RAPID ANALYTICS INTERACTIVE SCENARIO EXPLORER (RAISE) A tool for analysing and visualising land valuation in different development scenarios

Transcription

1 RAPID ANALYTICS INTERACTIVE SCENARIO EXPLORER (RAISE) A tool for analysing and visualising land valuation in different development scenarios

2 RAISE PROJECT COLLABORATION

3 RAISE DATA DRIVEN APPROACH Rapid Analytics Interactive Scenario Explorer Analysis Collaboration Location Intelligence

4 RAISE Objectives: Develop open, cloud-based architecture to combine data, models, and visualisation. Develop an interactive scenario explorer toolkit. Explore collaborative visualisation methods. Apply toolkit to automated valuation modelling. Apply toolkit to land value uplift modelling.

5 RAISE1 to RAISE2 RAISE1 Scenario exploration: Value uplift from transportation infrastructure deploy and evaluate RAISE to assist Council planners to rapidly quantify the nominal value uplift of transportation infrastructure proposal scenarios Study area: Western Sydney

6 RAISE1 to RAISE2 RAISE2 Scenario exploration: Value uplift from transportation infrastructure redeploy RAISE for Greater Sydney area and Brisbane City Council Scenario exploration: Value uplift from changing planning controls RAISE Community Version

7 DATA IN RAISE Data Source APM AVM ABS Census NSW Department of Planning & Environment (DPE) NSW Land & Property Information (LPI) NSW Department of Education (DET) Data Derived Property transaction records (transaction time, sold price, land area size, property type, property structural attributes); APM AVM value Digital boundaries and demographics (Employment rates, Aged population, Migration) Land zoning, FSR, Maximum building heights Cadastral and digital topographic database; LPI valuation components; Property characteristics, Points of Interest, Roads, Electric Transmission Lines. School locations and catchments NSW Bureau of Crime Statistics and Research (BOCSAR) NAPLAN PSMA Australia Crime data (breaking and entering dwelling) School performance compared to national average Geoscape (building footprints, building heights, materials)

8 DATA IN RAISE Data Source QLD Open Data Brisbane City Council Data Derived Cadastral and digital topographic database, State school locations and catchments, planning datasets Current land use, Building footprints

9 PROPERTY VALUATION MODEL 1. Data-driven knowledge of property values Using hedonic price models When we know a property s value (V), we assume it is a function of some set of known attributes of the property, neighbourhood and location (A1, A2, A3...). The modelling finds the relative weighting of these attributes (W1, W2, W3...): V = (A 1 W 1 ) + (A 2 W 2 ) + (A 3 W 3 ) C

10 PROPERTY VALUATION MODEL 1. Data-driven knowledge of property values Using hedonic price models When we don t know a property s value (V), we use the same known attributes and (now known) weightings to estimate the value: V = (A 1 W 1 ) + (A 2 W 2 ) + (A 3 W 3 ) C

11 RAISE HEDONIC MODEL ATTRIBUTES The RAISE 2 cross-sectional model was developed using a total of residential property sales records in the Greater Sydney City area in More than 60 hedonic attributes were tested and fitted into the models, finally 24 significant variables were selected and included in the RAISE2 hedonic model) Property/Land Attributes Property Type- House Property Type-Semi Property Type- Unit Number of Bedrooms Number of Bathrooms Number of Parking Swimming Pool Maximum Building Height Locational/Accessibility Attributes Distance to City Centres (m) Distance to Beaches (m) Distance to Primary Schools (m) Distance to Railway Stations Distance to Wharfs Distance to Universities Within 50m distance from a High Voltage Electricity Transmission Line Within 100m distance from a Main Road Within100m distance from a Railway Line (on ground) Within 20m distance from an Primary Easement Neighbourhood Attributes Family Income Crime Rates Percentage of Senior Aged People Percentage of People Born Overseas High School NAPLAN Indicator Primary School NAPLAN Indicator

12 RAISE RESIDUAL LAND VALUE MODEL 1. Objectives of the Residual Land Value Tool Predict the effects of planning controls on land value Predict the effects of planning controls on development feasibility Predict the effects of infrastructure projects on land value Predict the effects of infrastructure projects on development feasibility Identify the minimum planning controls if development is feasible

13 RAISE RESIDUAL LAND VALUE MODEL 2. Data-driven knowledge of the Residual Land Value Tool Using Residual Land Value Approach This approach more directly estimates the price of land that is set by the development potential of a site. Change planning controls Development yields Market value and development costs Input new market value (HPM) Force feasibility Residual land value Estimate change in land value Assess feasibility Estimate change in feasibility

14

15 RAISE FUNCTIONALITY

16 RAISE FUNCTIONALITY BREAKDOWN OF AUTOMATED LAND VALUE WITH CERTAIN VARIABLES DISTANCE TO SELECTED POINT OF INTEREST

17 VALUE UPLIFT FUNCTIONALITY CREATE NEW TRAIN STATIONS ON MAPS SELECT A PROPERTY VALUATION MODEL TO RUN THE VALUE UPLIFT SCENARIO

18 VALUE UPLIFT FUNCTIONALITY BREAKDOWN OF AUTOMATED LAND VALUE WITH CERTAIN VARIABLES DISTANCE TO SELECTED POINT OF INTEREST

19 VALUE UPLIFT FUNCTIONALITY

20 PARRAMATTA LIGHT RAIL OPTIONS

21 SYDNEY METRO NORTH WEST

22 LAND PLANNING ZONES

23 FLOOR SPACE RATIO (FSR)

24 MAXIMUM BUILDING HEIGHT

25 Digital planning tools can support the Digital Twin agenda Mobility data can be used to understand the people dimension to the Digital Twin City Analytics for near real-time decision support is crucial in the Digital Twin. The Digital Twin should support future city planning scenario exploration, through modelling, simulation, artificial intelligence and the involvement of experts and the community. Training and Education there is a need train the next generation of city planners to understand and work with the Digital Twin

26 Thank-you! Prof. Chris Pettit