitness than their parents Mutation allows infusion of features not present in parents Over several generations the best individuals survive and the worst are eradicated The reader should note that the pre selected population size remains unchanged from one generation to the next The following figure shows a flowchart of this methodology Initialize Population Evalu
189. signals For the first set it uses the HCM model to calculate average delay for each signal Then it adds average delays for all signals to determine the total system delay For the second set of calculations using DAR phasing sequences and offsets are also required In this case the tool uses current phasing sequences at all signals For offsets the tool provides two options If the user desires the absolute offsets to remain the same as those entered on the data screen the tool uses the same offsets for all cycle lengths However if the user selects the Proportional offset the tool recalculates the offset of each cycle in proportion to the change in the cycle length from that coded for a signal on the data screen 45 Program Limitations PASSER V analysis and optimization capabilities are limited to pretimed signal controlled intersections with three to eight signalized approaches The program currently supports controllers with two tings and multiple barriers only In addition all analysis and optimization of multiple signals diamonds or arterials assumes that the signals in the system operate under a common cycle length The program does not support double half cycling or conditional service 46 Chapter Theory of Two Way Stop Controlled Intersections Things you need to know about the analysis of TWSC intersections in PASSER 1 09 arterials Such facilities commonly have TWSC intersections or unsignalized driveway
190. signals within an arterial You might organize by groups to create logical groupings of controllers to mirror the way that your controllers are connected coordinated in the field or to allow for different cycle lengths coordination plans along different portions of any given arterial roadway To create a subsystem simply type in a name for the grouping next to Artery Name in the Subsystem dialog box and click Add Then click on each link that you want to include in the new subsystem you created right clicking the mouse when you are finished As you create a subsystem note that subsystems can only be composed of the links along one arterial roadway i e subsystems are subsets of selected links along one artery When you have finished the process of creating a subsystem you can click on the Subsystem button again You will notice that the artery along which you created a subsystem has a plus sign next to its name Clicking on the plus sign will expand the subsystem listing see figure below SUBSYSTEM g Artery Name Artery List Artery List Group 1 Artery 2 Artery 3 Artery 5 Artery 6 Once you have completed data entry traffic volumes geometric details signal parameters for each intersection along an artery the nodes for that artery should appear in light blue versus red If you think you have finished entering data for each node but one or more nodes remain red in color review your input data for internal cons
191. st be examined and approved by a licensed professional engineer who is trained in the field of traffic engineering PASSER V software and related documentation are copyrighted This software and documentation may not be copied or reproduced for commercial purposes Modifications or alterations in the meaning intent applications or operations of the software or documentation is absolutely prohibited unless prior approval has been obtained from the Texas Transportation Institute Trademark PASSER is a trademark of the Texas Transportation Institute As such any use of this trademark must have prior written approval from the Texas Transportation Institute COPYRIGHT 2009 Texas Transportation Institute All Rights Reserved Use of the PASSER trademark software or documentation in whole or part within the body of another work except for brief citations is prohibited Selling or redistributing the PASSER trademark software or documentation by any person or agency other than the Texas Transportation Institute and its authorized agents is prohibited Table of Contents CHAPTER 1 TRAFFIC SIGNAL THEORY Background Intersection Configuration NEMA Phasing Interchange Phasing Strategies Controller Types Modes of Control Theory Saturation Flow Rate Webster s Equation Minimum Green Time Yellow Change Red Clearance Pedestrian Treatment Traffic Detection Safety Considerations Analysis Tools Traffic Signal Analysis Models O
