An Overview of Urban Sim Center for Urban

An Overview of Urban. Sim Center for Urban Simulation and Policy Analysis University of Washington www. urbansim. org

Results of Needs Assessment (in order) 1. Analyze Effects of Land Use on Transportation 2. Analyze Multimodal Assignments 3. Promote Common Use of Data 4. Manage Data Needs 5. Analyze All Modes of Travel 6. Analyze Effects of Land Use Policies 7. Support Visualization Techniques 8. Analyze Effects of Transportation Pricing Policies 9. Analyze Effects of Growth Management Policies 10. Analyze Effects of Transportation on Land Use

Recommendations for New Model Design (made in 2001) • Model real estate, labor and transportation demand supply market interactions in an integrated microsimulation framework. § Eliminate artificial separation of household choices • Represent processes at appropriate temporal and spatial detail § Real estate and labor markets: annual, parcel § Activity and travel: mid-term model effort: 5 daily periods, parcel / term: continuous time long- • Integrate and extend recent advances in operational models § Disaggregate land use models (Eugene, Honolulu, Salt Lake City) § Activity based travel models (San Francisco, Portland) • Implement as a distributed model system § for use by PSRC, cities and counties, state agencies, public § with a web-based user-interface • Implementation plan balancing needs, costs, schedule, and risks

Linked Urban Markets Services Governments Infrastructure Land Housing Developers Households Labor Floorspace Businesses Flow of consumption from supplier to consumer Regulation or Pricing

Long-term Modeling Strategy: Integrated Model of Markets for Real Estate, Labor and Transportation Households Demographic Processes • Ageing • Household structure • Migration Long-term Choices • Residential Mobility • Housing Choice • Labor Supply • Workplace Choice • Vehicle Ownership Short-term Choices • Activity Generation • Activity Scheduling • Allocation of Vehicles • Activity Location Governments Land Use Regulation • Land use plans • Growth management Transportation • Infrastructure • Pricing Developers Real Estate Processes • Land development • Housing development • Non-res development • Redevelopment Businesses Economic Processes • Economic structure • Output goods/services • Inter-regional trade Long-term Choices • Mobility • Location Choice • Labor Demand Short-term Choices • Goods movement

An Illustration of the Approach • Consider a major decrease in transport costs • In reality, a household could substitute among any combination of interdependent changes: § § § Change travel routes, modes, times to take advantage Make more trips Make longer trips to other destinations Move residence to buy more space or amenities farther out Change jobs or enter job market Add a household vehicle • The proposed model would recognize these as interdependent choices, whereas current models would not

Phased Model Improvements: 2002 -5 • Implement Land Use Model based on Urban. Sim (current specification) § Phase I: Develop database § Phase II: Estimate model paramaters § Phase III: Test and validate model system • Integrate with current PSRC models: § STEP macroeconomic model § Travel model system

Urban. Sim Design Objectives • Support coordinated land use, transportation, and environmental planning and modeling • Use a transparent behavioral framework • Represent interactions of markets and policies • Represent sufficient detail to support: § Municipal applications; corridor studies § Non-motorized transport policies and TOD § Environmental and land use policies • Support trade-off analysis among objectives § Efficiency § Equity § Environmental Impact • Develop an Open Source modeling platform

Urban. Sim Model Design • Discrete Choice Framework § § Household Relocation and Location Choice Business Relocation and Location Choice Real Estate Development Explicit Markets for Real Estate • Governmental policies exogenous: scenarios • Dynamic § § Path-dependent (history matters) Supply fixed in short run (one year) Adjustment toward equilibrium in long-run Annual time steps

How is Urban. Sim Different? • Dynamic, non-equilibrium approach • Spatial detail very high • Behavioral approach empirically measures responses to policies • Focuses visioning on goals and objectives • Links visioning to planning and evaluation

