Buildings.sg Documentation
About the Project Project Structure Main Features Version Information Data Data Resource Mapping Data Dictionary Data Download Energy Simulation Templates Map GeoJSON Energy Simulation Before Start Quick Start Step 1 Step 2 Step 3 Step 4 Step 5 Embodied Carbon Simulation Before Start Quick Start Step 1 Step 2 Step 3 Step 4 Publication Team Disclaimer

Buildings.sg Documentation

About the Project

Buildings.sg is an open, interactive platform developed by the City Syntax Lab at the National University of Singapore (NUS) for urban building energy modeling (UBEM) and carbon emissions analysis in Singapore. The platform integrates spatial data with operational and embodied carbon simulations, supported by a comprehensive set of open-source EnergyPlus templates for typical building archetypes in Singapore.

Users can download and customize these EnergyPlus templates for detailed simulations. Buildings.sg enables visualization and analysis of building-level carbon footprints, facilitating identification of energy efficiency improvements and carbon reduction opportunities. The platform supports Singapore’s Green Plan 2030 by promoting energy-efficient designs, sustainable materials, and low-carbon construction practices. Designed to be scalable and adaptable, Buildings.sg can be deployed in any city or municipality with available building and climate data.


Building Archetypes

Singapore's buildings are classified into residential and non-residential categories, further divided into seven Building Subtypes and 23 Building Archetypes, forming a comprehensive three-tier system. This framework is specifically optimized to support accurate simulation of building carbon emissions in Singapore. For detailed information about each subtype and archetype, please refer to the Archetypes section of the map.

Operational Carbon Simulation

Operational carbon represents emissions arising from a building's daily energy consumption, including cooling, lighting, equipment, and heating systems. We adopt an Urban Building Energy Modeling (UBEM) approach, a city-scale shoebox-based methodology that simulates energy performance across the built environment. This analytical framework allows evaluation of multiple scenarios to identify effective carbon reduction strategies and improve energy efficiency.

Embodied Carbon Simulation

Embodied carbon accounts for emissions associated with building materials and construction throughout the building's lifecycle. While full life cycle assessments (LCA) consider multiple stages, this project focuses on emissions from raw material extraction, transportation, manufacturing, and construction (stages A1–A5, known as "cradle to practical completion"). Due to limited data and inherent uncertainties, a probabilistic calculation method is used. Results are presented as probability distributions rather than single values, reflecting uncertainty and enabling effective risk assessment.

Project Structure


    index.html          – Main entry point that loads and initializes the web application.
    documentation.html  – Documentationation for project pipelines and user guidance.
    data/               – GeoJSON, and spreadsheets supporting charts and analysis.
    download/           – EnergyPlus IDF templates, DesignBuilder files, and map data.
    image/              – Archetype illustration images for interface visualization.
    logo/               – Icons and branding graphics displayed across the site.
    mapbox.js           – Initializes and styles the interactive map using Mapbox GL JS.
    panel.js            – Manages UI panels, controls, and filtering logic.
    popup.js            – Generates informational popups for Data and About sections.

Main Features

The platform offers a range of interactive tools designed to support urban-scale building performance analysis and decision-making:

Version Information

Platform Version: 1.0          Status:Active
Building Dataset Version: 5.0          Last Updated: 2025-11-15


Energy Simulation

Version: 5.0           Last Updated: 2025-6-18


Embodied Carbon Simulation

Version: 3.0           Last Updated: 2025-6-11

Data

Data Resource

Buildings.sg Dataset

Buildings.sg provides an integrated urban-to-building-level dataset covering Singapore. Sources include:

The data is provided in unified GeoJSON format and visualized using Mapbox GL JS. Map layers include building archetypes, carbon metrics, EUI, and Green Mark ratings. Each feature represents a single building with attributes such as building levels, height, footprint, gross floor area, archetype, energy use, and carbon metrics.


