input_configuration - bellevuewa/BKRCast GitHub Wiki
Overview
This file contains model run settings such as data directory, what components to run, and number of iterations, among others. It is recommended to start with the primary settings. The following is a summary of settings available in the file:
- Primary Settings
- Settings that must be updated for a new setup
- Subprocess flags
- Switches to turn model subcomponents on or off
- Feedback Loops Settings
- Settings related to feedback iterations
- Accessibility Calculations
- Settings for the accessibility calculations subprocess
- Input Checks
- Settings for input checks
- DaySim
- Settings related to a DaySim run
- Bike Model Settings
- Settings for the bike model
- Truck Model Settings
- Settings for the truck model
- Calibration/Validation Settings
- Settings for generating calibration and validation summaries
The following is a complete list of settings and their descriptions:
File Content
| Setting | Value | Description |
|---|---|---|
| Primary Settings | ||
| project_folder | 'E:/Projects/Clients/ bkr/ model/ bkrcast_tod_new' | BKRCast model directory; this needs to be updated for a new setup |
| parcels_file_folder | 'E:/Projects/Clients/ bkr/ model/ bkrcast_tod_new/ inputs/ 2014/landuse' | Primary parcel file location; this needs to be updated for a new setup |
| base_year | '2018' | Base year |
| model_year | '2018' | Model year |
| supplemental_module_base_year | '2018' | base year for supplemental module data (supplemental module inputs from Soundcast model) |
| scenario_name | '2014' | Scenario name; this is the name of the folder with scenario inputs under the inputs folder |
| daysim_code | [working_folder]/ daysim_2016 |
Path of the folder with DaySim executable |
| main_inputs_folder | [working_folder]/ inputs |
Path of the inputs folder |
| base_inputs | main_inputs_folder + Scenario_name |
Path of the scenario inputs |
| modeller_initial | anything less than 4 characters | initial of the model operator |
| input_ensemble | inputs\\parking_gz.csv |
|
| Subprocess Flags | ||
| run_update_parking | False | Flag to update parking; set to True if new parking data or a future-year scenario |
| run_accessibility_calcs | True | Flag to run accessibility (parcel buffering) calculations; set to True if an updated primary parcel file with new land-use data |
| run_copy_daysim_code | True | Flag to copy DaySim code to the DaySim directory used in the model; set to True if a new setup is created or DaySim code is updated in the DaySim folder |
| run_copy_input_files | True | Flag to copy scenario input files to appropriate locations for the next model run |
| run_setup_emme_project_folders | True | Flag to set up Emme project folders; set to True for a new setup |
| run_setup_emme_bank_folders | True | Flag to set up Emme bank folder; set to True for a new setup |
| run_import_networks | True | Flag to import network; this switch can be set to False while debugging and once a previous run has already imported the network to the Emme database; it is recommended to set this to True while doing a full model run |
| create_no_toll_network | True | Flag to create no toll network |
| run_skims_and_paths_seed_trips | False | Flag to build starting skims (skimming and assignment) using DaySim seed trips; this flag is an alternative to run_copy_seed_skims; if both flags are set to False, make sure that starting skims are available under the inputs/ folder |
| should_build_shadow_price | True | Flag to build shadow prices; if set to True, work and school shadow prices are updated using converged skims after the feedback iterations and then DaySim and assignment are run; set to False if work and school location shadow prices are not a concern |
| run_skims_and_paths | True | Flag to run skimming and assignment; it is recommended to keep this switch as True as it is useful while debugging; setting the switch to False skips it |