192. sults for two scenarios The first part top rectangle shows results of analysis assuming that the TWSC intersection is isolated The second part bottom rectangle presents results that have been modified to account for platoons arriving at any nearby traffic signals The magnitude of impact of adjacent signals on calculations is a function of the proximity of an adjacent signal or signals its their timings and magnitudes of platoons By default the program applies the HCM platoon dispersion model one of the two options in the purple box to arrive at these modified values This model however does not account for platoon compression which occurs due to friction between vehicles under heavier demand scenarios If you desire results to account for both platoon dispersion and platoon compression then you should select the model developed by Manar and Baass the second option in the purple rectangle In this exercise however both sets of analyses shown in the two blue rectangles are the same because this is an isolated intersection Furthermore platoon dispersion model selection will 110 have no impact on the calculations The following criteria are used for generating delay and v c ratio LOS in the output illustrated above Level of Service Delay sec veh v c Ratio 0 10 0 lt 0 6 gt 10 15 0 6 lt 0 7 A115 25 0 7 lt 0 8 es at 0 8 lt 0 85 gt 35 50 0 85 lt 1 0 gt 50 gt 1 0
193. sured from the field On the other hand critical headway cannot be directly observed from the field It must be derived from headway data collected in the field As an alternate you can assess critical headway from follow up time Please see Chapter 2 for a detailed description of these and other related concepts Effects of an Upstream Signal To study the effects of an upstream signal we will use four cases These cases are saved as Isolated TWSC 1 Signal p51 Isolated TWSC 1 Signal V2 p5i Isolated TWSC 1 Signal V3 p5i and Isolated TWSC 1 Signal V4 p5i PONS The first three of these cases illustrated graphically in the following figure are provided to study the distance effect of an adjacent signal The only difference between these cases is the length of link between the two intersections Case 4 not shown graphically has the same geometric configuration as Case 3 but westbound traffic volume arriving from the adjacent signal to the TWSC intersection is 50 percent of the traffic in Case 3 112 lsolated TWSC 1 Signal p5i O O TWSC Sgnal 9 O TWSC Signal TWSC Signal lsolated TWSC 1 Signal V2 p5i O TWSC Signal TWSC Signal O DISMAL Signal lsolated TWSC 1 Signal V3 p5i O TWSC Sgnal TWSC Signal O DML Signal 113 Open Case 1 and compare the difference in MOEs for Isolated and With Platooning for the TWSC intersection As illustrated in the next figure you
194. t to note that the inhibit functions are not consistent with the use of lead lag or lag lag phasing since the applied inhibits would always not allow the lagged left turn phase to be displayed 1 e phase 2 inhibiting phase 1 would always not allow phase 1 to become active if phase 1 is lagging Lead lag phasing yellow trap solutions include the Dallas phasing option mentioned earlier or using protected operation only for the leading left If both opposing left turn displays are protected only you should only apply a dual lag lag left turn sequence if max calls are placed on the through phases and min recalls are placed on the left turn phases to ensure that the lag turns begin simultaneously otherwise you could create a yellow trap situation if one of the protected left turn phases is skipped As a general rule when using standard eight phase operation if you inhibit phase 1 with phase 2 phase 3 with phase 4 phase 5 with phase 6 and phase 7 with phase 8 you will ensure that the yellow trap never occurs in a protected permitted left turn display You can safely lag a left turn phase by following this rule Lift the Inhibit Phase if the opposing left turn is protected only 12 Note that limited visibility or programmable signal heads can also be used to alleviate the yellow trap situation By not allowing left turning vehicles to see the signal indications for their associated through movement and vice versa potential driver
195. ted left turns also increase the potential capacity of the movement by increasing the time the movement is allowed to proceed through the intersection Interchange Phasing Strategies The basic diamond interchange strategies are two phase three phase and four phase Each of these strategies uses a different phasing structure to serve the traffic at the interchange A discussion of the operation of each follows Two Phase Two phase operation can be used at diamond interchanges operating under low traffic demands The two phases from which the strategy derives its name are the arterial phase and the frontage road or ramp phase In this strategy the interior left turn movements do not have a protected phase Le a left turn arrow but proceed permissively during the arterial phase under a green ball indication Two phase operation is beneficial when the left turn and or opposing through traffic volumes are light however sufficient sight distance must be available to the left turning vehicles to determine whether it is safe to make the turn Three Phase For three phase control the three phases are the arterial phase the ramp frontage road phase and the interior left turn phase at each intersection The two intersections can operate independently using coordination or the intersections can be controlled by a single controller thus providing a more defined relationship between the intersections Protected only or protected plus permitted
