Generative Design Research Articles

Integrating Deep Learning with Generative Design and Topology Optimization for Efficient Additive Manufacturing

Abstract. Additive manufacturing (AM) through generative design and topology optimisation creates complex, lightweight structures with exceptional material efficiency and structural integrity. When coupled with deep learning functionality, generative design and topology optimisation can explore broader design spaces and optimise more efficiently, creating novel AM structures that utilise material more efficiently and have better strength and performance than their counterparts created through conventional AM methods. The study tackles how deep learning models such as convolutional neural networks (CNNs) can be integrated into generative design and topology optimisation and how these integration help optimise material usage, production time and performance. Case studies from the aerospace, automotive, and healthcare industries exemplify how these synergies resulted in more resilient, cost-effective designs that would not have been possible through conventional AM approaches. The study focuses on material usage efficiency, reduction in production time and performance improvement to showcase how deep learning integrations enhance the process from design conceptualisation, through iterations, to final production.

Innovative Applications of Parametric Design and Digital Tools in Architecture: Exploring the Integration of Generative Design and BIM Technology

Abstract. Through parametric design and generative design, as well as the broader Building Information Modeling (BIM) approach, there is now a whole body of new tools being used by architects to are opening up new ways of designing buildings in innovative, efficient and sustainably responsible ways. This paper presents the emergence of parametric design and its evolution from simple structural optimisations to complex fulling the design of complete buildings and entire urban areas. The paper discusses the integration of both parametric design and generative design with Building Information Modelling (BIM). This technology pairing enables architects to research and test hundreds of design alternatives and optimise for particular variables, such as structural integrity, material efficiency and energy use. This paper showcases a number of case studies of how these technologies are applied to develop innovative and sustainable urban environments (eg, Sidewalk Toronto) and discusses the challenges associated with adopting these technologies, such as computational requirements and a lack of digital interoperability between these technologies. The paper concludes by identifying the innovation opportunities for sustainable development and the future of architectural practice that these technologies are offering.

Deep generative design of RNA aptamers using structural predictions.

RNAs represent a class of programmable biomolecules capable of performing diverse biological functions. Recent studies have developed accurate RNA three-dimensional structure prediction methods, which may enable new RNAs to be designed in a structure-guided manner. Here, we develop a structure-to-sequence deep learning platform for the de novo generative design of RNA aptamers. We show that our approach can design RNA aptamers that are predicted to be structurally similar, yet sequence dissimilar, to known light-up aptamers that fluoresce in the presence of small molecules. We experimentally validate several generated RNA aptamers to have fluorescent activity, show that these aptamers can be optimized for activity in silico, and find that they exhibit a mechanism of fluorescence similar to that of known light-up aptamers. Our results demonstrate how structural predictions can guide the targeted and resource-efficient design of new RNA sequences.

CACHE Challenge #1: Targeting the WDR Domain of LRRK2, A Parkinson's Disease Associated Protein.

The CACHE challenges are a series of prospective benchmarking exercises to evaluate progress in the field of computational hit-finding. Here we report the results of the inaugural CACHE challenge in which 23 computational teams each selected up to 100 commercially available compounds that they predicted would bind to the WDR domain of the Parkinson's disease target LRRK2, a domain with no known ligand and only an apo structure in the PDB. The lack of known binding data and presumably low druggability of the target is a challenge to computational hit finding methods. Of the 1955 molecules predicted by participants in Round 1 of the challenge, 73 were found to bind to LRRK2 in an SPR assay with a KD lower than 150 μM. These 73 molecules were advanced to the Round 2 hit expansion phase, where computational teams each selected up to 50 analogs. Binding was observed in two orthogonal assays for seven chemically diverse series, with affinities ranging from 18 to 140 μM. The seven successful computational workflows varied in their screening strategies and techniques. Three used molecular dynamics to produce a conformational ensemble of the targeted site, three included a fragment docking step, three implemented a generative design strategy and five used one or more deep learning steps. CACHE #1 reflects a highly exploratory phase in computational drug design where participants adopted strikingly diverging screening strategies. Machine learning-accelerated methods achieved similar results to brute force (e.g., exhaustive) docking. First-in-class, experimentally confirmed compounds were rare and weakly potent, indicating that recent advances are not sufficient to effectively address challenging targets.

