Skip to content
HU DONG edited this page Jul 6, 2023 · 93 revisions

Scenario Specific Input Files

Scenario specific input files are saved in base_inputs folder. The following table lists the directories and files in base_inputs folder under the master model directory.

Name Type Description
Network
networks Directory Network inputs
vdfs Directory Volume-delay functions
Fares Directory Transit fare files
tolls Directory Toll inputs
IntraZonals Directory Input files for intrazonal travel time calculations
observed Directory traffic counts (link counts in peak hours)
extra_attributes Directory files of EMME extra attributes
Synthetic Population
popsim Directory synthetic population
Land Use
Land Use Directory Land use related files
accessibility Directory (created on the fly) Parcel accessibility (buffer) inputs
Park and Ride
pnr Directory Park-and-ride locations
Auxiliary Models
supplemental Directory Supplemental model inputs
bikes Directory Bike model inputs
trucks Directory Truck model inputs
shadow_pricing
shadow_pricing Directory shadow pricing for daysim if should_build_shadow_price is set to True.

Network

networks

The following table lists the network-related inputs:

Name Type Description Data Source
[tod]_roadway.in File Roadway network file by time of day (tod): am, md, pm, ni populate network Modeller tool
[tod]_transit.in File Transit network file by time of day (tod): am, md, pm, ni populate network Modeller tool
[tod]_turns.in File Turns network file by time of day (tod): am, md, pm, ni populate network Modeller tool
vehicles.txt File List of transit vehicles and characteristics populate network Modeller tool
modes.txt File List of modes and their characteristics populate network Modeller tool
sc_headways.csv File Transit headways by route and time period populate network Modeller tool
[tod]_linkshapes.in File Link shapes by time of day (tod): am, md, pm, ni populate network Modeller tool
transit_stops.csv File Transit stop location file populate network Modeller tool, for accessibility calculation.
zone_partitions.txt File Partition file populate network Modeller tool
rdly Directory Contains [tod]_rdly.txt files; road delays by time of day (tod): am, md, pm, ni populate network Modeller tool
fixes Directory Contains a subdirectory ferries that contains [tod]_roadway.in files; roadway network changes for ferries by time of day (tod): am, md, pm, ni manual edits

vdfs

The following table lists the text files (by time period) describing volume-delay functions (vdfs) for both roadway and transit assignments:

File Description Data Source
vdfs6to9.txt Volume-delay functions for AM period (6:00 a.m. to 9:00 a.m.) populate network Modeller tool
vdfs9to1530.txt Volume-delay functions for MD period (9:00 a.m. to 3:30 p.m.) populate network Modeller tool
vdfs1530to1830.txt Volume-delay functions for PM period (3:30 p.m. to 6:30 p.m.) populate network Modeller tool
vdfs1830to6.txt Volume-delay functions for NI period (6:30 p.m. to 6:00 a.m.) populate network Modeller tool

Fares

The following table lists input files providing transit fares, by transit zones:

File Description Format
am_fares_farebox.in AM off-peak farebox fares (2010 dollars); the structure of the file: origin zone, destination zone, fare($) EMME matrix by ensemble
md_fares_farebox.in MD off-peak farebox fares (2010 dollars); the structure of the file: origin zone, destination zone, fare($) EMME matrix by ensemble
am_fares_monthly_pass.in AM peak monthly pass cost (2010 dollars); the structure of the file: origin zone, destination zone, pass cost($) EMME matrix by ensemble
md_fares_monthly_pass.in MD off-peak monthly pass cost (2010 dollars); the structure of the file: origin zone, destination zone, pass cost($) EMME matrix by ensemble
transit_fare_zones.grt Transit fare zones: gt1 (seattle ride free), gt2 (seattle), gt3 (king county), gt4 (everett transit), gt5 (snohomish county), gt6 (pierce county), gt7 (kitsap county), gt8 (kitsap - walk to ferry) EMME ensemble batchout file

tolls

The following table lists input files providing cost (tolls or fare) to use the roadway system and include ferries:

File Description Data Source
[tod]_roadway_tolls.in Roadway tolls by time of day (tod): am, md, pm, ni; the file does not have a header manual edit
ferry_vehicle_fares.in Fare for vehicles on ferry; the file does not have a header
bridge_ferry_flags.in Bridge and ferry flags (not being used?) manual edit

IntraZonals

This directory contains files for intrazonal travel time calculations.