196. terchanges with Adjacent Traffic Signals TTI Report 4913 1 College Station Texas October 2000 Jain S An Enhanced Model for Calculating Delay as a Function of Offset M S Thesis Texas A amp M University College Station Texas 1996 Kell J and I Fullerton Manual of Traffic Signal Design Institute of Transportation Engineers Washington D C 1991 Manual of Transportation Engineering Studies Institute of Transportation Engineers Washington D C 1994 151 30 31 32 33 Approved Corrections and changes for the Highway Capacity Manual 2000 TRB Committee AHB40 Highway Capacity and Quality of Service http people sunyit edu lhmi ahb40 hem Approved 20Corrections 20to 20HCM 202000 200 ct 202006 pdf Accessed August 16 2007 Kyte M Z Tian Z Mir Z Hameedmansoor W Kittleson M Vandehey B Robinson W Brilon L Bondzio N Wu and R Troutbeck NCHRP Web Document 5 Capacity and Level of Service at Unsignahzed Intersections Final Report Volume 1 Two way Stop Controlled Intersections 1997 http www nap edu books nch005 html Baass K G and S Lefebvre Analysis of Platoon with Respect to Traffic Volume In Transportation Research Record 1194 TRB National Research Council Washington D C 1988 pp 64 76 Manar A and K G Baass Traffic Platoon Dispersion Modeling on Arterial Streets In Transportation Research Record 1566 TRB National Research Council Washington D C 1996
197. terion can be selected by clicking on the corresponding column heading As described below the result will depend on the criterion selected e Cycle lengths are arranged from smallest to largest e Bands efficiencies attainabilities entry and exit volumes are arranged from the largest best to smallest worst 137 e Delay stops starve time and blockages are arranged from lowest best to highest worst For example click on the column heading Avg Delay the corresponding column will be highlight and results will be rearranged from lowest to highest average delay Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis SubNetwork List Input Output Summary Report Detailed Report Plot Artery List SW Military Title pee Il Summary Report for SW Military CSV File Print Show All Sub Arts ee De MMM piua eai Cycle Total EB WB Total EB WB Total EB WB Avg Total Starve Vehicles Enty Exit Artery List sec Band Band Band Eff Eff Eff Att Attain Attain Delay Stops Time Blocked Vol Vol SW Mitay sec sec sec M W W a seciveh veh hn secht veh cyc veh hr veh hr 3 E 60 j 43 00 23 00 20 00 35 83 38 33 93 48 100 00 86 96 21 18 2889 52 0 00 0 00 1853 1922 3200 18 00 1400 2909 3273 84 21 94 74 73 68 21 35 3101 56 0 00 0 00 1853 1925 52 00 2600 26 00 4000 40 00 100 00 100 00 100 00 21 53 2718 38 0 00 0 00 185
198. that the time space diagram identifies the tool used to develop timings displayed Also note that PASSER II and GA Optimizer solutions to the same arterial analysis will be different but they will usually have similar splits and cycle lengths if a Bandwidth Based Fitness Routine is used by the GA Optimizer 83 Optimization Analysis Tools Select PASSER II GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis la Diagram Summary Report Detailed Report Artery List Timing Source GA Based Model Artery 1 Time meee Show All Sub Arts Gere fase Redraw Hide All Sub Arts Efficiency 45 3 Attainability 86 5 Ager EB EB Band 37 0sec EB Attain 841 Print aiina WB Band 40 0sec WB Attain 88 9 Atep 1 20 40 60 89 100 120 140 160 180 200 220 240 Artery 2 Ref Phase 6 No yj 73 3 32 2500 00 ft Artery 3 Ref Phase 6 No 33 3 78 2700 00 ft Artery 4 Ref Phase 2 No yj 74 3 30 1390 00 ft Artery 5 Ref Phase Note that if the artery you are currently analyzing contains a diamond interchange you cannot use the PASSER II tool Therefore you must use the GA Optimizer tool to have PASSER V generate optimal timings As stated earlier you may wish to optimize the interchange first probably using the PASSER III tool fix a subset i e phase sequence of its signal settings and then analyze the artery using the GA Optimizer Additional tools available beyo
199. the diamond interchange a subsystem and optimize it with the PASSER III tool Then by making the appropriate selections within the GA Optimizer for the overall arterial you can retain the diamond phasing sequence for the interchange while creating progression bands and optimizing signal settings for the other arterial intersections for the overall arterial 146 PASSER V 09 E PASSER V Accessory P5 07 Help Data SH1 95am p5i E File Yiew Window Help BEC belie Sa LILIES Select Panning Two Way One Way Delete Move System Subsystem Tools In Out Normal Oe x cy Aney 2 Diamond Interchange defined as a Subsystem x 9230 Y 1705 In this data set a subsystem that includes only the diamond interchange has been predefined As described before you can ask PASSSER V to show the defined subsystem on screen by using the Subsystem button from the function toolbar By doing so the group of links that compose the subsystem will be highlighted in red as shown in the figure on the next page 147 Aran S Artery 6 O mare gt O Next click the Tools button and select SH 195 from the Artery List Since this system is a collection of diamond interchanges and adjacent signalized intersection the only available optimization tool for either bandwidth or delay optimization is the GA Optimizer In addition three other analytical tools that apply to all types of facilit