Façade Design Pattern Optimization Workflow Through Visual Spatial Frequency Analysis and Structural Safety Assessment

As the demand for highly efficient yet aesthetically pleasing, complex building envelope structures is rising worldwide, computational analysis and generative design tools are becoming ever so relevant. Previous methods for achieving a natural distribution of structural or shading elements in non-uniform façades are mostly based either on computer-generated pseudo-randomness or a literal biomorphic approach where a naturally occurring pattern is directly projected on the façade surface. As an alternative, this research introduces a novel technique for optimisation that utilises a two-dimensional Power Spectrum Analysis, suitable for numerically assessing the alignment of designed geometry with natural patterns. By integrating this optimisation method into the design process, the façade pattern generation can be automated and optimal design can be selected by evaluating multiple design solutions. Instead of using repetitive geometrical patterns or generated pseudo-randomness, patterns objectively similar to those occurring in nature can be created without directly copying natural structures. The distribution of the structural and shading elements controls the way natural light permeates the building and, considering the data gathered from images of natural scenes, this method can be used to design structures not only with optimal structural and energy performance but also with visual and psychological occupant comfort in mind.

Role of Design Thinking in Socio-Technical Activity for Product Manufacturing

Human interaction intensively impacts design in product manufacturing through design thinking, interpreted product recognition and perceptual product experience (PPE). While the dominant digitalization and artificial intelligence in generative design present significant opportunities for innovation and optimization, there is a gap in adopting holistic, human-centered approaches. Despite numerous advancements, these methods, including design thinking, are not widely practiced, and there is limited discussion on effective design methodologies. This research aims to address this issue by investigating the influence of human interaction on design processes, integrating socio-technical principles, and analyzing how product form affects user perception and recognition.

TamGen: drug design with target-aware molecule generation through a chemical language model

Generative drug design facilitates the creation of compounds effective against pathogenic target proteins. This opens up the potential to discover novel compounds within the vast chemical space and fosters the development of innovative therapeutic strategies. However, the practicality of generated molecules is often limited, as many designs focus on a narrow set of drug-related properties, failing to improve the success rate of subsequent drug discovery process. To overcome these challenges, we develop TamGen, a method that employs a GPT-like chemical language model and enables target-aware molecule generation and compound refinement. We demonstrate that the compounds generated by TamGen have improved molecular quality and viability. Additionally, we have integrated TamGen into a drug discovery pipeline and identified 14 compounds showing compelling inhibitory activity against the Tuberculosis ClpP protease, with the most effective compound exhibiting a half maximal inhibitory concentration (IC50) of 1.9 μM. Our findings underscore the practical potential and real-world applicability of generative drug design approaches, paving the way for future advancements in the field.

Global, Green, and Generative Design

Global, Green, and Generative Design

Automated construction site layout design system for prefabricated buildings using transformer based conditional GAN

Construction site layout plans (CSLP) are crucial for efficient prefabricated construction project management. Traditional manual design process is costly and time-consuming, while optimization methods heavily depend on expert knowledge. Recent advancements in deep generative models present promising alternatives. However, their application to the generation of prefabricated construction site layouts is hindered by several challenges, including limited datasets, significant overlap between facilities, and the necessity to generate layouts based on fixed facilities with specific attributes such as minimal transportation costs. These challenges constrain the efficacy and applicability of the generated layouts. To address these issues, this study introduces an innovative automated generative design system for prefabricated construction site layouts, leveraging a novel Transformer-based conditional generative adversarial network (GAN). The data preparation module of the system collects and augments layout data for training. The CSLGAN module is designed to generate layouts that conform to spatial constraints and desired attributes, with minimal facility overlap. Furthermore, this study establishes benchmarks in terms of model capacity and specialized performance metrics. Extensive experiments demonstrate the effectiveness of the proposed system in automated construction site layout generation.