| run_truck_model | True | Flag to run truck model; it is recommended to keep this switch as True as it is useful while debugging; setting the switch to False skips it |
| run_supplemental_trips | True | Flag to run supplemental trips model; it is recommended to keep this switch as True as it is useful while debugging; setting the switch to False skips it |
| run_daysim | True | Flag to run DaySim; it is recommended to keep this switch as True so that DaySim is run; setting the switch to False skips running DaySim |
| run_daysim_popsampler | False | Flag to run DaySim population sampler; this feature is no longer used. |
| run_accessibility_summary | True | Flag to run accessibility summaries; set to False if buffered file statistics are not needed |
| run_bkrcast_summary | True | Flag to run BKRCast demand summaries; set to False if demand summaries are not needed |
| run_create_daily_bank | True | Flag to create daily network with volumes; this will add counts and daily assignment summary to a new Emme project. It outputs a csv file daily link-level results; this flag is useful for assignment validation |
| run_truck_summary | True | Flag to run truck summaries after truck model is run; set to False truck summaries are not needed |
| run_daysim_report | True | Flag to run DaySim report; this flag is activated only if run_bkrcast_summary is set to True |
| run_day_pattern_report | True | Flag to run day pattern report; this flag is activated only if run_bkrcast_summary is set to True |
| run_mode_choice_report | True | Flag to run mode choice report; this flag is activated only if run_bkrcast_summary is set to True |
| run_dest_choice_report | True | Flag to run destination choice report; this flag is activated only if run_bkrcast_summary is set to True |
| run_long_term_report | True | Flag to run long-term choice report; this flag is activated only if run_bkrcast_summary is set to True |
| run_time_choice_report | True | Flag to run time of day choice report; this flag is activated only if run_bkrcast_summary is set to True |
| run_district_summary_report | True | Flag to run district summary; this flag is activated only if run_bkrcast_summary is set to True |
| run_landuse_summary | True | Flag to run land-use inputs summary; this summarizes parcels and the synthetic population |
| Feedback Loop Settings | ||
| pop_sample | [1, 1, 1] | DaySim population sampling rate by feedback loop, used in pop sampler. Its length decides how many feedback loops. |
| max_iterations_list | [50, 100, 100] | Assignment iterations by feedback loop |
| min_pop_sample_convergence_test | 10 | |
| shadow_work | [1, 1, 1] | |
| shadow_con | 30 | %RMSE for shadow pricing to consider being averaged |
| Log File Settings | ||
| main_log_file | 'bkrcast_log.txt' | primary log file |
| good_thing | ["cookie", "run", "puppy", "seal sighting", "beer", "snack", "nap", "venti cinnamon dolce latte"] | A list of strings to print out at the end of a successful run |
| transit_modes | {"lbus": "bus", "ebus": "express","fry": "ferry", "crt": "commuter_rail", "lrt": "light_rail"} | Transit modes; would compute nearest distance to these; key=resulting parcel columns (minus suffix), value=names in the add-on dataset |
| Input Checks | ||
| commonly_missing_files | ['buffered_parcels.dat', 'tazdata.in'] | Files that are usually missing |
| DaySim | ||
| households_persons_file | r'inputs\ hh_and_persons.h5' | Synthetic population file |
| sampling_option | 2 | District sampling option to be picked from five options available in pop_sample_district. This setting is used in population sampler to create a synthetic population. This setting is not used by DaySim. DaySIm settings are in the properties file. |
| pop_sample_district | {'BKR':[1,4,2],'Seattle':[1,0.50,0.50],'Rest of King':[1,0.20,0.20],'Pierce':[1,0.10,0.10],'Snohomish':[1,0.10,0.10],'Kitsap':[1,0.10,0.10]} | Three population sampling options; each column represents an option. Sampling plans were derived from Daysim ABM stochastic variation analysis, April 2017. To change sampling rate for a region in a sampling option, change the corresponding column for the region. For example, for outside, change the same column for every region other thn BKR |