Household Grid Cell HOUSEHOLD_ID GRID_ID PLAN_TYPE_ID GRID_ID COMMERCIAL_SQFT PERCENT_WATER PERSONS GOVERNMENTAL_SQFT PERCENT_WETLAND WORKERS INDUSTRIAL_SQFT PERCENT_STREAM_BUFFER AGE_OF_HEAD COMMERCIAL_IMPROVEMENT_VALUE PERCENT_FLOODPLAIN INCOME INDUSTRIAL_IMPROVEMENT_VALUE PERCENT_SLOPE CHILDREN GOVERNMENTAL_IMPROVEMENT_VALUE PERCENT_OPEN_SPACE RACE_ID NONRESIDENTIAL_LAND_VALUE PERCENT_PUBLIC_SPACE CARS RESIDENTIAL_LAND_VALUE PERCENT_ROADS RESIDENTIAL_IMPROVEMENT_VALUE IS_OUTSIDE_URBAN_ RESIDENTIAL_UNITS GROWTH_BOUNDARY YEAR_BUILT IS_INSIDE_NATIONAL_FOREST FRACTION_RESIDENTIAL_LAND IS_INSIDE_TRIBAL_LAND JOB_ID PERCENT_UNDEVELOPABLE IS_INSIDE_MILITARY_BASE GRID_ID TOTAL_NONRES_SQFT ZONE_ID SECTOR_ID DEVELOPMENT_TYPE_ID CITY_ID HOME_BASED DISTANCE_TO_ARTERIAL COUNTY_ID SIC DISTANCE_TO_HIGHWAY PERCENT_AGRICULTURAL_ Job RELATIVE_X RELATIVE_Y PROTECTED_LANDS ACRES

Key Model Components • Household and Employment Location § Standard multinomial logit § Grid cell is unit of choice • Real Estate Development § Multinomial logit § 24 development type outcomes • Real Estate Price Estimation § Hedonic regression

Data Inputs • • Regional Control Totals Parcel Data Business Establishments Household Data (Census, Travel Survey) Land Use Plan/Zoning Environmental Features Public Land ownership

Residential Location Variables • Housing Characteristics § Prices (interacted with income) § Development types (density, land use mix) § Housing age • Regional accessibility § Job accessibility by auto-ownership group § Travel time to CBD and airport • Urban design-scale (local accessibility) § Neighborhood land use mix and density § Neighborhood employment

Employment Location Variables • Real Estate Characteristics § Prices § Development type (land use mix, density) • Regional accessibility § Access to population § Travel time to CBD, airport • Urban design-scale § Proximity to highway, arterials § Local agglomeration economies within & between sectors: center formation

Development Variables • Site characteristics § Existing development characteristics § Land use plan § Environmental constraints • Urban design-scale § § Proximity to highway and arterials Proximity to existing development Neighborhood land use mix and property values Recent development in neighborhood • Regional § Access to population and employment § Travel time to CBD, airport

Land Price Variables • Site characteristics § Development type § Land use plan § Environmental constraints • Regional accessibility § Access to population and employment • Urban design-scale § Land use mix and density § Proximity to highways and arterials

Urban. Sim – Travel Model Interactions Households by Income Age of head Size Workers Children Employment by sector Travel Models Urban. Sim Composite Utility by Auto Ownership Highway Travel Times Vehicle Ownership Probabilities

Creating Policy Scenarios • Macroeconomic Assumptions § Household and employment control totals • Development constraints § Can select any combination of • Political and planning overlays • Environmental overlays • Land use plan designation § Constraints determine which development types cannot be built • Transportation infrastructure • User-specified events

Puget Sound Regional Council 4 Counties 72 Cities 3 Ports WS Dept of Transp WS Transp Comm also: 6 transit agencies 6 Associate Members 2 adjacent counties 2 tribes 1 port Evans School of Public Affairs, UW SNOHOMISH Everett Bremerton Bellevue Seattle KITSAP Tacoma PIERCE KING

Puget Sound Regional Council Population: 3, 275, 847 Area: 6, 287. 8 sq mi SNOHOMISH Everett Bremerton Bellevue Seattle KITSAP Tacoma PIERCE KING

Puget Sound Regional Council Population: 3, 275, 847 Area: 6, 287. 8 sq mi Urban Growth Area Population: 2, 804, 125 Area: 980. 1 sq mi SNOHOMISH Everett Bremerton Bellevue Seattle KITSAP Tacoma PIERCE KING

Major Steps in Data Preparation 1. Determine study area boundary 2. Generate grid over study area 3. Assemble and standardize parcel data 4. Impute missing data on parcels 5. Assemble employment data 6. Assign employment to parcels 7. Convert Parcel data to grid 8. Convert other GIS layers to grid 9. Assign Development Types 10. Synthesize household database 11. Diagnose data quality and make refinements