Immediate Correction of Idiopathic Scoliosis With Nighttime Braces Created by a Fully Automated Generative Design Algorithm: A Randomized Controlled Crossover Trial.

Single-center, double-blinded, prospective crossover randomized controlled trial. To clinically validate the efficacy of nighttime braces designed automatically by a generative design algorithm to treat idiopathic scoliosis (IS). The tested hypothesis was the clinical equivalence of immediate in-brace correction for the new automatically generated brace design versus a standard Providence-type brace. Documented efficacy of brace treatment varies between centers, and depends on the empirical expertise of the treating orthotist. Our group previously developed a fully automated generative brace design algorithm that leverages a patient-specific finite-element model (FEM) to optimize brace geometry and correction before its fabrication. Fifty-eight skeletally immature patients diagnosed with IS, aged between 10 and 16 years were recruited. All patients received both a nighttime brace automatically generated by the algorithm (test) and a Providence-type brace designed by an expert orthotist (control). Radiographs were taken for each patient with both braces in a randomized crossover approach to evaluate immediate in-brace correction. The targeted 55 patients (48 females, 7 males) completed the study. The immediate Cobb angle correction was 57% 19 (test) versus 58% 21 (control) for the main thoracic (MT) curve, whereas it was 89% 25 (test) versus 87% 28 (control) for the thoracolumbar/lumbar (TLL) spine. The immediate correction with the test brace was noninferior to that of the Control brace ( P 0.001). The order in which the braces were tested did not have a residual effect on the immediate correction. The fully automated generative brace design algorithm proves to be clinically relevant, allowing for immediate in-brace correction equivalent to that of braces designed by expert orthotists. Patient 2 years follow-up will continue. This method's integration could help design and rationalize the design of braces for the treatment of IS. Level 2.

Generative artificial intelligence and optimisation framework for concrete mixture design with low cost and embodied carbon dioxide

This research presents a generative Artificial Intelligence (AI) and design framework that integrates machine learning (ML) and optimisation methodologies to discover new concrete mixture designs. Unlike traditional ML models that predict based on existing data, this framework innovatively generates new concrete mix designs that meet specific requirements such as strength, cost-efficiency, and reduced embodied CO2. To propose a powerful and reliable generative AI model, several advanced ML algorithms were considered, e.g., CatBoost, XGBoost, and LGBM. These models were trained on a unique dataset consisting of 4,936 data points collected from five different batching plants and have not been published yet. Bayesian Optimisation was employed to fine-tune model hyperparameters, resulting in the most effective models attaining R2 values of 0.94 and 0.89 for raw and grouped data, respectively. To verify the trained generative AI model, a case study was conducted, in which the model was requested to provide designs of a mix with pre-determined strength and optimised cost and embodied CO2. The mix designs generated by the framework were successfully validated through experimental tests, corroborating the predictive outcomes. The research culminated in the development of a web application, a tool crafted to streamline the concrete mixture design and optimisation process. This generative AI design framework can be applied to many other aspects of material design and engineering problems.

Deep Analogical Generative Design and Evaluation: Integration of Stable Diffusion and LoRA

Abstract The rapid evolution of generative design through artificial intelligence has opened new avenues for innovative product styling. Integrating this efficient generative technology with established professional theories presents a novel challenge in contemporary international design research. In response to this challenge, this paper introduces a pioneering and collaborative approach for the swift generation of automobile styling designs. The primary objective is to investigate an intelligent generation method that incorporates analogical reasoning and Stable Diffusion to support industrial designers in innovating product styling. This study scrutinizes traditional analogical reasoning design alongside the intelligent analogical reasoning design proposed herein, elucidating the distinctions through multidimensional comparisons using illustrative examples. The proposed methodological framework encompasses several key steps. Initially, a dataset comprising branded automobile images is meticulously constructed. Subsequently, an exclusive style model is trained leveraging Stable Diffusion techniques, coupled with advanced computer graphics and machine learning methodologies. Following this, design requirements are inputted, facilitating intelligent analogical reasoning design across multiple spatial dimensions to yield diverse and innovative automobile styling solutions. Finally, eye-tracking experiments are conducted to quantitatively compare the traditional analogical reasoning design approach with the Stable Diffusion-based analogical reasoning design method. The results substantiate that the latter effectively generates innovative and diversified automobile design solutions. This research contributes to enhancing the quality of automobile styling design, optimizing the design efficiency of enterprises, and catalyzing innovation in the automobile styling design process.

