Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
⢠Stop- location log sums (behaviorally interpreted as density of stop attractions on the way to and from the primary destination) are used as variables in the stop- frequency model; these log sums are calculated for each half tour and take into account stop- location access by the chosen tour mode and transit submode. The tour- based structure imposes some problems in the way that model outputs can be compared for the base and build scenarios. The core complication is that both trip distribution and mode choice stages are closely inter- twined and cannot be fully separated. Indeed, the former single- trip distribution stage is divided into primary- tour destination choice and stop- frequencyâlocation choice. The former mode choice stage is divided into entire- tour mode choice and trip mode choice. These choices are sequenced in a way that pure trip distribution and mode choice stages cannot be recombined in a simple trip for- mat. Consequently, the basic requirement of a fixed trip table cannot be met without some enforcement in the model chain. However, there are three possible construc- tive ways to meet the FTA requirements, at least on a partial basis, and to provide meaningful inputs for UB calculation that can be processed by SUMMIT. These three options are outlined in Table 1. Currently, the first (simplest) option has been adopted for the MORPC model and implemented programmati- cally. According to this approach, a full microsimulation model (with several global iterations that include all steps) is run for the base scenario. Then, all tours are fixed with their primary destinations, and the build scenario is run for several iterations, with the inclusion of only mode, stop fre- quency, and stop location choices, as well as assignments. The model output directly used for the UB calculation relates to the tour- level mode choice statistics only. Tech- nically, it is similar to the conventional model output in which trip units are replaced with tour units. The impact of other choices (stop frequency, stop location, trip mode choice) is taken into account implicitly through the over- all iterative equilibration of travel times and cost and upward log sums included into the tour- level mode choice utilities. The individual- record format can be converted into the quasi- aggregate format that corresponds exactly to the conventional SUMMIT input. The conversion is based on the following rules: ⢠All tour records with identical production zone, attraction zone, socioeconomic market segment, travel purpose, and time of day are collapsed into a single (aggre- gate) record with the corresponding values for these fields. ⢠The other fields for the aggregate record are processed in the following way: â Person trips are totaled across the aggregated records. â Fractions of trips that have a walk- to- transit path and drive- to- transit- only, as well as walk- to- transit and drive- to- transit- only shares, are aver- aged across the aggregated records (which makes their values fractional). 35APPLICATION OF A MICROSIMULATION MODEL TABLE 1 Three Options for Comparison of Transit Alternatives Model StageâFeature Option 1 Option 2 Option 3 Tour generation Fixed Fixed Fixed Tour primary destination Fixed Fixed Fixed Tour time of day Fixed Fixed Fixed Tour mode Variable Variable Variable Stop frequency Variable Fixed Fixed Stop location Variable Variable Fixed Trip mode choice Variable Variable Variable Tour OD tables by TOD periods Fixed Fixed Fixed Trip OD tables by TOD periods Variable Variable but totals are fixed Fixed Report to SUMMIT Simplified tour-level mode Simplified tour-level mode Full tour-level mode choice log sum choice log sum w/o choice log sum w/o with LOS variables reflecting intermediate stops intermediate stops frequency and location of intermedi- ate stops Included components Differences in mode utilities Differences in mode utilities Full differences in entire-tour mode w/o stops w/o stops utilities including LOS variables associated with making stops Ignored components Additional LOS components Additional LOS components associated with making stops; associated with making stops; a certain incomparability of a certain (but less significant) LOS variables for alternatives incomparability of LOS with different number of trips variables for alternatives with the on the tour and different same number of trips on the tour stop locations but different stop locations