File Name Description Format
destination_tt.in Terminal times (in minutes) destination/attraction trip end EMME vector format
origin_tt.in Terminal times (in minutes) origin/production trip end EMME vector format
taz_acres.in Zone area in acres EMME vector format

observed

File Name Description
@slcnt_pm_2014.txt 2014 screenline pm counts
@slcnt_md_2014.txt 2014 screenline md counts
@slcnt_am_2014.txt 2014 screenline am counts
@slcnt_pm_2018.txt 2018 screenline pm counts
@slcnt_am_2018.txt 2018 screenline am counts
@red_tubecnt_pm_2018.txt 2018 Redmond PM tube counts
@red_tubecnt_am_2018.txt 2018 Redmond AM tube counts
@red_tube_cnt_tag.txt Redmond tube count location
@local_cnts_pm_2014.txt 2014 PM local counts
@local_cnts_md_2014.txt 2014 MD local counts
@local_cnts_am_2014.txt 2014 AM local counts
@local_cnts_am_2019.txt 2019 AM local counts
@local_cnts_pm_2019.txt 2019 PM local counts
@local_cnts_am_2018.txt 2018 AM local counts
@local_cnts_pm_2018.txt 2018 PM local counts
ScreenLine_ID.txt Screenline locations
@kirkland_slid.txt Screenline tags for Kirkland's screenlines.

They are in EMME extra attribute batchout format, created by EMME modeler tool ExportAllAttributes.py.

extra attributes

They are in EMME extra attribute batchout format, created by EMME modeler tool ExportAllAttributes.py.

Land Use

File Name Description Format
distributed_jblm_jobs.csv JBLM jobs csv file
parcels_military.csv jobs in military base csv
parking_gz.csv ensemble gz - PSRC TAZ lookup table csv
lu_type.csv Land-use type definitions (land use type data is not used in the Daysim model) csv

popsim

File Name Description Format
hh_and_persons.h5 synthetic population file including household and person data h5 format

seed_skims (No longer needed)

The seed_skims sub-directory contains seed skims for four time periods. These skims are used to speed up run time. The seed skims can be updated by simply copying them from a previous model run and put in this sub-folder. When run_skims_and_paths_seed_trips is set to False and run_copy_seed_skims is set to True in input_configuration.py, the model copies these skims to the inputs folder for DaySim to use. Alternatively, if you do not want to use seed skims, set run_skims_and_paths_seed_trips to True and the model will build an initial set of skims (travel times, costs and distances) using DaySim seed trips.

File Description
6to9.h5 Skims for AM period (6:00 a.m. to 9:00 a.m.)
9to1530.h5 Skims for MD period (9:00 a.m. to 3:30 p.m.)
1530to1830.h5 Skims for AM period (3:30 p.m. to 6:30 p.m.)
1830to6.h5 Skims for AM period (6:30 p.m. to 6:00 a.m.)

accessibility

The accessibility folder has four files.

File Name Description Format
all_streets_links_2014.csv all street network link file
all_streets_nodes_2014.csv all street network node file
daily_parking_costs.csv Updated daily parking cost in EMME ensemble format
hourly_parking_costs.csv Updated hourly parking cost in EMME ensemble format

pnr

File Description
p_r_nodes.csv Park-and-ride locations and capacities

This file is for park and ride lot/path choice in DaySim.
Relevant DaySim configuration parameters:

  • RawParkAndRideNodePath (string file name and path)
  • RawParkAndRidePathDelimiter (ASCII code, 9=tab, 32=space, 44=comma)
  • ShouldReadParkAndRideNodeSkim (if true, DaySim will get park and ride mode LOS from zone-to-zone skims listed in the Roster, rather than using lot/path choice)
  • MaximumMilesToDriveToParkAndRide (to save run time, a limit on the drive search radius for park-and-ride lot/path choice)
  • MaximumRatioDriveToParkAndRdieVersusDriveToDestination (to save run time and avoid illogical lot choices, a limit on the ratio of the drive access distance to park-and-ride and the drive access distance to the final destination)

supplemental

Supplemental module was updated in 2022. Click here to read the supplemental module input files.

