Layout Generation Research Articles

Exploring spatial reasoning performances of CNN on linear layout dataset

Abstract Spatial reasoning, a fundamental aspect of human intelligence, is essential for machine learning models to understand and interpret object relationships. It is crucial for numerous real-world applications, ranging from autonomous navigation to urban planning. The lack of comprehensive datasets limits the development and evaluation of models that can effectively handle spatial reasoning tasks. Existing datasets often contain complex spatial reasoning problems with overlapping spatial relationships, making it challenging to diagnose specific aspects that a model struggles with. We address this gap by introducing a new dataset of linear layouts. This dataset is systematically designed to exhibit a range of spatial relations and complexity levels. Analyzing spatial reasoning through linear layout generation offers a more structured and manageable approach to understanding how models learn and interpret spatial relationships. Linear layout generation has broad applicability and is of fundamental importance in design and optimization. To benchmark dataset, we develop LinLayCNN, a generic data-driven method that applies shallow, one-dimensional convolutional neural network (CNN), to generate linear layouts in an iterative process. Experimental results reveal that LinLayCNN can effectively solve fundamental spatial challenges even with the relatively small size of the training set. It is capable of precise object placement, making it a robust tool for linear layout generation. Current layout generation methods focus on domain-specific solutions, often fail to maintain the precision needed for technical domains, such as accurate sizing, and object counting. They also require a substantial amount of data to function effectively. LinLayCNN overcame these issues. This study also enhances our understanding of CNNs' capabilities in spatial reasoning, highlight their potential to advance the field of layout generation. As a result, our approach establishes a clear benchmark for evaluating spatial reasoning and aids in development of models that can more effectively understand and reason about space.

SWA: SoftWare for Analog Design Automation

We have developed SWA: SoftWare for Analog design automation. Its commands can describe analog and mixed-signal (AMS) layouts and schematics to replace the graphic editor with a program reflecting the knowledge of design experts. Also, it is able to utilize variables to parameterize schematics and layouts to fulfill design needs. We programmed a 10b 1 GS/s DAC using SWA with 8.3 K lines of code, which is about 1/10 compared with conventional programs. The programmed DAC is configurable with multiple voltages and multiple resolutions from 4 to 12 bits. The DAC schematic and layout generation with DRC and LVS SWA API can be finished in about 1 min.

ArchiDiffusion: A novel diffusion model connecting architectural layout generation from sketches to Shear Wall Design

Shear wall structures are widely used in residential and office buildings, but the overall design process is complex and inefficient. This study introduces a novel diffusion-based model, named ArchiDiffusion, designed to revolutionize the generation of architectural floor plans from user-drawn sketches. ArchiDiffusion employs a two-stage process that ensures both creativity and practicality, setting a new standard in automated design. Embedded within the End-to-End Shear Wall Design (EESD) framework, ArchiDiffusion seamlessly integrates with existing structural design processes, including the previously developed StructDiffusion model, to streamline the traditionally complex and inefficient design workflow. Experimental results demonstrate that ArchiDiffusion significantly outperforms existing models in layout generation, reducing design time and enhancing project quality. A case study reveals that the EESD framework, powered by ArchiDiffusion, increases the design efficiency by 50 times compared to traditional workflows. Notably, ArchiDiffusion achieves a 9.6%–15.2% improvement in layout generation accuracy over traditional GAN-based models, while the EESD framework demonstrates an approximate 20% enhancement in overall building performance, as measured by energy efficiency and spatial optimization metrics. This research highlights the significant potential of ArchiDiffusion in intelligent design processes for shear wall residential buildings, offering a promising solution to the challenges faced in integrating architectural and structural design.

Line-of-Sight Probability Models for 5G/6G Millimeter-Wave Networks: A Review

In the fifth-generation (5G) and future wireless networks, coverage planning for millimeter-wave (mm-wave) networks in a three-dimensional (3D) urban environment depends on the communication channel and line-of-sight (LoS) probability models. This is especially true when there are big stationary obstructions. The current literature lacks a complete review of the existing LoS probability models. To bridge this gap, this paper reviews modeling approaches and mathematical models of the LoS probability. We discovered that, due to the complexity of modeling a large number of 3D layouts, most existing models have focused on a few urban layouts. We developed a 3D synthetic urban layout generator that is based on game engine technology to simulate any ITU-R P.1410 urban layouts and collect the ray-tracing. We provided a proof-of-concept to show that the game technology could handle the complexity of simulating ray-tracing in 3D urban environments. Moreover, we addressed some of the biggest challenges and gaps in the current LoS probability literature and suggested some valid opportunities.

Enhancing healthcare operations: a systematic literature review onapproaches for hospital facility layout planning.

