Flexible building stock modelling with array-programming

Many building stock models employ archetype-buildings in order to capture the essential characteristics of a diverse building stock. However, these models often
require multiple archetypes, which make them inflexible. This paper proposes an array-programming based model, which calculates the heat demand of each building separately. This approach makes it possible to study any subset of the building stock without loss of information. Moreover, the calculated heat demand can readily be compared to a registered consumption in individual buildings to account for discrepancies. Results show, that we tend to overestimate potential energy-savings, if we do not consider these discrepancies. The proposed model makes it possible to compute and visualize potential energy-savings in a flexible and transparent way.