bikes

Click here for bike model input files.

trucks

File Description
districts19_ga.ens
input_skims.txt
truck_inputs.csv
truck_matrices.csv
truck_time_of_day_factors_bkr.csv

Shadow Pricing

This directory contains "shadow_pricing.txt" file. This file, if should_build_shadow_price is set to False, is used as a starting point for shadow prices in order to obtain stable results. Fixed starting shadow prices is also helpful in replicating an activity-based run with the same inputs. The shadow prices are intermediate outputs of a DaySim run. In absence of initial shadow prices, set should_build_shadow_price is to True and the model will create shadow prices from scratch. The shadow prices should be built from scratch when the model is run with new sets of inputs. Once a model run produces satisfactory results, the shadow prices from that run can be used as starting shadow prices to future runs get stable results.

Shadow prices are used to constrain employment and school enrollment in a parcel by its actual capacity. In every feedback loop, BKRCast runs three iterations of DaySim with only work and school location choice models to create stable shadow prices. At each iteration of running the work location model, the shadow price for a particular parcel (or the corresponding TAZ if the number of workers in the parcel is smaller than a user-specified threshold) is incremented by an amount equal to LN(Jobs in the land use data / Workers predicted to work there)*StepSize, where StepSize is also a user-specified input. There are other user-specified threshold-values that tell DaySim what is “close enough”, so that it doesn’t try to match everything exactly.

General Input Files

Model specific files are saved in inputs/ folder.

Name Type Description Location
DaySim
parcels_urbansim.txt File Primary parcel file; this is an input to accessibility calculations that produces a buffered parcel file for DaySim parcels_file_folder
node_index_2014.txt File Node index file inputs/model
node_to_node_distance_2014.h5 File Node-to-node shortest distances inputs/model
parcel_nodes_2014.txt File Nearest all-street nodes for parcels inputs/model
bkr_worker_ixxifractions.dat File Zone-based shares of workers who are in-commuting and out-commuting outputs/landuse
TAZIndex.txt File Zone index file inputs/model
hh_and_persons.h5 File Synthetic population base_input/popsim
TAZ_District_CrossWalk.csv File Zone-to-district correspondence inputs/model
bkr_roster.csv File DaySim impedance roster file inputs/model
bkr-roster.combinations.csv File Input for mode/path-type combinations in DaySim inputs/model
coefficients Directory DaySim coefficient files inputs/
shadow_pricing Directory Starting shadow prices inputs/
etc Directory Other inputs; see trips and household survey file inputs/
skim_params Directory Parameter settings for skimming and assignment inputs/

DaySim

parcels_urbansim.txt

The accessibility calculations use this parcel file and produce a buffered parcel file (buffered_parcels.dat) that is then used by DaySim. The relevant DaySim configuration parameters for the buffered parcel file include the following:

  • RawParcelPath (string file name and path)
  • RawParcelDelimiter (ASCII code, 9=tab, 32=space, 44=comma)
  • UseShortDistanceNodeToNodeMeasures (set True if node distance files are available)
  • UseShortDistanceCircuityMeasures (set True if circuity measures are available on the parcel file; the NodeToNodeMeasures setting takes precedence over this one)
  • MaximumBlendingDistance (the maximum distance, in miles, to which the short-distance measures are used to adjust zone-to-zone skim-based travel times)

File format can be found here.

node_index_2014.txt

Relevant DaySim configuration parameters:

  • NodeIndexPath (string file name and path)
  • NodeIndexDelimiter (ASCII code, 9=tab, 32=space, 44=comma)

The file format can be found here.