200. the PASSER III tab under the Summary or Detailed Reports right click your mouse and select the desired cycle length and phasing sequence When you go back to the T Sp Diagram tab your selection s details will be displayed Note from the figure that the source of the time space diagram is PASSER HI e PASSER IPs least delay solution PASSER Ill GA Optimizer Volume Analysis T Sp Diagram Delay Cycle Analysis Input Summary Report Detailed Report Plot CSV File Title ie Ill Summary Report for Harvey Starve Vehicles Volume Time Blocked Enter Texas Length Diamond sec Phase Cycle Average Total Delay Stops sec veh veh hr Volume Exit veh hr sec hr veh cyc veh hr 45 Ext 3 Phase 0 00 0 00 1816 50 Ext 3 Phase 45 Basic 3 Phase 50 Basic 3 Phase 40 Basic 3 Phase 40 Ext 3 Phase 55 Ext 3 Phase 55 Basic 3 Phase 60 Ext 3 Phase 60 Basic 3 Phase 70 TTI 4 Phase Ext 3 Phase Basic 3 Phase TTI 4 Phase Ext 3 Phase TTI 4 Phase 17 21 2507 59 17 51 2509 04 17 57 2458 30 17 91 2471 62 18 01 2574 66 18 01 2574 66 18 24 2643 32 18 63 2619 99 19 38 2740 42 19 64 2725 56 20 21 2667 57 20 44 2759 87 20 70 2748 55 21 05 2620 03 21 41 2773 14 21 91 2589 39 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 1816 1816 1816 1816 1816 1816 1816 1816 1816
201. the genetic algorithm PASSER III Tool Similar in functionality to the PASSER HI program this tool is for isolated interchanges operating in three phase or four phase mode It performs an exhaustive search for cycle length splits and phasing sequence and it selects a timing plan that minimizes total delay Since it uses exhaustive search it provides access to timings and MOEs for all combinations of timings possible based on user selected options This tool accurately predicts delay for all types of traffic conditions as opposed to the PASSER III program whose delay model was applicable to undersaturated traffic conditions only GA Based Tool This tool uses a genetic algorithm to provide users the ability to time signalized arterials for maximizing arterial progression or for minimizing system delay Depending on the optimization type selected delay based or bandwidth based it uses either DAR or BAR for calculating the fitness values of population members during the optimization process Since each run of DAR performs multiple stages of more intense calculations than BAR delay minimization requires significantly more computational time than the bandwidth based optimization The ability of this tool to find a good timing plan requires that the user selects large enough values for population size and number of generations These values should be selected based on problem size number of signals and the number of variables cycle offsets sp
202. tialization step the analysis step applies the oversaturated routine for a specified number of cycles In the current version of PASSER V this number is fixed to two cycles DAR uses four types of movements external to external external to internal internal to internal and internal to external Undersaturated Routine This routine assumes undersaturated flow conditions irrespective of the actual conditions in the network The routine builds flows and queue profiles by applying an extended version of the delay difference of offset ADOF model used by PASSER III 27 This methodology is similar to TRANSYT 7F s link wise simulation model In the undersaturated step the analysis is conducted one link at a time Starting from the upstream link each link is simulated The upstream flow profiles are created and projected downstream At the downstream intersection the outflows and inflows are calculated and queue profiles and delay profiles are built This process is repeated for each link During this process the routine applies the TRANSYT 7F platoon dispersion model 14 Queue storage on each link at the end of one cycle is obtained by building an input output queue profile If the queue at the end of the cycle is greater than the queue storage space it is set equal to the storage space The throughputs and delays for upstream and downstream movements are then calculated using flow profiles calculated using internal logic The only exception is
203. timizer for arterial optimization You can also use the Volume Analysis T Sp Diagram and Delay Cycle Analysis tools that we first used when analyzing a diamond interchange To begin our arterial optimization we want to first find out what range of cycle lengths is most appropriate for our artery Click on the Delay Cycle Analysis tool and then click on Draw Notice that the minimum delay cycle is around 55 to 60 seconds This result is not surprising since the minimum delay cycle length that we found for each intersection is in the neighborhood of this value We will make use of this information to constrain our cycle length range for analysis from the default of 40 to 120 seconds down to 40 to 90 seconds Next we will use the Volume Analysis tool to find out how much throughput capacity we need to provide to avoid congestion problems on the artery S W Military After clicking on the Volume Analysis tab and then on Chart you can see that as long as we provide a cycle length that is greater than about 50 seconds we will have enough capacity at the arterial intersections to accommodate our volumes 136 Now that we know the cycle length range for our analysis click on the PASSER II tool and enter a cycle length range from 40 to 90 seconds We will use Volume Weighted Band under Options to have the PASSER II tool try and create progression bands in proportion to the amount of through traffic either eastbound or westbound on the art
