TRANSYT-7F Release 10

HCS™ > TRANSYT-7F™ > TRANSYT-7F Release 10 (2004-2006)

About TRANSYT-7F™ Release 10

TRANSYT-7F Release 10 introduced several important features listed below:

  • Multi-Period Optimization
  • Direct CORSIM Optimization
  • Time-Space Diagram Screen
  • New Output Reports
  • Lane Configuration Screen
  • Saturation Flow Rate Calculator
  • “Profile View” Static Graphics

Please contact McTrans with any new issues raised that may be added to this list.

  1. When coding link connections (on the Map View or the Edit > Feeders screen), automatic flow balancing does not work properly if the summation of upstream left-turn and right-turn volumes exceeds the downstream thru volume. In this situation, it might be necessary for the user to manually code feeding flows in the Feeders screen.*
  2. When importing a CORSIM input file into T7F10, two-way stop-controlled (TWSC) intersections are not imported. The next version of TRANSYT-7F will allow the user to import TWSC nodes if desired, or import signalized intersections only.*
  3. In the time-space diagram screen, efficiency and attainability are not computed correctly when progression bands “wrap” at a two-way stop-controlled (TWSC) node, such that the “bottom” band is above the “top” band at the TWSC node.*
  4. In the Lane Configuration screen, if volumes are coded prior to lane channelizations, subsequent changes to the lane channelizations might automatically change the volumes. This might make it necessary to re-code some of the volume data (i.e., volume, peak hour factor, and turn bay storage).*
  5. In the Route Summary Report, the route-wide Disutility Index (DI) is not computed correctly for the first route in the data file. To get around this problem, the DI for route #1 could be computed manually, by taking the summation of link-specific DI’s within the route.*
  6. When importing a CORSIM input file into T7F10, T-intersections operating under actuated control are sometimes not assigned the correct approach direction (NB, SB, EB, WB). This does not affect the ability to perform direct CORSIM processing or optimization, but if the user tries to perform a macroscopic TRANSYT analysis, volumes and timings might be assigned to the wrong approach.*

*Indicates that this issue will be resolved in the next software release.

Please contact McTrans with any new issues raised that may be added to this list.

  1.  If “double-cycled” nodes are present within the network, optimization sometimes does not function properly. Contact McTrans technical support for assistance with this issue.*
  2. If the “Uncoordinated Network” checkbox is toggled on the Edit > Analysis screen, interconnected signals are still simulated with non-uniform flow profiles. If the user wishes to simulate uncoordinated nodes with uniform flow profiles, the “Coordinated Signal” checkbox on the Edit > Timing screen must be deactivated.*
  3. “Free” right-turns, which receive green time throughout the cycle, cannot be coded in the graphical timing screen (Edit > Timing) when a conflicting thru movement exists. The Timing View screen instead assumes right-turn-on-red (RTOR), and specifies that the right-turn is a permitted movement that must yield to the conflicting thru movement. Therefore, free right-turns must be coded using the non-graphical timing screen (Edit > Optional > Timing). Regarding left-hand drive, this issue applies to free left-turns, for which the Timing View screen assumes left-turn-on-red.*
  4. When a lane-by-lane analysis is in effect for dual or triple left-turn or right-turn lanes, the graphical lane configuration screen (Edit > Lanes) sometimes does not allow the user to properly change between shared and exclusive lane use. In release 10.2, this problem can avoided by temporarily disabling lane-by-lane analysis when changing between shared and exclusive lane use.*
  5. Double-cycled nodes are not portrayed properly by the Time Space Progression Diagram screen.*
  6. When importing HCS files, if right-turns from “adjacent” approaches (e.g., northbound and eastbound) move in the same phase, one of these right-turns might not be included in the TRANSYT timing plan. In this event, the user should open the Edit > Timing screen, and code the missing right-turn.*
  7. At two-way stop controlled (TWSC) intersections, the Timing View (Edit > Timing) screen sometimes codes and displays incorrect turn protection (protected vs. permitted) for minor-street movements. To get around this problem, the user can specify turn protection on the Edit > Optional > Timing screen.*
  8. When importing CORSIM (*.TRF) files containing coordinated actuated controllers, the timing plan might not be imported properly if phases 1&5 or 3&7 (typically the left-turn phases) were coded as “lag phases”. This issue has no effect on CORSIM processing, which simply runs CORSIM and post-processes the output. This issue also has no effect on direct CORSIM optimization, which uses minimum phase times from the TRANSYT (*.TIN) file, but ignores the rest of the TRANSYT timing plan. This issue only affects any potential conversion to macroscopic (TRANSYT-based) analysis.  If a companion macroscopic analysis is needed, the timing plan described above should be manually corrected within TRANSYT, by using the Edit > Timing screen.
  9. The Progression Diagram screen cannot handle progression routes that contain two-way stop controlled (TWSC) intersections.*
  10. HCS+ Signals module files can be imported into TRANSYT-7F, but only if they have an *.HXS extension. In reality, HCS+ Signals files have an *.XHS entension, so they can only be imported into release 10.2 if they are temporarily renamed to have an *.HXS extension.*
  11. When using metric units, the Route Summary Report shows maximum back of queue in units of feet instead of meters.*
  12. If an entire input data file has been marked as an “Uncoordinated Network”, genetic algorithm optimization of cycle lengths does not function properly.*
  13. If an intersection has been marked as an “Uncoordinated Signal”, mid-block source volume is not handled properly. This problem can be avoided by coding mid-block source volumes of zero at any uncoordinated signals.*
  14. HCS files are not imported properly when any yellow times are coded as zero seconds, or when there are (extra) period characters within the filename. HCS files with T-intersections are not imported properly when the stem of the T contains both exclusive left-turn and exclusive right-turn lanes.*
  15. Queue spillback is sometimes simulated from “secondary shared stopline” links. However, such links do not have their own queue length.*