Node-to-Node Distance File (node_to_node_distance_2014.h5)

DaySim is designed to take advantage of more accurate distances for short distance on-street trips than are available from the TAZ-to-TAZ road network skims. These more accurate shortest-path node-to-node distances are calculated from an “All Streets Network” which is a network (as the name indicates) that consists of all the streets in the model region. A typical model network would only consist of freeways, expressways, and major and minor arterials. This coarser network cannot accurately represent the accessibility between small geographies such as parcels. These short distance microzone-level impedances are most important for non-motorized trips such as bike and walk trips. Thus, facilities such a freeways and expressways, freeway ramps, and other facilities that are not available to non-motorized modes are excluded from the all-streets based network path building. In addition, short driveways (if present) are also typically dropped from the all-streets network in order to reduce the computation burden. The modified all streets network is then used to generate node-to-node shortest path distance. These short distance impedances are directly used by the Daysim choice models, in conjunction with more typical TAZ-based impedances.

The node to node distance file needs to be updated only when there are significant changes in the all-street network (more or fewer links). As currently the node-to-node distances are created using based on shortest distance path (link length), the link attributes other than length do not matter. PSRC has a process to generate the distance file, so if an updated all-street network is available, the process can be run and an updated distance file can be generated.

This file is in HDF5 format.

Relevant DaySim configuration parameters:

  • NodeDistancesPath (string file name and path)
  • NodeDistancesDelimiter (ASCII code, 9=tab, 32=space, 44=comma, 0=binary)

The file format can be found here.

Parcel-to-Nodes File (parcel_nodes_2014.txt)

This file is ASCII delimited with a header record.

Relevant DaySim configuration parameters:

  • RawParcelNodePath (string file name and path)
  • RawParcelNodeDelimiter (ASCII code, 9=tab, 32=space, 44=comma)

The file format can be found here.

IXXI Worker Fractions (bkr_worker_ixxifractions.dat)

This is a file that DaySim uses for the work location model. This is the file that is used to set the percentage of workers living in each zone that work outside the region and the percentage of jobs in each zone that are filled by workers living outside the region. DaySim does not select a usual workplace or simulate internal home-work tours for the I-X fraction of workers, and makes the X-I fraction of jobs unavailable as usual work locations for regional residents. Both fractions tend to be larger toward the edges of modeled regions.

The file is an ASCII delimited file with optional header.

The relevant DaySim configuration parameters are:

  • IxxiPath (string file name and path)
  • IxxiDelimiter (ASCII code, 9=tab, 32=space, 44=comma)
  • IxxiFirstLineIsHeader (if true, DaySim requires a header record)

The file format can be found here.

Zone File (TAZIndex.txt)

This file indicates to DaySim which zone numbers are valid. If there are gaps in the zone numbering (unused zone numbers), then this file is used to set up a mapping from the nominal zone numbers to an internal zone numbering that is used to reduce the amount of memory required to save zone-to-zone skim matrices.

The file is ASCII delimited with a header record.

Relevant DaySim configuration parameters include the following:

  • RawZonePath (string file name and path)
  • RawZoneDelimiter (ASCII code, 9=tab, 32=space, 44=comma)

The file format can be found here.

Zone-to-District Correspondence (TAZ_District_CrossWalk.csv)

This file is used by the popsampler process to identify zones corresponding to a district. Seven districts are identified: BKR, Kitsap, Pierce, Rest of King, Seattle, and Snohomish.

This is a comma-delimited file with a header record. The file format can be found here.

DaySim Roster File (bkr_roster.csv)

The roster file is a convenient and flexible way to define which skim matrices are used for all modes/path types and level-of-service (LOS) variables for all times of day.

The file is a CSV file with a header record. Its file format can be found here.

DaySim Roster Combinations File (bkr-roster.combinations.csv)

The roster combinations file is used with the Roster file to determine which mode-path type combinations are valid in the Roster file. A “path type” is essentially a “submode” in DaySim, such as tolled versus nontolled networks for auto modes or local bus versus light rail (plus optional bus) networks for transit. The file has columns that enumerate the nine modes used in the current model system (walk, bike, SOV, HOV2, HOV3, transit, park-and-ride, school-bus, and other) and seven rows that enumerate the path types currently used (full-network, no-tolls, local-bus, light-rail, premium-bus, commuter rail, ferry). The cells are TRUE for valid combinations within DaySim and FALSE otherwise.