204. tion when any approach first receives a green signal indication i e it takes motorists a few seconds to react to a green indication take their foot off of the brake and begin accelerating As 4 seconds is a good estimate of lost time we will leave this default value in place The next step is to advance to the Sat Flow Data tab i e the next tab within the Node Data window A screenshot of this tab folder is shown below The items we will need to check and or modify here include the peak hour factor the growth factor the ideal saturation flow rate and the truck heavy vehicle percentage 98 Node Data Export Intersections Controller Type Coord Phase Offset Reference Point Controller Id 3 Artery 1 at Artery 2 Pretimed Signal Zi Begin of Green Cycle Length fo Area Type Other gt NTCIP Offset Referencing fv Begin of Yellow Ofso S Timing Data Sat Flow Data Optimization Data Performance Analysis Controller Signal MOEs a k h 3 1 gt a h em g s 1908 11 100 10 10 10 100 10 100 i0 10 100 100 fioo 1 00 100 100 100 fioo 1 00 100 100 1 00 1 00 100 100 100 fioo 1 00 20 200 200 200 200 200 200 200 20 1900 00 1900 00 1900 00 1900 00 1900 00 1800 00 1900 00 1900 00 1300 00 1300 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1900 00 1300 00 1400 00 1400 00 1400 00 Update Ok
205. tion Research Board National Research Council Washington D C 2000 Traffic Engineering Handbook 4 Edition Institute of Transportation Engineers Prentice Hall Inc 1992 McShane W R and R P Roess Traffic Engineering Prentice Hall Inc 1990 Determining Vehicle Signal Change and Clearance Intervals ITE Technical Council Task Force 4TF 1 Institute of Transportation Engineers Washington D C 1994 Gordon R L R A Reiss H Haenel E R Case R L French A Mohaddes and R Wolcott Traffic Control Systems Handbook US Department of Transportation Federal Highway Administration Washington D C 1996 Traffic Detector Handbook Second Edition Institute of Transportation Engineers Washington D C 1997 NAZTEC Inc Internet site http www naztec com tecnotes tn3013 htm Site and page accessed on September 26 2002 150 13 14 15 16 17 18 19 20 21 22 23 24 25 26 2T 28 29 CORSIM Version 1 03 Users Manual Kaman Science Corporation Colorado Springs Colorado 80933 7463 April 1997 TRANSYT 7F User s Guide Methodology for Optimizing Signal Timing MOST Volume 4 University of Florida Gainesville Florida March 1998 Husch D and J Albeck Synchro 4 0 User Guide Albany California 1999 Chang E C and C J Messer Arterial Signal Timing Optimization Using PASSER II 90 Program User s Manual TTI Report 467 2F College Stati
206. tion bracket would be between 4 30 PM and 6 30 PM TMC information i e number of vehicles turning left through and right is often collected using manual or computerized counters in 15 minute increments Each 15 minutes of data is transferred from the counting device to a computer program or a written sheet Data for all intersection approaches must be collected simultaneously so it is often necessary to have more than one technician at the intersection at one time each person counting two approaches Once the data have been assembled from each approach for the entire intersection a calculation is made from all approaches to determine which 15 minute periods have the highest volume The following table shows data from a four leg intersection with an intersection total volume in the right column The peak hour is determined from the highest four consecutive 15 minute periods In the table the highest peak hour is from 4 45 to 5 45 PM Data sheets may contain more data than are shown in the example table The example shows the minimum amount of data necessary to determine the peak period and perform an analysis An example of more detailed information would be a separate count for automobiles and trucks heavy vehicles Under each Left Thru and Right heading there would be two columns one for autos and one for trucks instead of the one column of numbers shown in the example Time Northbound Southbound Eastbound Westbou
207. tive the primary fitness value will be the delay to motorists resulting from a specific scenario Such an optimization model will evaluate the delay value for each alternative timing plan and select the timing plan that results in the least amount of delay In other words search algorithms are wrappers around traffic simulation or analytic models to provide the optimization function Search algorithms can be simple or extremely sophisticated Some of the common search algorithms are described below Exhaustive Search Algorithms As the name implies these algorithms calculate and compare the selected fitness values for all possible signal timing scenarios It should be noted that there can be millions of such combinations of signal timing parameters depending on the size of the facility and how many variables are to be optimized simultaneously Thus exhaustive search may require hours of computer time Unless a model is designed for small facilities the sheer number of possible scenarios usually requires the use of a divide and conquer strategy For instance computational time can be drastically reduced by stage wise optimization of each variable instead of all variables simultaneously and or by using a simple analytic or simulation model Such strategies increase computational efficiency by sacrificing accuracy The positive feature of exhaustive algorithms is that full information is available for each scenario analyzed Most optimization algo
