The National Academies Press

Currently Skimming:

Pages 344-370

The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.

From page 344... ... I-1 A p p e n d i x i Transit Travel Time Analysis Contents I-2 Transit Path-Building Assumptions I-7 Transit Trip Characteristics I-9 Preparing the Data for Comparisons to Modeled Component Travel Times I-20 Respondent Perceptions of Travel Time Compared to Schedule-Based Paths I-11 Respondent Perceptions of Travel Time Compared to Travel Model Paths I-25 Summary of Findings Transit ridership forecasts generated by regional travel demand models are dependent on estimates of routing, travel time, and cost generated by network path-finding software. These estimates are key inputs to several travel forecasting model components, including: • Transit trip assignment procedures that determine boarding stop, alighting stop, and transit route • Mode split procedures that determine market shares for transit versus automobile and non-motorized modes, walk versus kiss-and-ride and park-and-ride access modes, and bus versus rail sub-modes • Destination choice procedures that develop matrices of total travel demand (trip tables) Read the entire page →
From page 345... ... I-2 Characteristics of premium Transit Services that Affect Choice of Mode One possible explanation for reported times not matching simulated times is that survey respondents do not accurately portray their own travel time. Although this may occur in some cases, travel time reporting is one of easiest parts of the survey questionnaire and should not be assumed to be the sole cause of the mismatch. Read the entire page →
From page 346... ... Transit Travel Time Analysis I-3 the fall of 2007. Transit networks were validated by comparing modeled bus running times to timetable information and by assigning survey-derived transit trip tables to confirm that overall assigned survey trips generate ridership by station and route that match observed values. Read the entire page →
From page 347... ... I-4 Characteristics of Premium Transit Services that Affect Choice of Mode Like most other travel forecasting models, paths are built from the zone centroid using access links that represent either average trip distances to any transit station/stop or specific FIGURE I-1. Sample zone systems for urbanized areas in Charlotte and Chicago. Read the entire page →
From page 348... ... Transit Travel Time Analysis I-5 Path-Weighting Path-weighting parameters for each model are presented in TABLE I-1. Both the Charlotte and Chicago models generally follow conventional practice with out-of-vehicle time being weighted between two and three times the weight assigned to in-vehicle time. Read the entire page →
From page 349... ... I-6 Characteristics of premium Transit Services that Affect Choice of Mode Consistent with many of the transit forecasting models employed today, the Charlotte model assumes that all egress from the transit alighting location to the ultimate attraction zone occurs by the walk mode only. This means that automobile egress to the attraction end of the trip is not considered by the current Charlotte model transit path-builder. Read the entire page →
From page 350... ... Transit Travel Time Analysis I-7 The Charlotte schedule path-builder operates on an origin-destination basis with a specific representation of trip departure time. It includes the capability to build paths for each of the following: • Walk access, walk egress • Kiss-and-ride access, walk egress • Park-and-ride access, walk egress • Walk access, kiss-and ride egress • Walk access, park-and-ride egress Transit Trip Characteristics TABLE I-2 presents an overview of the mode shares and transit trip characteristics for the survey responses collected in Charlotte and Chicago. Read the entire page →
From page 351... ... I-8 Characteristics of premium Transit Services that Affect Choice of Mode TABLE I-2. Characteristics of surveyed trips in Charlotte and Chicago. Read the entire page →
From page 352... ... Transit Travel Time Analysis I-9 TABLE I-3. Reported Charlotte Area Transit System (CATS) Read the entire page →
From page 353... ... I-10 Characteristics of premium Transit Services that Affect Choice of Mode TABLE I-4. Summary of usable records for survey-model comparisons in Charlotte. Read the entire page →
From page 354... ... Transit Travel Time Analysis I-11 failure of the model to represent reported behavior, these records were analyzed in more detail to understand why paths were not available. Key findings include: Nearly 20% of the transit paths not found by the transit path-builder (9 out of 52 for Charlotte and 23 out of 119 for Chicago) Read the entire page →
From page 355... ... I-12 Characteristics of premium Transit Services that Affect Choice of Mode Transit In-Vehicle Travel Times FIGURE I-2, FIGURE I-3 and FIGURE I-4 present a comparison of reported and modeled transit IVTT for walk and drive access trips in Charlotte and for total trips in Chicago. Survey observations are color coded according to mode reported to be used during the trip: • Blue -- bus only (trip contains bus time but no rail time) Read the entire page →
