Topological Enumeration Heuristics in Constraint-Based Space Layout Planning

Abstract

We deal with the problem of space layout planning. We present an approach of architectural CAD softwares which starts with design specifications in terms of constraints over spaces. This approach is based on an intermediate topological solutions level thanks to specific topological enumeration heuristics. Then the prototype ARCHiPLAN that we have developed enumerates all the topologically different solutions, without presuming any precise definitive dimension. These solutions correspond to the sketching step that an architect carries out from the Design specifications during a preliminary design in architecture. Next, the architect may evolve in this topological solution space, which is of rather small size; afterwards, he can choose some of them for a more detailed study. For each topological solution, the best dimensional solution is determined from cost, useful surface or wall length. This topological solutions level limits the combinatorial explosion and fits to the designer habitudes.