Buildings and constructions account for a substantial portion of global energy use and greenhouse gas emissions. Indeed, globally, buildings and structures are responsible for 36% of the final energy use and 39% of the energy-related CO2 emissions. Therefore lowering the energy consumption in this sector is a crucial component for achieving a sustainable energy transition. The need for accurate estimations of current energy use and predictions of future energy demand is evident. This calls for precise and high-quality energy models of building stocks at a large scale.
Reducing energy demand in the building sector, or improving energy efficiency in buildings, has proved to be an extremely complex undertaking. Moreover, the building sector is characterized by its great heterogeneity, with buildings of different types, sizes, and operational uses. Owners, tenants, contractors, architects, researchers, regulators, and politicians, among others, all have different motivations for energy efficiency measures, which results in barriers to improving energy efficiency.
The building sector, a lever for the energy transition
Energy demand in buildings is determined by various factors such as climate, building characteristics, or occupant behavior. In this context, one primary method to analyze the current building stock is integrating simulation on a large spatiotemporal scale. This, however, is very demanding in terms of computing resources. Detailed simulations for a large number of buildings, i.e., thousands or millions, are time-consuming and data-intensive when considering multiple climate, weather, or retrofit scenarios.
To meet the challenge of simulating the energy demand of many buildings on a regional or national scale, grouping and clustering techniques are commonly applied. These techniques enable simulations to be carried out on a subset of representative buildings, also widely referred to as building archetypes, which can then be used for scaling. The advantage of relying on a simplified representation of the building stock is not only the reduction in computational costs for analysis but also the reduction in the need for detailed input data for individual buildings, which is often protected and challenging to obtain.
Clustering is one of the most common general techniques for building performance data. It is used to generate subgroups of similar types of observations based on certain features. The archetype aggregation approach aims at defining typical buildings that represent the studied stock, which is a preferential option under the circumstances of scarce/unavailable data. Clustering buildings by type and classifying actions based on asset types can be a valuable approach to save time and money in various industries, such as real estate management, facilities management, or urban planning.
How to create building archetypes?
Segmentation of the building stock and identification of building archetypes are critical elements in this process, which can be more or less complex. There is no standard methodology for this process, and it may involve several different approaches and techniques. Nevertheless, the process generally occurs in three main stages: classification (or segmentation), characterization, and calibration.
Firstly, buildings are classified, which involves sorting them into sub-groups to represent differences in typology or simulation behavior better. Typical segmentation parameters can include building age, type, and size of HVAC (heating, ventilation and cooling) systems. After the classification of the building stock, the identified archetypes are characterized by all relevant building parameters, not just those used for segmentation. Finally, the archetypes are calibrated, and the results are aggregated at a larger scale or the appropriate spatial resolution.
When should you apply building archetypes clustering?
Here are six reasons why clustering buildings by type is useful:
1. Urban planning
It helps urban planners understand the distribution and composition of different building types within a city or region. This information can advise decisions regarding zoning regulations, land use planning, and the allocation of resources for infrastructure development.
2. Real estate market analysis
It enables real estate professionals to analyze market trends and property values based on building types. This information can be used to make educated decisions regarding investments, property development, and pricing strategies.
3. Infrastructure management
It helps in managing and maintaining infrastructure efficiently. Different types of buildings may have different requirements regarding utility connections, maintenance schedules, and service providers. By grouping similar buildings together, infrastructure management can be optimized, resulting in cost savings and improved service delivery.
4. Energy efficiency
Different building types have varying energy consumption patterns. Clustering buildings by type allows for the identification of energy-efficient practices and the development of targeted strategies for energy conservation. It enables the implementation of building codes, standards, and policies that promote sustainable practices and reduce energy waste.
5. Risk assessment
Clustering buildings by type can also help assess risks associated with different building types. For example, certain building types may be more prone to specific maintenance issues or safety hazards.
Focus on Climate Risk
Clustering buildings based on their vulnerability to specific climate-related risks (such as floods, hurricanes, or heatwaves) allows for developing specific resilience strategies. By identifying clusters of buildings that face similar risks and potential impact, we can efficiently allocated resources to implement mitigation measures, such as improved drainage systems, flood barriers, or green infrastructure.
This targeted approach saves time by focusing efforts where they are most needed and thus enables the prioritization of actions. This approach thereby enhances the overall efficiency of climate resilience efforts. In short, it provides insight into how future climate conditions might affect buildings and their energy use.
Clustering buildings allows for easier benchmarking against similar types of buildings. Organizations can compare the performance of their assets against others within the same cluster to identify areas for improvement.
Overall, clustering buildings provides valuable insights into the composition, characteristics, and specific needs of different types of buildings. This information can support decision-making processes, optimize resource allocation, and improve the overall management and efficiency of urban environments.
The success of this approach heavily relies on the quality and relevance of the data collected, as well as the accuracy of the clustering and classification models. This requires a regular review and refine the models to ensure their effectiveness and adapt them as building portfolios evolve over time.
Deepki has some of the most qualified data scientists that have been able to build a cutting-edge methodology with a genuinely scientific approach: ensure a scientific approach to the quality checks that exist behind the scenes.