| zone_district_file | TAZ_District_CrossWalk.csv | Zone-to-district correspondence |
| taz_sample_rate_file | taz_sample_rate.txt | Intermediate output |
| Bike Model Settings | ||
| bike_assignment_tod | ['6to9'] | Time period for bike assignment |
| aadt_dict | {'volume_wt': {1: 0.368, 2: 1.40, 3: 7.157}} | Distance perception penalties for link AADT; 1 is 10k-20k, 2 is 20k-30k, 3 is 30k+; no penalty applied for AADT < 10k |
| aadt_bins | [0, 10000, 20000, 30000, 9999999] | AADT segmentation breaks to apply volume penalties |
| aadt_labels | [0,1,2,3] | Corresponding "bucket" labels for AADT segmentation for aadt_dict |
| bike_facility_crosswalk | {'@bkfac': { 0:'none', 1:'premium', 2:'standard',3:'standard', 4:'standard'}} | bike facility classification |
| facility_dict | {'facility_wt': {'premium': -0.160,'standard': -0.108, 'none': 0}} | Perception factor values corresponding to these tiers, from Broch et al., 2012 |
| slope_dict | {'slope_wt': {1: .371,2: 1.203,3: 3.239}} | Perception factor values for 3-tiered measure of elevation gain per link gain per link; 1=between 2-4% grade, 2=between 4-6% grade, 3=greater than 6% grade |
| slope_bins | [-1, 0.02, 0.04, 0.06, 1] | Bin definition of total elevation gain (per link) |
| slope_labels | [0,1,2,3] | Labels |
| avg_bike_speed | 10 | Average bike speed in miles per hour |
| bike_link_vol | 'outputs/ bike_volumes.csv' | Output file for bike volumes |
| bike_count_data | 'inputs/bikes/ bike_counts.csv' | Output file for bike counts |
| bike_skim_mult | 100 | Multiplier for storing skim results |
| extra_attributes_dict | {'@tveh' : 'total vehicles', '@mveh' : 'medium trucks', '@hveh' : 'heavy trucks', '@vmt' : 'vmt', '@vht' : 'vht', '@trnv' : 'buses in auto equivalents', '@ovol' : 'observed volume', '@bveh' : 'number of buses'} | |
| transit_extra_attributes_dict | {'@board' : 'total boardings', '@timtr' : 'transit line time'} | |
| Truck Model Settings | ||
| truck_model_project | 'Projects/TruckModel/ TruckModel.emp' | Emme project name for truck model |
| districts_file | 'districts19_ga.ens' | District file name |
| truck_trips_h5_filename | 'inputs/4k/ auto.h5' | Truck trips file name |
| truck_base_net_name | 'am_roadway.in' | Base network file name for truck model |
| truck_generalized_cost_tod | {'5to9':'am', '15to18':'pm'} | Time periods for truck generalized cost |
| LOW_STATION | 1511 | Minimum external zone ID |
| HIGH_STATION | 1528 | Maximum external zone ID |
| EXTERNAL_DISTRICT | 'ga08' | External district label |
| truck_adjustment_factor | {'ltpro':0.544, 'mtpro':0.545, 'htpro':0.530, 'ltatt':0.749, 'mtatt':0.75, 'htatt':1.0} | Truck adjustment factors by truck class (light, medium, and heavy) and production/attraction; key=truck class by prod/att, value = adjustment factor |
| Calibration/Validation Settings | ||
| h5_results_file | 'outputs/ daysim_outputs.h5' | Filename for DaySim output in HDF5 format |
| h5_results_name | 'DaysimOutputs' | Name of the output |
| h5_comparison_file | 'scripts/summarize/ inputs/ calibration/ survey.h5' | Data in HDF5 format to compare with the model |
| h5_comparison_name | 'Survey' | Name of the data |
| guidefile | 'scripts/summarize/ inputs/ calibration/ CatVarDict.xlsx' | |
| districtfile | 'scripts/summarize/ inputs/ calibration/ TAZ_TAD_County.csv' | Zone-to-district and county correspondence |
| FAZ_TAZ | 'scripts/summarize/inputs/ calibration/ FAZ_TAZ.xlsx' | FAZ-to-TAZ correspondence |
| LEHD_work_flows | 'scripts/summarize/ inputs/ calibration/ HFAZ_WFAZ_LEHD2014.xlsx' | Longitudinal Employer Household Dynamics (LEHD) worker flows |
| acs_data | 'scripts/summarize/ inputs/ calibration/ ACS_2014.xlsx' | ACS 2014 data file |
| report_output_location | 'outputs' | Location to output reports |