Characterization, generative design, and fabrication of a carbon fiber-reinforced industrial robot gripper via additive manufacturing

Robot grippers are crucial components across various industrial applications, requiring special design and production for obtaining the optimal performance. Conventional plastic injection moulding techniques fall short in achieving the specificity needed for these grippers. To address this challenge, current paper focuses on developing a robot gripper using carbon fiber-reinforced polyamide with a next-generation composite filament and employing the innovative Generative Design technique. In the work, we began by characterizing and optimizing the composite material specifications. Then, the tensile strength and fracture mechanics of standard samples based on printing parameters, applying Taguchi experimental design for optimization were evaluated. Analysis of Variance (ANOVA) was used for factor analysis to fine-tune the process. Using the Generative Design technique, we determined optimal geometries, which were then fabricated through Fused Deposition Modeling (FDM). As a result, the optimization efforts led to significant improvements i.e., tensile strength increased from 103.2 to 116 MPa, and the elasticity modulus from 8386 to 8990 MPa. In practical industrial applications, we achieved a reduction in material weight from 14 to 4 g, lowered production costs from $5.16 to $1.50, and cut production time from 58 to 28 min. This study presents a validated method for developing industrial products with reduced material usage and costs, promoting sustainable production practices.

Life cycle assessment and generative design: development of a national LCA tool for exterior walls

PurposeThe promotion of sustainable design is demanded globally. The life cycle assessment (LCA) proved its reliability in this mission, but the difficulty and time required to apply it discouraged designers. This research aims to integrate LCA into the building design process through a software tool, taking advantage of generative design features. This will facilitate decision-making by architects and construction professionals.Design/methodology/approachThe study develops the EGY-LCA (http://egy-lca.com/). This prototype tool suggests exterior wall design alternatives for residential buildings in Egypt, using the environmental impact indicators of LCA data and other criteria related to national codes, materials, construction methods and required thermal resistance. Within a generative design process, the algorithm tests every possible wall method with materials and thickness combinations for each layer in compliance with inputs. The paper begins by explaining the tool’s working method. Afterward, different sets of inputs are examined and the values of the resultant environmental impacts of several suggested wall solutions are statistically analyzed. The application demonstrates the importance of the generative design tool. Proposing several solutions based on a set of inputs facilitates the selection of sustainable choices and allows comparisons between alternatives.FindingsThe prototype experiment confirms the research hypothesis. Unlike the available LCA tools, architects can make decisions with limited LCA experience if the data and equations are integrated into a generative design tool. The prototype proves its applicability for exterior wall alternatives.Research limitations/implicationsThe prototype is the initial step toward a whole-building LCA tool. It includes limited LCA stages and materials for the external wall. Future research is required to expand this parametric tool concept to include all the building components. The framework in Section 5 proposes a visualization.Practical implicationsThe prototype tool: EGY-LCA (http://egy-lca.com/). The value added to the design and construction sectors through the uncomplicated LCA application is fostering sustainable design, generative design tools can achieve this.Originality/valueThe novelty of this work is that it is the first initiative offering a parametric LCA tool. It promotes the application of LCA at the design stage using generative design, which contributes to sustainable development.

Artificial intelligence‐driven sustainability: Enhancing carbon capture for sustainable development goals– A review