208. tle or no traffic growth and traffic patterns are regular and predictable Downtowns and smaller towns not experiencing growth are typical locations for effective pretimed operation e Semi actuated Semi actuated or coordinated actuated operation uses detectors on non coordinated phases to offer more flexible use of green time than is possible in pretimed operation A fixed cycle length is still in force and the main street s through movements are also fixed with respect to when they must be present in the background cycle However the main street s left turn phases and the crossing street s through and left turn phases can all be skipped shortened or lengthened in comparison to pretimed operation depending on demand Unfortunately this added degree of freedom in managing green time is obtained by the use of detectors which must be installed wired back to the cabinet and maintained Semi actuated operation along an arterial roadway managed by a closed loop system and monitored by agency staff is a practical goal for state of the practice signal control Semi actuated operation is most appropriate along arterials that have a high volume with respect to crossing roadways Moderate changes in volume and traffic flow pattern are easily accommodated e Fully Actuated Intersections operating in fully actuated mode have no background cycle length Phase durations splits are determined by the number of vehicles that pass through the
209. tools use DAR for calculating performance measures for all internal movements and the HCM delay model for calculating delay for all external movements A description of each tool follows PASSER I Tool The PASSER II tool is applicable to a signalized arterial that contains no interchanges operating in three phase or four phase mode Similar in capability to the PASSER II program this tool allows the user to develop arterial signal timings for providing maximum arterial progression It performs exhaustive cycle length search in the user selected range ower limit upper limit and increment and maximizes bandwidth efficiency for each cycle using the interference minimization algorithm Because of this feature a timing plan is available for each cycle length By default this tool reports best timings and MOEs for the optimal solution However the user can request the tool to report best timing plans and MOEs for all cycle lengths In the latter case the tool displays a summary of MOEs for all plans It also allows the user to load into memory any selected timing plan to view the detailed results The PASSER II tool is capable of providing perfect one way progression for a selected direction or two way volume weighted bands In addition the user can request the tool to fine tune offsets to further minimize delay without affecting progression bands When a user selects this feature the program performs bandwidth constrained delay minimization using
210. ts It is the time in seconds between the departure of the vehicle at the head of the queue and the next vehicle in the queue using the same gap This parameter is analogous to the lost time at signalized intersections As opposed to the critical headway follow up time can be directly measured in the field The only implied requirement for obtaining this parameter from field observations is the presence of a queue Average follow up time must be obtained using a large enough sample size Calibration of Field Data If follow up time data for a movement is collected during a long enough period say 10 or 15 minutes during which the queue persists the count of vehicles leaving the stop bar will provide field measurement of movement capacity during that period Such data are extremely useful because they provide a way to calibrate the factor relating critical headway and follow up time The following steps are recommended 1 Record the data follow up times counts of vehicles leaving the stop bar and duration of persistent queue 2 Obtain average follow up time from data and critical headway using the above equation 3 Calculate movement capacity Count x 60 Data Collection Period Length 4 Lock use calculated follow up time and critical headway in PASSER V and compare capacity calculated by the program with that obtained in step 3 If needed adjust critical headway entered in PASSER V until the program calculated capacity matches fiel