Archetypes


Energy Simulation


Embodied Carbon Simulation

Mapping Data Dictionary

Features Name Dataset Field Name Description Data Type Data Source
Building ID id Unique identifier for each building, formatted as 'relation/' or 'way/' followed by a 7-digit number. Text OpenStreetMap
Building Name addr_housename Official name of the building, which may sometimes match its address. Text OpenStreetMap/BCA
Building Address (Housenumber) addr_housenumber Building's street or unit number, usually appearing before the street name in the full address. Text OpenStreetMap/BCA
Building Address (Street) addr_street Official street or road name where the building is situated. Text OpenStreetMap/BCA
Postal Code addr_postcode Singapore postal code assigned to the building location. Text OpenStreetMap/BCA
Building Levels building_levels Total number of above-ground floors in the building, excluding basement levels. Text OpenStreetMap/HDB*/Google Map
Building Archetype building_archetype Classification category from the 23 defined building archetypes, spanning both residential and non-residential uses. Text OpenStreetMap/HDB*/BCA/Machine Learning
Machine Learning Predicted Archetype Probability ml_probability The probability that a building belongs to a specific archetype, as predicted by a random forest classification model, based on input data from buildings in Singapore with known archetypes. Number Machine Learning
Height height Total vertical height of the building in meters, measured from ground level to the highest point. Number OpenStreetMap/HDB*
Building Footprint building_footprint Ground-level area covered by the building, calculated from its geometric polygon outline (m²). Number OpenStreetMap
Gross Floor Area gross_floor_area Total usable floor space in the building (m²), calculated or estimated based on footprint and number of floors. Number BCA/OpenStreetMap
Built Year built_year Year of the building's original construction completion. Text HDB*/OpenStreetMap/BCA
Data Source data_source Origin of the building data, with HDB and BCA sources specially marked for Mapbox layer filtering. Default: OpenStreetMap. Text HDB*/BCA
Green Mark Rating greenmark_rating Building's sustainability performance rating under Singapore's Green Mark certification system. Text BCA
Green Mark Year of Award greenmark_year Year when the Green Mark certification was granted to the building. Text BCA
Green Mark Version greenmark_version Specific edition of the Green Mark assessment criteria used for the building's certification. Text BCA
AC Floor Area Percentage aircon_area Percentage of total floor area with air-conditioning, primarily relevant for non-residential buildings. Number BCA
Type of AC System aircon_type Specific air-conditioning technology deployed in the building (e.g., central, split, VRF). Text BCA
Average Monthly Building Occupancy Rate occupancy Average monthly occupant density measured in persons per square meter (P/sqm). Number BCA
2021 EUI eui2021 Measured Energy Use Intensity for 2021, representing annual energy consumption per unit floor area (kWh/m²/year). Number BCA
2022 EUI eui2022 Measured Energy Use Intensity (EUI) for 2022, representing annual energy consumption per unit floor area (kWh/m²/year). Number BCA
2023 EUI eui2023 Measured Energy Use Intensity for 2023, representing annual energy consumption per unit floor area (kWh/m²/year). Number BCA
Total Dwelling Units total_dwelling_units Complete count of individual residential units within the building. Number HDB*
Total Embodied Carbon embodied_carbon Total carbon emissions associated with building materials and construction processes (kgCO₂). Number Simulation
Cooling Carbon energy_cooling Carbon emissions resulting from energy used for space cooling and air conditioning (kgCO₂). Number Simulation
Lighting Carbon energy_lighting Carbon emissions resulting from energy consumed for interior and exterior lighting (kgCO₂). Number Simulation
Equipment Carbon energy_equipment Carbon emissions from energy used by appliances, office equipment, and other electrical devices (kgCO₂). Number Simulation
Water Carbon energy_water Carbon emissions associated with energy used for water heating, pumping, and treatment systems (kgCO₂). Number Simulation
Total End Use energy_total Aggregate carbon emissions from all operational energy consumption categories (kgCO₂). Number Simulation
Geometry Type type GeoJSON geometry classification, typically "Polygon" for simple buildings or "MultiPolygon" for more complex structures. Text OpenStreetMap
Geometry Coordinates coordinates Ordered array of geographic coordinates defining the building outline, with each point represented as [longitude, latitude]. Array OpenStreetMap
Land Use ID Name Unique identifier assigned to each land parcel in the Master Plan Text URA
Land Use GPR GPR Maximum permissible gross plot ratio specified for the parcel Number URA
Land Use Description LU_DESC Official land use designation under the URA Master Plan 2019 (e.g., Residential, Commercial, Park) Text URA

*Data for HDB buildings is derived from various data sources including HDB, OpenStreetMap, and the HDB dataset from the NUS Urban Analytics Lab.