*Indicates that this issue will be resolved in the next software release.

Please contact McTrans with any new issues raised that may be added to this list.

  1. When loading HCS2000-Signals files into TRANSYT-7F, the gap setting (a.k.a. unit extension, for actuated phases) is not imported. If the value within HCS is not 3.0 seconds (the TRANSYT-7F default), this could cause an unnecessary discrepancy in results, unless the user codes the correct gap setting into TRANSYT-7F manually.*
  2. User-coded queuing capacities are treated differently by the software, depending on whether link-wise simulation or step-wise simulation has been selected. If link-wise simulation has been selected, the program will always adjust the user-coded queuing capacity to reflect oversaturated conditions. If step-wise simulation has been selected, the program will never adjust the coded queuing capacity to reflect oversaturated conditions, but will always adjust the internally computed default value (in the absence of any coded value).
  3. When link-wise simulation is used, incremental delays for “secondary shared” links are computed using the 1985 HCM delay equation, instead of the HCM2000 delay equation.
  4. For certain complex timing plans, “effective green” is not reported correctly in the output file. This issue does not affect other results within the model. Effective green is only reported when the “detailed” output format has been selected.
  5. The start-up lost time (SLT) and extension of effective green time (EEG) can be calibrated for individual links, but this adjustment must be at least 1.0 seconds. For example, if the global SLT is 2.5 seconds, the SLT for a given link could not be modeled as 2.0 seconds or 3.1 seconds. It would have to be modeled as 2.5 seconds, or at least 3.5 seconds, or at most 1.5 seconds.*
  6. If “double-cycled” nodes are present within the network, hill-climb optimization does not function properly. Genetic algorithm optimization appears to handle double-cycled nodes properly, unless cycle length optimization has been requested.
  7. If the “Uncoordinated Network” checkbox is toggled on the Edit > Analysis screen, interconnected signals are still simulated with non-uniform flow profiles. If the user wishes to simulate uncoordinated nodes with uniform flow profiles, the “Coordinated Signal” checkbox on the Edit > Timing screen must be deactivated.
  8. Progression diagrams are sometimes not painted properly for one-way routes, or when phase #1 is coordinated, or when all-red clearance times straddle the cycle length boundary.*
  9. When optimizing splits via genetic algorithm, intersections with seven signal phases may cause a run-time error. This problem may be avoided by omitting seven-phase nodes from the optimization node list, or by selecting hill-climb optimization.*
  10. I-value calculations (within the random delay equation) are compliant with the 1997 HCM by default, and an option is available to select the HCM2000 I-value calculation. In release 10.2, the HCM2000 I-value calculation will be applied automatically.*
  11. In the Node Input Summary reports, modifications to the Start-Up Lost Time (SLT) and Extension of Effective Green Time (EEG) are reported in units of seconds. The actual units for these input parameters are sec*10. For example, a value of ’15′ would indicate a 1.5 second adjustment.*
  12. When using genetic algorithm optimization on a relatively fast computer, the program sometimes performs many simulation runs in a very short period of time, such that the computer doesn’t have enough time to reclaim memory. This can sometimes cause the computer to pause for several seconds or minutes before resuming optimization. Contact technical support for assistance with this issue.*
  13. Release 10.1 was the first version in which permitted-only links could be “critical”, in actuated estimation. However, saturation flow rates coded as zero cause a divide-by-zero when determining the number of lanes. The problem can be avoided by coding token saturation flow rates for permitted-only links, to reflect the number of lanes.*
  14. The actuated estimation model uses the global start-up lost time (SLT), and the global extension of effective green time (EEG). However, it’s calculations do not reflect any user-coded, link-specific modifications to the SLT’s and EEG’s.*
  15. When performing a multi-period analysis, if an oversaturated movement was undersaturated in the previous time period, random delay and random queue are underestimated for these movements.*

*Indicates that this issue will be resolved in the next software release.

Download Release 10.3 (8.8 MB, requires release 10.2)
Download Release 10.2 (10.9 MB, requires release 10.1)