211. ulation DAR also restricts outflow of any TWSC intersection movement affected by blocking due to queues at a downstream traffic signal 3 Lastly it adjusts capacities of any TWSC intersection movements whose capacities are reduced due to downstream blocking Impact on Optimization Tools PASSER I Tool This tool assists users in developing arterial signal timings for maximizing arterial progression During the optimization process the PASSER II tool ignores TWSC intersections by assuming that the through phases at these intersections have continuous green indications In other words it is assumed that the presence of TWSC intersections will not affect the progression bands It should be noted that all solutions from this tool are simulated using DAR to obtain performance measures Because DAR explicitly considers TWSC intersections the MOEs corresponding to each solution generated by this tool do account for such intersections 59 PASSER III Tool This tool applies to isolated signalized diamond interchanges It does not consider TWSC intersections GA Based Tool This tool uses a genetic algorithm to provide users the ability to time signalized arterials for maximizing arterial progression or for minimizing system delay Depending on the optimization type selected delay based or bandwidth based it uses either the delay analysis routine or bandwidth analysis routine for calculating the fitness values of population members during th
212. ve finished with this approach data entry step your Node Data dialog box window should look like the screenshot on the next page 92 Right Turn Geometry Queuing Condition Queue Does Not Block Queue Blocks Access to Access to Channel or Bay Channel or Bay No Right Turn Channel or Pay Full right turn hourly Nit volume less the number of right turn on red vehicles per hour Right Turn Channelized Right Turn Bay Lane yi Island NTS or Zero unless right turn volume is so high that right turn volume cannot be cleared as right turn on red If this is the case enter unsetved right turn on red volume in vehicles per hour If the right turn volume is zero according to this rule do not code a right turn movement for this approach Enter number of right turning vehicles per hour that ate blocked from the right turn channel by the queue Node Data Intersections Controller Type Coord Phase Controller Id 3 AreaType Other v Timing Data Artery 1 at Artery 2 Sat Flow Data Optimization Data Pretimed Signal Zi NTCIP Offset Referencing V Performance Analysis Controller Signal MOEs Begin of Green C Begin of Yellow Offset Reference Point Cycle Length 90 Offset 0S Optimization Settings Lock Sat Flows Lock Green Splits 2 gt lt 1 100 100 Prot Prot Lead Update Cancel 93 Volume
213. ve it before a complete analysis can be performed We will assume protected permitted operation on the crossing roadways and protected only operation on S W Military 131 Bay is i 1 161 long I I ith BY AM PM l i 1 AM pM L 9 35 l ANEM L 14 4 T 179 29 l AM PM L 79 7 7 a T 381 705 R 27 38 i L 56 101 T 145 161 aA IR A ag eA 2 T 329 615 R 128 204 J fay I R 14 10 4 i 3 0 Bay is my 140 long i Se Re ee sees Ad i SW ee 4 11 gt Milltafy s anm ei ecto eee O i Wo o o eee eee uae ape 13 AM PM L 246 115 T 610 762 L he D7 R 18 15 j L 1 T 231 159 2 2 Pa i T 184128 3 1 i IIR 7 104 A R 46 80 Se SG 4 2 2 a Z i 12 12 42 l Bay is i iy l 168 long 1 1 New Laredo f Somerset Highway Bay is 145 long Assume all lanes at Somerset are 12 wide Another piece of information that is essential to an arterial multiple intersection analysis is the average speed along the segment of roadway that links the analyzed intersections If a speed that is too high is used in the analysis the offsets will be set so that the platoon arrives late in the downstream intersection s main street through phase If the speed entered is too low the resulting offset will cause the platoon to arrive on red A good approximation of the average speed is the speed limit Field based speed limit or average speed information must be obtained before appropriate steps can be taken
214. ve right turns onto an arterial that you are analyzing in PASSER V for optimization it is important to include at least some right turn volume In our case we will include just enough of our northbound right turning 133 traffic that the right turn volume does not influence signal timing at the intersection A simple way to do this is to take the through volume for the same approach and divide it by the number of through lanes In our case this is 128 through vehicles divided by 2 lanes for a per lane through volume of 64 vehicles per hour Apply the right turn factor of 0 85 from the HCM to this value to produce 54 64 X 0 85 and use this value as the number of right turns This calculation includes some right turn volume for use in flow and delay calculations in PASSER V but not so much right turn volume that the right turn lane influences the split time for the through phase For the eastbound approach at New Laredo Highway the right turns do not turn onto a coordinated arterial so we can simply not code a right turn movement for this approach and or set this right turn volume to zero Le all rights can be made through the channelized right turn lane as right turns on red For the southbound right turns the rightmost lane will mainly be carrying only right turning traffic and at least half and perhaps up to 75 percent of right turns will get through on red For the westbound approach you might want to halve the right turning volume