From page 356... ... Transit Travel Time Analysis I-13 FIGURE I-3. Comparison of reported and modeled IVTTs for Charlotte drive access transit trips. Read the entire page →
From page 357... ... I-14 Characteristics of premium Transit Services that Affect Choice of Mode In Charlotte, this headway represents the entire 2.5 hour a.m. peak period between 7:00 a.m. Read the entire page →
From page 358... ... Transit Travel Time Analysis I-15 FIGURE I-6. Comparison of reported and modeled wait times for Charlotte drive access transit trips. Read the entire page →
From page 359... ... I-16 Characteristics of premium Transit Services that Affect Choice of Mode synthesizes walk access time as part of the demand models, and skimmed values of walk access time are not generated. The Charlotte model estimates walk access time by connecting zone centroids to transit stations and stops using highway centroid connectors and highway network links. Read the entire page →
From page 360... ... Transit Travel Time Analysis I-17 FIGURE I-9. Comparison of reported and modeled total walk times for Charlotte drive access transit trips. Read the entire page →
From page 361... ... I-18 Characteristics of premium Transit Services that Affect Choice of Mode FIGURE I-10. Comparison of reported and modeled drive access time for Charlotte transit trips. Read the entire page →
From page 362... ... Transit Travel Time Analysis I-19 FIGURE I-12. Comparison of reported and modeled total travel time for Charlotte peak period drive access transit trips. Read the entire page →
From page 363... ... I-20 Characteristics of premium Transit Services that Affect Choice of Mode FIGURE I-14. Comparison of reported and modeled total travel time for Charlotte off-peak period drive access transit trips. Read the entire page →
From page 364... ... Transit Travel Time Analysis I-21 The schedule-based path-building procedures differ from conventional travel model skims in that they attempt to represent, as faithfully as possible, the characteristics of a transit trip as experienced by each specific traveler beginning at a precise origin, ending at a precise destination, and departing at a specific time. Key aspects of this process that are different from typical travel model path-finding algorithms include: • Origin and destination locations are based on the reported survey latitude and longitude for each point. Read the entire page →
From page 365... ... I-22 Characteristics of premium Transit Services that Affect Choice of Mode FIGURE I-15. Comparison of travel model- and schedule-based estimates of transit travel time. Read the entire page →
From page 366... ... Transit Travel Time Analysis I-23 As shown in TABLE I-6, approximately one-quarter of all trips were discarded because the manual review concluded that the trip records that contained inconsistencies between origin location, destination location, and path that were unlikely to occur as reported. This percentage is similar to the fraction of trips that had no modeled skim time (19%) Read the entire page →
From page 367... ... I-24 Characteristics of premium Transit Services that Affect Choice of Mode TABLE I-7. Frequency of survey records where automatic and manual paths best replicated reported paths and reasons for differences. Read the entire page →
From page 368... ... Transit Travel Time Analysis I-25 Not surprisingly, the manual paths introduced different characteristics in terms of total time and each constituent time component (e.g., access time, in-vehicle time, and waiting time) Read the entire page →
From page 369... ... I-26 Characteristics of premium Transit Services that Affect Choice of Mode FIGURE I-16. Comparison of automatic and adjusted schedule travel times by travel time type. Read the entire page →
From page 370... ... Transit Travel Time Analysis I-27 FIGURE I-17. Comparison of reported and adjusted schedule travel times by travel time type. Read the entire page →

From page 344...

... I-1 A p p e n d i x i Transit Travel Time Analysis Contents I-2 Transit Path-Building Assumptions I-7 Transit Trip Characteristics I-9 Preparing the Data for Comparisons to Modeled Component Travel Times I-20 Respondent Perceptions of Travel Time Compared to Schedule-Based Paths I-11 Respondent Perceptions of Travel Time Compared to Travel Model Paths I-25 Summary of Findings Transit ridership forecasts generated by regional travel demand models are dependent on estimates of routing, travel time, and cost generated by network path-finding software. These estimates are key inputs to several travel forecasting model components, including: • Transit trip assignment procedures that determine boarding stop, alighting stop, and transit route • Mode split procedures that determine market shares for transit versus automobile and non-motorized modes, walk versus kiss-and-ride and park-and-ride access modes, and bus versus rail sub-modes • Destination choice procedures that develop matrices of total travel demand (trip tables)