Data Centre Compass

Team Name

Dual Core

Team Members

Project Description

The goal is to analyze infrastructure, energy, and geographic data to find the best locations for data centres, helping position Australia as a leading AI hub.

Data Story

Methodology for Data Centre Site Selection

To identify optimal locations for data centres in Queensland, a multi-criteria evaluation (MCE) model was developed by analysing open government spatial data.

1. Data Preparation and Analysis
   The initial datasets were aligned to a common cartographic projection to ensure spatial accuracy. For the Minimum Viable Product (MVP), the analysis was focused on the South East Queensland (SEQ) region, covering 23 Local Government Areas (LGAs).

2. Scoring of Suitability Factors
   Each factor was assigned a score on a scale of 1 (least suitable) to 10 (most suitable).

Energy Supply (Substations): Proximity to substations is a key factor. Buffer zones were created to define both an "ideal" connection zone (100-1,000 m, 10 points) and a "setback" zone (0-100 m, 1 point) to ensure safety and security.

Energy Supply (Transmission Lines): Proximity to high-voltage transmission lines is also important. Zones within 1 km were awarded 10 points, with the score decreasing to 2 points at a distance of 40 km.

Transport Access: To assess logistics and personnel access, proximity to major roads was analysed. Zones within 500 m received 10 points.

Operational Expenditure (Temperature): To minimise cooling costs, mean annual temperature data was used. Through raster reclassification, areas with the lowest temperatures were assigned the maximum score (10 points).

1. Risk Assessment and Constraints To exclude unsuitable areas, mask layers were created from official data sources. Locations falling within these hazard zones received a coefficient of 0 in the final formula.

Flood Risk: The Queensland Floodplain Assessment Overlay was used.

Bushfire Risk: Data from Historical Bushfire Boundaries was used.

1. Final Evaluation Formula The final score for each point on the map was calculated using a weighted sum. The weights (coefficients) were assigned based on the importance of each factor for a data centre:

Proximity to Substations: 35%

Proximity to Transmission Lines: 15%

Transport Access: 25%

Mean Annual Temperature: 25%

Final Formula:
Final Score = (FloodFilter * BushfireFilter) * ((SubstationScore * 0.35) + (TransmissionLineScore * 0.15) + (MajorRoadScore * 0.25) + (TemperatureScore * 0.25))

1. Final Output
   The result of the analysis is presented as a uniform point grid layer. Each point in the attribute table contains the final suitability score, allowing the results to be easily visualised as a heatmap for an intuitive identification of the most promising zones.

2. Future Development
   The model is flexible and scalable. For future development, we recommend:

Integrating renewable energy sources (solar, wind farms) as a suitability factor.

Assessing resource availability for cooling systems (e.g., proximity to water sources).

Expanding the risk analysis to include seismic hazard data and land use restriction zones (e.g., aerodromes, protected areas).

Conducting detailed site-level analysis, incorporating cadastral data (DCDB), land pricing, land ownership, and approval procedures.

Tools: All spatial analysis, from data processing to the final calculation, was performed using the free and open-source software QGIS.