215. vidual s fitness such as bandwidth or delay tournament selection a number of individuals are picked using roulette wheel selection then the best of these are chosen for mating and rank selection pick the best individual every time Population Size The size of the population in each generation quite often affects the solution A population size of five to a population size of tens of thousands is used depending on the evolutionary strategy and the nature of the problem that one is trying to solve In a solution space of NV possible solutions a population of N individuals can solve the problem in one generation however N is often far too big or unknown to do that Solution space affects the population size hence multiple runs need to be conducted for each kind of problem to select the optimal population size Termination Criteria GAs ate terminated using two criteria 1 convergence and 2 number of generations In PASSER V convergence is defined as the ratio of the average score of N previous best generations to the score of the current best of generation One can also define the maximum number of generations after which the GA evolution should stop 34 Crossover Probability Crossover probability is the probability that two parents mate An appropriate probability will allow parents to mate and thus make possible the search of new solution spaces In effect evolutionary techniques are most useful for problems where the
216. will be displayed with the opposing direction s protected left turn arrow The phasing sequence leading up to the yellow trap is depicted below Southbound Demonstration of Lead Lag YELLOW TRAP SYA ow hw Ne SA OM WN Northbound One solution to the yellow trap problem associated with protected permissive lead lag phasing is to use Dallas phasing which maintains the permissive left turn from the leading direction until the opposing direction s through movement terminates This type of operation is shown in the following figure Southbound Dallas Phasing SA On BW NR SAB uU Nel Northbound Note that the use of protected permissive lead lag phasing is not the only phasing situation that can create the yellow trap problem In semi actuated or fully actuated operation the skipping of the cross street phases can also result in situations where a permissive left turn phase in one direction is terminated while the opposing direction s through movement remains green To remedy these situations a minimum recall can be placed on cross street phases or phase inhibit functions can be used to ensure that the phasing situation that causes the yellow trap situation does not occur For instance using phase 2 to inhibit phase 1 phase 4 to inhibit phase 3 phase 6 to inhibit phase 5 and phase 8 to inhibit phase 7 you can eliminate the sequence of phase indications that result in a yellow trap However it is importan
217. y Click on the Controller tab to view the ring structure for this intersection and do the same for the Somerset intersection Always be aware of all of the safety considerations discussed in this manual when you make final decisions about what type of timing plan you implement in the field For instance our optimal timing solutions included lead lag phasing for the S W Military arterial If we had allowed permitted left turns from S W Military we could have created a yellow trap safety consideration Before implementing any timing solution you will want to review all pertinent phasing and configuration information to ensure that your recommendation is free of potential safety concerns and fully and appropriately meets the needs of the vehicles and pedestrians using the intersections If you do find safety concerns you can always make changes in your input file to remove those concerns and rerun your arterial in PASSER V Signalized Arterial with TWSC Intersections Now we are ready to analyze a signalized arterial with a TWSC intersection For this exercise we will use a data set named Arterial TWSC p51 that was created by adding a TWSC intersection between the two signalized intersections of the artery you just created The following screen capture shows the sketch of this arterial displayed by the program when this data file is opened 143 PASSER V 09 E PASSER V Accessory5 0 Help Data rtertal IWSC p5i Ele yow
218. y that your timings 62 should be longer i e require a slightly longer cycle length or longer splits for approaches that peak heavily than otherwise indicated by the total peak hour volume Key Point The peak hour factor is a measure of how much volumes vary within the peak hour When the PHF is less than 0 85 you should account for volume variations within the peak period when you are computing your signal timings Average Daily Traffic ADT information is a valuable resource for checking the accuracy of peak hour TMCs checking for the location of the peak hour for each intersection approach and monitoring increases in overall traffic volume over time The figure below shows a hypothetical intersection and hourly counts that develop the one way ADT counts for each intersection approach roadway Time NB ADT SB ADT EB ADT WB ADT 12 lam 0 0 0 0 123 o o o 0 SB ADT 2 3 0 0 o o 3 4 o o o o 4 5 0 o o 0 5 6 50 50 50 50 627 100 100 00 00 7 8 200 200 200 200 WB ADT amp 9 200 200 200 200 9 10 100 100 00 00 10 11 50 50 50 50 11 12 50 50 50 50 12 1pm 00 00 00 00 1 2 50 50 50 50 EB ADT 2 3 50 50 50 50 3 4 100 100 00 00 4 5 200 200 200 200 5 6 350 350 350 350 6 7 200 200 200 200 7 8 50 50 50 50 NB ADT 8 9 50 50 50 50 9 10 50 50 50 50 10 11 o 0 o 0 11 12 0 o o 0 TOTAL 1950 1950 1950 1950 Examination of the hourly figures used to calculate ADT confirms t

User Guide Only - passer

Contents

Download Pdf Manuals

Related Search

Related Contents