Dynamic, Integrated Model System: Sacramento-Area Application, Volume 2: Network Report (2014)

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.

2The second Strategic Highway Research Program (SHRP 2) Project C10B, Partnership to Develop an Integrated, Advanced Travel Demand Model with Fine-Grained, Time-Sensitive Net- works, is an important step in the evolution of travel model- ing from an aggregate, trip-based approach to a completely dynamic, disaggregate methodology. In this project, an existing disaggregate activity-based model (ABM) was integrated with an existing traffic simulation model to create a new, completely disaggregate model. Both models were implemented using open source software. At the same time that travel demand models have been evolv- ing, traffic simulation models, which simulate the movements of vehicles through a highway network, have become more sophisticated due to improvements in computing. The product of SHRP 2 C10B is an integrated model that simulates individu- als’ activity patterns and travel and their vehicle and transit trips as they move on a real-time basis through the transpor- tation system. It produces a true regional simulation of the travel within a region, for the first time using individually simulated travel patterns as input rather than aggregate trip tables to which temporal and spatial distributions have been applied to create synthetic patterns. A unique feature of this model is the simulation of transit vehicles as well as individ- ual person tours using transit. The new integrated model has been developed and imple- mented for the entire Sacramento, California, region. The integrated model components include SACSIM, the regional travel model maintained by the Sacramento Area Council of Gov ernments (SACOG), the regional metropolitan planning organization (MPO), and DynusT (Dynamic Urban Systems for Transportation), a mesoscopic traffic simulation model developed by the University of Arizona. SACSIM includes an activity-based demand model, DaySim. The transit simula- tion is performed by FAST-TrIPs (Flexible Assignment and Simulation Tool for Transit and Intermodal Passengers), also developed by the University of Arizona. The integrated model also includes the ability to run MOVES, the air quality analysis program developed by the U.S. Environmental Protection Agency. While the C10B integrated model produces reasonable results for regional travel patterns and behavior, the true value of the model is its ability to provide analysis results that demonstrate sensitivity to policy variables more accurately than models that use aggregate demand or assignment proce- dures. This sensitivity was tested through a series of policy and project tests conducted by SACOG using the new inte- grated model and the existing SACSIM model with aggregate assignment. The SHRP 2 C10B project has been documented in a series of four reports: • SHRP 2 C10B summary report, • User’s manual for the integrated model, • Network report on SHRP 2 C10B version of DynusT and FAST-TrIPs, and • Software start-up guide. This report, the third in the series, describes the theoretical background and methodology for the integration of DynusT, a mesoscopic dynamic traffic assignment (DTA) model, and FAST-TrIPs, a public transit passenger assignment and simu- lation model. The DynusT and FAST-TrIPs integrated system is designed to be a “loosely coupled” system. This design allows the DynusT and FAST-TrIPs teams to develop and test com- ponents in parallel following separate software development cycles. In the SHRP 2 C10B project, the travel demand data are simulated by the DaySim ABM implemented in Sacramento, California. The overall FAST-TrIPs model structure is illustrated in Fig- ure 1.1, which depicts the software components as well as the files needed for the DynusT and FAST-TrIPs communication. First, data from the General Transit Feed Specification (GTFS), originally known as the Google Transit Feed Specification, are processed and converted to transit route layouts, stops, and C h a p t e r 1 Introduction

3 measures that result from the DynusT assignment become inputs to the FAST-TrIPs model in an iterative process until convergence is reached. As shown in Figure 1.2, DynusT and FAST-TrIPs maintain a loose coupling integration architecture. This integration archi- tecture is of particular advantage in reducing development route schedule input files used by FAST-TrIPs. The FAST-TrIPs assignment procedure accepts demand data from DaySim and roadway link travel times from DynusT and assigns travelers to transit paths and transit vehicles. The DynusT DTA model simulates the DaySim trip and tour rosters and the transit vehi- cles from the FAST-TrIPs component. The network-wide LOS Figure 1.1. DynusT–FAST-TrIPs integration framework. Figure 1.2. Interfacing mechanism in DynusT–FAST-TrIPs integrated system.

4risk. This method allows FAST-TrIPs, a completely new devel- opment from the C10B project, to be prototyped, developed, and tested as a stand-alone module, allowing robust and con- trolled testing and debugging along the development process. As a result, communications between DynusT and FAST-TrIPs are conducted via flat text files. In addition to the GTFS files, other inputs to the entire system include the DaySim output files containing trip and tour rosters (sout.dbf) and non-ABM trips, such as airport or exogenous origin–destination (OD) (exogenous.dbf and air .dbf, ixxi.dbf, etc.). The ft_BST script processes sout.dbf and produces demand_auto.dat. Demand_auto.dat contains the auto trip roster and is further processed into vehicle.dat and path.dat as the main input files for DynusT. The transit route and trip schedule information is organized and prepared as transitRouteSchedule.dat file from GTFS file format. DynusT takes both the airport demand (processed into demand.dat) and external freight traffic (processed into demand_truck.dat) as the trip roster (vehicle.dat and path .dat) and performs simulation and dynamic assignment until reaching convergence in auto demand. Once DynusT reaches convergence transit vehicle trajecto- ries (AltTime_transit.dat and AltTime_Intermodal.dat), combining demand_transit.dat are processed and fed into FAST-TrIPs for transit assignment. Essentially AltTime_ transit.dat and AltTime_intermodal.dat describe the transit vehicle movements, which can be considered as the supply- side information, whereas demand_transit.dat contains the demand-side information. The transit assignment procedure essentially determines the assignment of passengers to vari- ous transit route and transit stop choices. The outcome of FAST-TrIPs procedure determines how many passengers will get on and off at each stop, and conse- quently determines the transit vehicle dwell time at each stop (TransitDwellTime.dat). If FAST-TrIPs is deemed not con- verged, DynusT is rerun again with the updated transit dwell time as the important input to DynusT. On the other hand, if the FAST-TrIPs convergence criterion is met, the entire DynusT–FAST-TrIPs integrated procedure is considered completely converged, and the system will output necessary skim information in terms of skim.dat, skimXXX .dat. Such information will be fed into DaySim ABM. This report is structured as follows: A detailed description of the DynusT model is presented in Chapter 2; Chapter 3 describes the enhancements that were made to DynusT to work with the DaySim ABM model; and Chapter 4 provides an in-depth exposition of the theoretical models used in FAST-TrIPs.