The global focus on sustainable development has intensified in recent years, with the United Nations (UN) establishing the 2030 Agenda for Sustainable Development (SDGs). Traditional methods of assessing the SDGs, based on administrative units such as cities or counties, however, often lack the necessary granularity for effective urban interventions. This approach can overlook the nuances of urban dynamics, including environmental degradation or widening income inequalities, and does not always offer clear, practical recommendations tailored to the unique challenges of individual cities.

Concepts such as the “15-minute city”, which aim to provide access to essential services and amenities within a short walking or cycling distance, are conceptually strong, but often they overlook accessibility barriers for all residentsFor example, while cities such as Copenhagen, Melbourne, Barcelona and Paris have made significant strides towards sustainable urbanisation, their policies have also demonstrated social and economic challenges, such as rising property prices or gentrification, which can exclude lower-income residents from fully reaping the benefits. At the same time, principles of environmental justice emphasise the importance of a fair distribution of resources and opportunities, particularly in historically disadvantaged areas.

Innovative solution: Building Level Sustainability Scoring (BLSS)

In an attempt to overcome these limitations, a new framework was developed to generate Building Level Sustainability Score (BLSS), which integrates equality into the 15-minute city framework. This approach, published in partnership with RMIT University, represents a robust tool for data-driven decision-making, which helps urban planners and policymakers promote equitable and sustainable urban growth.

The BLSS methodology uses advanced geospatial modeling and machine learning techniquesIn the case of Hong Kong, this approach was implemented using data from over 40,000 residential buildings and 100 government departments. It includes the collection of data from OpenStreetMaps, the census, and extensive facility data (such as schools, hospitals, aged and child care facilities, police, and waste management).

How BLSS works:

• Mapping SDGs indicators: The system maps data to 17 Sustainable Development Goals (SDGs) and their 232 unique building-level target indicators. Some indicators, such as the gender pay gap (SDG 8.5.2) or access to affordable medicines (SDG 3.b.3), contribute to multiple SDGs, reflecting the interconnectedness of urban challenges.
• Calculating the shortest distance: Using Python libraries, distances between buildings and nearby facilities are calculated, providing a comprehensive assessment of urban accessibility.
• Calculating the SDGs score: The scores for each individual indicator are first normalized on a scale of 0 to 10. These normalized scores are then summed to obtain the building’s score for that SDG, and finally the overall SDG score for the building is calculated. The validity of the methodology was verified by comparison with real-world UN scores for Hong Kong, with the results confirming high accuracy.

Key insights and applications: The BLSS framework provides detailed analysis at multiple levels – buildings, districts, and regions.

• Identifying inequalities: Using statistical measures such as Gini coefficients, inequalities in access and distribution of resources are identified. For example, in Hong Kong, high Gini coefficients were found in waste management and elderly care, signaling significant inequalities in access. Differences were also found between age groups and regions, with older age groups facing greater inequalities in access to sustainable resources.
• Scenario simulations: A new simulation framework allows analyzing the impact of infrastructure development on BLSS. For example, the simulation showed that the introduction of an integrated facility for senior and child care led to a significant improvement in the sustainability score for both components. This provides decision makers with a methodological basis for optimizing equipment configuration.
• Targeted housing interventions and recommendations: Based on the insights gained, urban planners and policymakers can formulate evidence-based policies that promote economic prosperity, social equity, and environmental protection. The framework also includes a machine learning-based housing recommendation system that integrates user preferences with quantifiable criteria.

Methodology BLSS is adaptable to urban areas around the world, including cities with limited data availability, by leveraging data from Open Street Maps. This framework effectively addresses challenges related to urban planning, infrastructure optimization, and inclusive growth, helping to create smart cities that adapt and thrive based on real-time data. JRi