Computation-based approaches in design have emerged in the last decades and rapidly became popular among architects and other designers. Design professionals and researchers adopted different ...terminologies to address these approaches. However, some terms are used ambiguously and inconsistently, and different terms are commonly used to express the same concept. This paper discusses computational design (CD) and proposes an improved and sound taxonomy for a set of key CD terms, namely, parametric, generative, and algorithmic design, based on an extensive literature review from which different definitions by various authors were collected, analyzed, and compared.
Parametric Modeling, Generative Design, and Performance-Based Design have gained increasing attention in the AEC field as a way to create a wide range of design variants while focusing on performance ...attributes rather than building codes. However, the relationships between design parameters and performance attributes are often very complex, resulting in a highly iterative and unguided process. In this paper, we argue that a more goal-oriented design process is enabled by an inverse formulation that starts with performance attributes instead of design parameters. A Deep Conditional Generative Design workflow is proposed that takes a set of performance attributes and partially defined design features as input and produces a complete set of design parameters as output. A model architecture based on a Conditional Variational Autoencoder is presented along with different approximate posteriors, and evaluated on four different case studies. Compared to Genetic Algorithms, our method proves superior when utilizing a pre-trained model.
•Complex 2D/3D models hinder understanding of relationships in Generative Design.•Deep Conditional Generative Design learns joint distribution for targeted designs.•Inverse formulation enables precise control for Performance-Based Generative Design.•Our Conditional Variational Autoencoder is evaluated against Genetic Algorithms.•Expressive posterior improves performance; partial conditioning shows promise.
Generative Adversarial Networks (GANs) are a type of deep neural network that have achieved many state-of-the-art results for generative tasks. GANs can be useful in the built environment, from ...processing large-scale urban mobility data and remote sensing images at the regional level, to performance analysis and design generation at the building level. We analyzed 100 articles to provide a comprehensive state-of-the-art review on how GANs are currently applied to solve challenging tasks in the built environment. Our results show that: (i) GANs are replacing older methods in some problems and setting state-of-the-art performances; (ii) GANs are opening new frontiers in previously overlooked problems, such as automatically generating spatially accurate floorplan layouts; (iii) GANs can be applied to different scales in the built environment, from entire cities to neighborhoods and buildings; and (iv) GANs are being used in a variety of problems and data types, from remote sensing data augmentation, vector data generation, spatio-temporal data privacy protection, to building design generation. In total, there are 26 unique application domains enabled by GANs; (v) however, one common challenge in this field currently is the lack of high-quality datasets curated specifically for problems in the built environment. With more data in the future, GANs could potentially produce even better results than today.
Structural design is a search for the best trade-off between multiple architecture, engineering, and construction objectives, not only mechanical efficiency or construction rationality. Producing ...hybrid designs from single-objective optimal designs to explore multi-objective trade-offs is common in the design of structural forms, constrained to a single parametric design space. However, producing topological hybrids offers a more complex challenge, as a combinatorial problem that is not encoded as a finite set of real numbers but as an unbonded series of grammar rules. This paper presents a strategy for the generation of hybrid designs of quad-mesh pattern topologies for surface structures. Based on a quad-mesh grammar, an algebra is introduced to measure the distance between designs, find their similar features, and enumerate designs with different degrees of topological similarity. Structural design applications are shown to highlight the use of topologically hybrid designs as a surrogate for obtaining multi-objective trade-offs.
•Topological similarity is used to explore multi-objective performance trade-offs.•A novel algebra for generative design of quad-mesh patterns is introduced.•Several structural design applications illustrate this approach.
Recent advances in generative modeling, namely Diffusion models, have revolutionized generative modeling, enabling high-quality image generation tailored to user needs. This paper proposes a ...framework for the generative design of structural components. Specifically, we employ a Latent Diffusion model to generate potential designs of a component that can satisfy a set of problem-specific loading conditions. One of the distinct advantages our approach offers over other generative approaches is the editing of existing designs. We train our model using a dataset of geometries obtained from structural topology optimization utilizing the SIMP algorithm. Consequently, our framework generates inherently near-optimal designs. Our work presents quantitative results that support the structural performance of the generated designs and the variability in potential candidate designs. Furthermore, we provide evidence of the scalability of our framework by operating over voxel domains with resolutions varying from 323 to 1283. Our framework can be used as a starting point for generating novel near-optimal designs similar to topology-optimized designs.
•Latent diffusion model for generating 3D structural component designs.•Framework for generating component designs consistent with topology optimization.•Generated designs have similar (near-optimal) strain energy to SIMP designs.•Large scale 3D voxel dataset for structural topology optimization.