AbstractArtificial intelligence (AI) and environmental points are equally important components within the response to local weather change. Therefore, based on the efforts of reducing carbon emissions more efficiently and effectively, this study tries to focus on AI integration with carbon capture technology. The urgency of tackling climate change means we need more advanced carbon capture, and this is an area where AI can make a huge impact in how these technologies are operated and managed. It will minimize manufacturing emissions and improve both resource efficiency as well as our planet's environmental footprint by turning waste into something of value again. Artificial intelligence could be leveraged to analyze huge data sets from carbon capture plants, searching for optimal system settings and more efficient ways of identifying patterns in the available information at a larger scale than currently possible. In addition, AI incorporated sensors and monitoring mechanisms in the supply chain can identify any operational failure at reception itself allowing for timely action to protect those areas. Artificial intelligence also helps generative design for carbon capture materials, which allows researchers to explore new types of carbon‐absorbing material, including metal–organic frameworks and polymeric materials that are important in industrial CO2, such as moisture. In addition, it increases the accuracy of reservoir simulations and controls CO2 injection systems for storage or enhanced oil recovery. Through applying AI algorithms on reservoir geology, production performance and real‐time data this study would like to facilitate the optimization of injection processes as well as minimize CO2 emissions while assuring a maximum efficiency. Artificial intelligence integrates with renewable‐based carbon capture efforts that can be employed by AI‐driven smart grid systems to improve carbon capture methods.

Generative Design and Experimental Validation of Non-Fullerene Acceptors for Photovoltaics

Generative Design and Experimental Validation of Non-Fullerene Acceptors for Photovoltaics

3D constructing: Exploring the potential of 3D concrete and clay printing with generative design for architectural innovation

The architectural and construction sectors are experiencing a transformative shift with the advent of 3D constructing. This research delves into the potential of 3D printing technologies when combined with generative design principles to reshape traditional construction methods. Through a comprehensive lens, the study evaluates the generative design workflow, decision-making algorithms, machines capacities and material considerations. Results indicate that, when synergized with generative design, these technologies offer sustainable, efficient, and tailored solutions. The use of accesible materials, such as clay and cement, demonstrates their superiority over conventional alternatives. However, the challenge lies in harmoniously integrating these innovations into architectural practice, ensuring a balance between cutting-edge advancements and practical application. Conclusively, this new approach heralds a future of sustainable and user-centric design, but its full potential can only be tapped with a holistic strategy that encompasses both design and material facets.

Radar waveform generative design method

AbstractTo address the issue of radar detection performance degradation caused by interference in complex electromagnetic environments, this paper proposes a generative waveform design method and demonstrates its advantages. This method fully exploits waveform degrees of freedom to achieve a larger design space, generating a multitude of agile waveform parameters within the constraint envelope through algorithmic processes. This approach enables the radar to maintain detection performance in complex electromagnetic environments. Compared with the traditional waveform design method, the simulation experiment verifies the effectiveness of the generative waveform design method.

Structural optimization of hydraulic valve block based on generative design and selective laser melting process simulation

Structural optimization of hydraulic valve block based on generative design and selective laser melting process simulation

Denoising diffusion probabilistic models for generative alloy design

Inverse material design is an extremely challenging optimization task made difficult by, in part, the highly nonlinear relationship linking performance with composition. Quantitative approaches have improved significantly owing to advances in high throughput experimentation and computational thermodynamics. However, existing physics-based tools are mostly forward models; input a chemistry and obtain a prediction. More recently the materials community has leveraged advances in the machine learning community to establish novel inverse design frameworks. Very recently denoising diffusion probabilistic models have been shown to be extremely powerful generators producing synthetic data of various modalities e.g. images, text, audio, tables, etc.. In this work a novel framework for alloy design and optimization is proposed leveraging these class of models. Five key generative tasks are demonstrated (1) unconditional generation (2) composition conditioned generation (3) property conditioned generation (4) multi-feedstock conditioned generation and (5) generative optimization. These methods were tested on three case studies: high entropy alloy design, superalloy binder jet additive manufacturing, and in-situ dual-feedstock wire-arc additive manufacturing. Results indicate that the established models are extremely flexible, expressive, and robust. The architecture’s flexibility and training procedure empower the model to learn complex intra-compositional and composition-property relationships. Furthermore, the probabilistic nature of these models makes them well suited for addressing solution non-uniqueness and tackling uncertainty quantification tasks. While the fidelity and quantity of the underlying training data is paramount, we envision that future alloy design frameworks will make extensive use of these kinds of machine learning models as “search” tools bolstering the utility of experimental and computational approaches.