Data Download

Energy Simulation Templates

Archetype Name Description Download IDF Download DesignBuilder
HDB PPVC Prefabricated Pre-finished Volumetric Construction (PPVC) is a method of building that uses modular units, pre-built offsite and assembled onsite. This approach helps reduce construction time and enhances quality control. IDF DesignBuilder
HDB non-PPVC Traditional method of construction used by the Housing and Development Board (HDB) without prefabricated units. It involves the use of concrete and steel to construct public housing units. IDF DesignBuilder
Landed property Landed properties are individual houses that have their own plot of land. They include detached houses, semi-detached houses, and terraced houses, offering more space and privacy compared to high-rise flats. IDF DesignBuilder
Private apartment & condo Private apartments and condominiums are residential buildings that are owned privately. They usually offer amenities such as swimming pools, gyms, and security, and can be found in both city-centre and suburban areas. IDF DesignBuilder
Shophouse Shophouses are traditional buildings that combine both commercial and residential uses. They typically consist of a ground-floor retail space with living quarters above. IDF DesignBuilder
Hotel Hotels are commercial establishments offering accommodations to travelers, with amenities such as room service, concierge, and recreational facilities. IDF DesignBuilder
Retail Retail properties include spaces for selling goods directly to consumers, such as shopping malls, stores, and boutiques. IDF DesignBuilder
Mixed development Mixed development properties are those that combine different types of uses, typically commercial, retail, and residential, within a single complex. IDF DesignBuilder
Office Office buildings are spaces designed for business and administrative purposes. They can range from small office buildings to large office towers. IDF DesignBuilder
Business park Business parks are specialized areas designed to accommodate various types of businesses, particularly in research and development, information technology, and industrial services. IDF DesignBuilder
Hospital Hospitals are large healthcare facilities that provide medical treatment and emergency care, with inpatient and outpatient services. IDF DesignBuilder
Clinic Clinics are smaller healthcare facilities that offer outpatient medical services, typically for general practitioners or specialized services. IDF DesignBuilder
Nursing home Nursing homes are facilities that provide residential care and support for elderly individuals, including medical assistance and daily living activities. IDF DesignBuilder
Inst. of higher learning (IHL) Institutions of higher learning (IHL) include universities, polytechnics, and other higher education institutions offering advanced degrees and diplomas. IDF DesignBuilder
Non-IHL Non-Institution of Higher Learning refers to other types of educational institutions, including schools and training centres that provide education at a pre-university or vocational level. IDF DesignBuilder
B1 B1 industrial properties are those zoned for light industry or research and development activities. They often accommodate businesses that do not involve heavy manufacturing processes. IDF DesignBuilder
B2 B2 industrial properties are those zoned for general industrial uses, including manufacturing and processing activities that may produce noise, pollution, or waste. IDF DesignBuilder
Data centre Data centres are facilities used to house computer systems and associated components, such as telecommunications and storage systems, often with high levels of security and redundancy. IDF DesignBuilder
Civic, community & cultural inst. Civic, community, and cultural institutions include public buildings that serve community and cultural purposes, such as libraries, museums, and community centres. IDF DesignBuilder
Sports & recreation Sports and recreation properties include facilities such as sports complexes, stadiums, gyms, swimming pools, and recreational parks. IDF DesignBuilder
Restaurant Restaurants are establishments that provide food and beverages to customers in exchange for money, typically served at tables in a dining setting. IDF DesignBuilder
Hawker centre Hawker centres are open-air food courts where various food vendors sell affordable local dishes, providing an essential part of Singapore's food culture. IDF DesignBuilder
Supermarket Supermarkets are large retail spaces that sell a wide variety of food, beverages, and household products, typically arranged in aisles. IDF DesignBuilder

Download All Archetypes

Map GeoJSON

Download Buildings.sg Mapping Dataset

Energy Simulation

Before Start

Before beginning your urban- to district-scale energy simulation, make sure you have all the necessary software, tools, and files ready. This preparation will help you efficiently set up your model, run simulations, and analyze results without interruptions.

Required Software

Required Files

Optional Online Tools

Quick Start

step 4.2-3

Step 1: Study Area Selection

Start by defining the area you want to study. This can be a single building, a block, a neighborhood, or a larger district. Identify the following information for all buildings in your study area:

You can obtain this information from Buildings.sg (for Singapore) or other public sources such as Google Maps or OpenStreetMap.

Step 2: Geometry Export & Cleaning

If you already have a complete 3D model in Rhino3D, you can skip this step. Otherwise, follow these steps to obtain geometry data:

Example using Cadmapper:

  1. Enable Include 3D Buildings (if available).
  2. Draw a rectangle that fully encloses your study area and click CREATE FILE.
  3. Review the 3D Axonometric View on the right to ensure it is correct.
  4. Click DOWNLOAD to obtain your .3dm model.
Cadmapper Screenshot

Step 3: Attribute Completion

Open your 3D model in Rhino and perform the following:

  1. Check all buildings for completeness and fix any duplicates or anomalies.
  2. Remove non-building objects (roads, green spaces, etc.) that are irrelevant to energy simulation.
  3. Update building geometry with the correct height and function/type based on your research from Step 1. (You may need to manually adjust heights for individual buildings.)

