Analysing optimisation data for multicriteria building spatial design

Domain experts can benefit from optimisation simply by getting better solutions, or by obtaining knowledge about possible trade-offs from a Pareto front. However, just providing a better solution based on objective function values is often not sufficient. It is desirable for domain experts to understand design principles that lead to a better solution concerning different objectives. Such insights will help the domain expert to gain confidence in a solution provided by the optimiser. In this paper, the aim is to learn heuristic rules on building spatial design by data-mining multi-objective optimisation results. From the optimisation data a domain expert can gain new insights that can help engineers in the future; this is termed innovization. Originally used for applications in mechanical engineering, innovization is here applied for the first time for optimisation of building spatial designs with respect to thermal and structural performance.