Here is an example of the file:

# Walk Bike SOV HOV2 HOV3 Transit Park-and-Ride School Bus Other
Full-Network TRUE TRUE TRUE TRUE TRUE FALSE FALSE TRUE FALSE
No-Tolls FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE
Local-Bus FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
Light-Rail FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
Premium-Bus FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
Commuter-Rail FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
Ferry FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE

Coefficients

This directory contains 23 files providing coefficients for different DaySim models. The coefficient files are in *.F12 format and can be edited by the user for calibration purposes.

Here is a list of coefficient files in the directory:

Coefficient File Model
AutoOwnershipModel_psrcper1.F12 Auto ownership model
EscortTourModeModel.F12 Escort tour mode model
IndividualPersonDayPatternModel.F12 Individual person day pattern model
IntermediateStopGenerationModel.F12 Intermediate stop generation model
IntermediateStopGenerationModel_psrcper1.F12 Intermediate stop generation model
IntermediateStopLocationModel_psrcper1.F12 Intermediate stop location model
OtherHomeBasedTourModeModel.F12 Other home based tour mode model
OtherHomeBasedTourTimeModel.F12 Other home based tour time model
OtherTourDestinationModel.F12 Other tour destination model
PayToParkAtWorkplaceModel_psrcper1.F12 Pay to park at workplace model
PersonExactNumberOfToursModel.F12 Person exact number of tours model
SchoolLocationModel.F12 School location model
SchoolTourModeModel.F12 School tour mode model
SchoolTourTimeModel.F12 School tour time model
TransitPassOwnershipModel.F12 Transit pass ownership model
TripModeModel.F12 Trip mode model
TripTimeModel.F12 Trip time model
WorkBasedSubtourModeModel.F12 Work-based subtour mode model
WorkBasedSubtourTimeModel.F12 Work-based subtour time model
WorkLocationModel.F12 Work location model
WorkTourDestinationModel_psrcper1.F12 Work tour destination model
WorkTourModeModel.F12 Work tour mode model
WorkTourTimeModel.F12 Work tour time model
WorkbasedSubtourGenerationCoefficients_SACOG-v1.5.F12 Work-based subtour generation model

Skim and Assignment

skim_params

This directory contains settings for skimming and assignment.

File Name Description
time_of_day.json
time_of_day_crosswalk_ab_4k_dictionary.json
transit_fare_dictionary.json
emme_bank_dimensions.json
matrix_balancing_spec.json
temp_link_Calculation.json
bike_assignment.json
bike_network_setup.json
bike_skim_setup.json
bike_walk_assignment.json
bike_walk_matrix_dict.json
demand_crosswalk_ab_4k_dictionary.json
demand_matrix_dictionary.json
extended_transit_assignment.json
general_attribute_based_skim.json
general_generalized_cost_skim.json
general_path_based_assignment.json
general_path_based_volume.json
intrazonal_dictionary.json
link_calculation.json
matrix_calc_spec.json
matrix_calculation.json
node_calculation.json
tolls.json
transit_line_calculation.json
transit_line_definition.json
transit_segment_calculation.json
transit_skim_aggregate_matrix_names.json
transit_skim_setup.json
truck_model.json
user_classes.json

demand_matrix_dictionary.json

record example: (a, b, c):matrix_name
The value of dictionary is matrix name. The key is a tuple of three values:

  • a: trip mode
  • b: value of time (three levels of VOT, 1 if income < 15K, 2 if income < 25K, or 3 if income >= 25K)
  • c: trip path type

etc

This directory contains other inputs, such as seed trips for DaySim and survey files for DaySim outputs validation.

File Name Type Description Location
Media Directory EMME media files `inputs\model\Media
worksheets Directory EMME predefined worksheets `inputs\model\worksheets
survey.h5 File Household travel survey for DaySim estimation validation summaries 'inputs\model\survey`

Clone this wiki locally