Project Description

Written Description of The Workflow

What does this workflow/idea accomplish?

• Lack of easy workflow that facilitates utilization of GIS data in architectural design process acts as a barrier between design phases especially in planning and concept design phases. Our team is developing a backbone workflow that brings GIS data to RHINO environment via using existing tool like ARCGIS and developing GH script. Since this process establishes the workflow and script that understand GIS data structure, adding more layers’ information from different set of GIS data will be simple variation of the script we developed. This script / workflow will act as good core script of numerous variants.
• The workflow we created is in two steps. The first step is finding a site with the highest potential for a certain type of development via using GIS data as decision making tool. In this case, we used residential development for example. The second step is importing GIS data (GeoJSON, shapefiles, CSV, etc.) into GH and visualizing the analyze within Rhino environment.

Detailed Methodologies for the Workflow

GIS SITE SELECTION Methodology

Pre-processing

1. Download Data

• Demographic Characteristics. 2014-2018 5-years estimates. [csv] American Community Survey. Census Bureau
  – Total Population
  – Median Household Income
  – Median Age
  – Median Housing Value
  – Median Gross Rent
  – Total Number of Housing Units
  – % of occupied housing units
• Census Tracts Geographies. [shapefile] Tiger Lines
• Tax Lot Data. MapPLUTO 20v4. Shoreline clipped [shapefile] NYC Department of City Planning from Department of Finance’s Digital Tax Map (DTM)
• Subway Entrances. [shapefile] NYC Open Data. Metropolitan Transportation Authority (MTA)
• Open Space (Parks). [shapefile]. NYC Open Data. Department of Information Technology & Telecommunications (DoITT)

2. Clean Data

• Python script that cleans census data – (drops not needed columns, renames fields, calculates percentages… etc)

Multi Criteria Decision Model

1. Project all layers to State Plane Projected Coordinate System for New York ( NAD 1983 StatePlane New York Long Isl FIPS 3104 (US Feet))

2. Using specific criteria to narrow down the number of parcels selected. Objective: Find those lots that would be more attractive for a residential developer

Select only those lots that:

• Are vacant
• Are zoned as multi-family residential
• Have a private owner
• Are not landmarks or in a historic district
• Their lot depth is higher than 74 feet and their frontage is wider than 30 feet
• Residential FAR allowed is greater than 3
• Are within walking distance to a park
• Are within walking distance to a subway entrance

3. Rank those lots from more attractive to less, based on:

• Lot Type
  – Corner: 4
  – Through: 3
  – Inside: 2
  – Unknown: 1
  – Waterfront: – 1
  – Interior Lot: -2
  – Island Lot: – 10
  – Alley Lot: -10
  – Submerged Land Lot -10
• Zoning residential sub-type: 0-30 weighted after for 0-5
  – Zoning that allows for more development is ranked higher
  From 0-30 afterwards re-weighted to from 0-5 to keep the same scale
  Info from: https://www1.nyc.gov/site/planning/zoning/districts-tools/r1.page
• Lot Surface – bigger ranked higher
  – < 3949: 1
  – 3959-6050:2
  – 6050-285727: 4
  – >284727: 5
• Median Housing Value around them (not yet accomplished)
• Median Gross Rent around them (not yet accomplished)

4. Map Algebra. Set up a formula that calculates the final ranking by prioritizing some of the criteria higher than others

Calculated Final Score: 1 * (Lot Type) + 2 * (Zoning residential sub type) + 3 * (Lot surface) + 4 *(MHV) + 4*(MGR)

5. Export the highest ranked lot and area around

• Make a buffer of half a mile around it
• Select the census tracts that intersect with that buffer – export to shapefile
• Select the parcels that intersect with that buffer – export to shapefile
• Export parcel to shapefile

IMPORTING GIS DATA TO ARCHITECTURE FRIENDLY PLATFORM (RHINO)

Import GIS data (GeoJSON, shapefiles, CSV, etc.) into GH and visualize the analyze within Rhino environment. Later can either export as diagrams or help designers to make plausible decision

Pre-processing

Export Data type selection (GeoJSON, shapefiles, CSV, OSM) shapefiles can set CRS

1. Import tool selection:

• Heron (can only pure ESRI data through internet, has bugs, cannot import shapefiles)
• ELK (can only accept OSM file)
• SHRIMPGIS (can handle limited amount of data, bugs, one reference system)
• GHOPPERGIS (excessive module, not condense, bugs)
• MEERKAT GIS (need to bill google cloud for API Key but cannot download .shp info)
• Humpback (GeoJSON limit, small amount of data handle)
• DeCodingSpaces (GeoJSON file limit)
• “Finches.py” (Pure, clean, no extra plugin needed, least bugs)

2. Data and Plugin Preparation

• Get 3d building .shp file through QGIS using QuickOSM to download from OSM and set CRS
• (NAD 1983 StatePlane New York Long Isl FIPS 3104 (US Feet))
• Prepare reference .xlsx or .csv file, e.p. LandUse Color Index, Residential Zone Rank
• Using LunchBox, Human, TT Toolbox
• Get purged .shp date from Urban Design Colleagues

3. Data processing

• Import .shp, .csv, .xlsx, file into grasshopper
• Select the Specific data type need to show
• Coorelating index data with attributions need to visualize
• Set gradients or color template
• Create a dashboard to show legend
• Organize the input and output

4. Next step (in time sequence)

• Expend attribution library and gradually set up a database
• Setup a comprehensive color legends dashboard template and reference data
• Design a more user-friendly UI, to make the operating process easier
• Export to Mapbox to render interactive animation (client oriented, easier to operate UI)

Pre-Existing Content & 3rd Party Resources

• ArcGIS Pro
• Jupyter notebooks
• Anaconda
• Pinches.py (to import shp files to Grasshopper environment)
• QGIS
• QuickOSM
• LunchBOX, Human, and TT toolbox