Before and after this step:

Rhino3D before step 3 Rhino3D after step 3

Step 4: Simulation Template Setup

4.1 Install Plugins

Ensure Ladybug Tools / OpenStudio plugins are installed. Open Grasshopper in Rhino to confirm successful installation.

step 4.1

4.2 Import Archetype Template

Import the provided EnergySim_regional.gh file (containing all building archetypes) into Grasshopper. You can find the file at our GitHub repository

step 4.2-1

Each archetype as shown below.

step 4.2-2

The internal structure of the EnergySimCluster is shown below.

step 4.2-3

Troubleshooting: If any components are red or display errors:

  1. Confirm that plugins from Step 4.1 are installed correctly.
  2. Check that OpenStudio is added to your system PATH.
  3. Ensure the template version is compatible; incompatible components can be replaced with the new ones.

4.3 Configure Weather & Geometry

  1. Replace the Weather File module with your own .epw file.
  2. Right-click the Geometry module, select buildings corresponding to your study area by name, and press Enter. Selected buildings will appear red in Rhino.
step 4.3-1
step 4.3-2

4.4 Customize Input Parameters

Double-click EnergySim Cluster to access internal input/output parameters. Modify inputs based on your data to generate a benchmark simulation.

4.5 Run Archetype Simulations

Double-click FalseStart under each archetype to start the simulation. Results for that archetype will appear in the panel to the right. Repeat for all archetypes. Calculate the overall district EUI by weighting individual archetype results.

step 4.5

Step 5: Scenario Setup & Batch Simulation

With the benchmark results obtained, you can perform scenario analysis:

Notes:

Embodied Carbon Simulation

Before Start

Before running the probabilistic embodied carbon estimation for UBEM, make sure you have all the necessary software and tools ready. This preparation ensures smooth execution of the scripts and accurate generation of results.

Required Software for Code Execution

Software for District Modeling (Optional)

Quick Start

Step 1: Prepare Data

Edit the Excel and CSV files in the /data subfolder as needed before running any scripts. This ensures that your embodied carbon calculation will use the most accurate and relevant inputs for your study.

Detailed Tutorial for Building Embodied Carbon Simulation

Step 2: Generate District-Independent Results

Run the following scripts in order to produce general EC results not tied to a specific district:

  1. EC_Calibration_A1A3ECF_Weights.py calibrates hyperparameters to meet the error threshold.
  2. EC_execution.py generates archetype_container objects containing generalized EC results for each archetype; saves them as .pkl files.
  3. EC_results.py produces reports and diagrams in .pkl, .xlsx, and .csv formats.

Step 3: Provide Study District Details

To generate results for a specific district, you need to provide:

  1. GFA of each building Use the Grasshopper script district_modelling/Building GFA Calculator.gh or manually input data. Refer to the tutorial district_modelling/Rhino & Grasshopper Tutorial.pptx for guidance.
  2. Archetype classification Assign each building to an archetype listed in data/Available Archetypes.txt.
  3. Input data Enter building GFAs and archetypes into data/District Individual Building GFAs.xlsx.

Step 4: Generate District-Specific Results

After providing district details, run the following scripts:

  1. EC_district_execution.py calculates embodied carbon based on the district data and saves results as .pkl files.
  2. EC_district_result.py generates district-level reports and diagrams for analysis.

Publication

Team

City Syntax Lab & NUS Logo

Yu Qian Ang: Principal Investigator, project coordination, and research guidance
Hongyao Wong: Data analysis and Energy Template modeling
Royden Soh Wei Jun: Data analysis and Embodied Carbon Template modeling
Pengdi LYu: Platform development, data analysis, and Energy Template modeling

Disclaimer

Data and results presented in Buildings.sg are for informational purposes only, and should not be used as the basis for legal, financial, or regulatory decisions. Despite our commitment to accuracy, we make no warranties regarding the precision, completeness, or reliability of this data. Users should independently verify all information through official sources before taking any action based on this platform. The developers, contributors, and affiliated organizations expressly disclaim all liability for any consequences arising from the use of this tool.