Hybridization of an evolutionary algorithm and simulations of co-evolutionary design processes for early-stage building spatial design optimization

Automation in Construction, vol. 124, pp. 103522

Abstract

Three methods for early-stage building spatial design optimization are presented, demonstrated, and compared for their qualities and limitations. The first, an evolutionary algorithm, can find well-distributed approximations of the Pareto front, but it uses many design evaluations and it can only explore a limited part of the entire design search space (i.e. the collection of all possible design solutions). The second, simulations of co-evolutionary design processes, can find improved design solutions relatively fast within an unrestricted design search space, however, they typically only find discretely distributed Pareto front approximations. For the third method, hybridization is proposed to combine the first two methods into two new hybrid methods, such that their advantages are combined and their disadvantages are diminished. The methods have been applied in an initial case study, which shows that hybridization can improve search efficiency and speed, and it can search larger design search spaces.

Authors 4

1. Sjonnie Boonstra corresponding

   Eindhoven University of Technology

   Affiliation as printed

   Eindhoven University of Technology, The Netherlands

2. Koen van der Blom Aachen Leiden Institute of Advanced Computer Science

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science, Leiden University, The Netherlands

3. Hèrm Hofmeyer

   Eindhoven University of Technology

   Affiliation as printed

   Eindhoven University of Technology, The Netherlands

4. Michael Emmerich Aachen Leiden Institute of Advanced Computer Science

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science, Leiden University, The Netherlands