The appropriate physical layout of hospital services can help resolve management problems by streamlining the work of medical teams, improving the flow of patients between specific areas and the medical support environment. Nevertheless, the academic literature lacks structured research into how the physical layout of hospitals might be improved. Our study aims to fill this research gap, providing information for researchers and professionals who intend to guide the hospital facility layout planning (HFLP) from the steps and prescribed approaches found in the literature. This study analyzes the current literature status and concerning approaches that support HFLP and identifies their strengths and weaknesses. The literature was classified using the following criteria: approaches for layout generation, approaches for layout evaluation and healthcare facility layout outcomes. The hospital facility layout outcomes achieved for each phase served as a basis for identifying a list of strengths and weaknesses for the hospital layout facility generation and evaluation approaches. Readers can refer to this paper to identify the approach that best fits the desired goal and the HFLP step. This is a contribution to current studies into HFLP, and it provides guidelines for selecting the approach to be utilized based on the desired outcome. The paper describes how to conduct an HFLP and lists the strengths and weaknesses of each approach. The research may be used as a strategy for determining which tool is most suited based on the practitioner's target purpose.

Automated urban landscape design: an AI-driven model for emotion-based layout generation and appraisal

The evolution of a city is significantly shaped by the design of its urban landscape. The advancement of artificial intelligence has substantially increased convenience for individuals. This research proposes an urban landscape layout model powered by artificial intelligence that automatically generates urban landscape design based on deep learning (URDDL) with two dimensions: emotional tendency and urban landscape appraisal. The input image represents land use and surrounding road conditions, while the output image depicts the selection of the main entrance and the internal spatial function layout. The Pix2Pix model is trained to learn the internal function layout based on varying land use and road conditions. Additionally, a domain-specific dictionary is constructed using an existing semantic resource vocabulary, where positive and negative sentiment words are compared with their corresponding sentiment values, focusing on categories such as Stimulate, Sense, and Action. Experimental results indicate that the absolute average error of the URDDL model is 91.31%, with a maximum error of 96.87%. The degree of fit is highly appropriate for evaluating the emotional prediction of urban landscapes. The findings demonstrate that the URDDL model outperforms traditional design methods regarding generated results, suggesting its potential for future applications in automated landscape design.

A novel framework for automated warehouse layout generation.

Optimizing warehouse layouts is crucial due to its significant impact on efficiency and productivity. We present an AI-driven framework for automated warehouse layout generation. This framework employs constrained beam search to derive optimal layouts within given spatial parameters, adhering to all functional requirements. The feasibility of the generated layouts is verified based on criteria such as item accessibility, required minimum clearances, and aisle connectivity. A scoring function is then used to evaluate the feasible layouts considering the number of storage locations, access points, and accessibility costs. We demonstrate our method's ability to produce feasible, optimal layouts for a variety of warehouse dimensions and shapes, diverse door placements, and interconnections. This approach, currently being prepared for deployment, will enable human designers to rapidly explore and confirm options, facilitating the selection of the most appropriate layout for their use-case.

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.

Image-aware layout generation with user constraints for poster design

Image-aware layout generation with user constraints for poster design

REVOLUTIONIZING SEMICONDUCTOR DESIGN AND MANUFACTURING WITH AI

The semiconductor industry plays a vital role in driving technological advancements, and the incorporation of AI (Artificial Intelligence) can greatly enhance its efficiency and productivity. Through optimizing material usage and reducing defects, AI can significantly reduce costs and enhance production efficiency and product quality. However, despite the increasing interest in AI applications in the semiconductor industry, comprehensive reviews are lacking to systematically analyze existing research and identify the challenges and opportunities in this field. This review aims to bridge this gap by providing a thorough overview of AI-driven techniques in optimizing semiconductor manufacturing and offering valuable insights for future research directions. The integration of Artificial Intelligence (AI) into chip design marks a transformative phase for the semiconductor industry. Traditional design methodologies, often labor-intensive and time-consuming, are increasingly constrained by human expertise and iterative processes. Generative AI, utilizing advanced machine learning models such as Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), offers innovative approaches to automate and optimize various stages of chip design. This paper examines how generative AI can revolutionize chip design by automating complex tasks, including architecture exploration, circuit optimization, and layout generation. Through case studies, we demonstrate significant improvements in design efficiency, performance optimization, and reduced time-to-market. Additionally, we address challenges such as data availability, model interpretability, and the integration of AI-generated designs into existing verification workflows. The findings highlight the potential of generative AI to enhance design capabilities, reduce development costs, and accelerate innovation in semiconductor technology.

Loads and fatigue characteristics assessment of wind farm based on dynamic wake meandering model

As the scale of wind farms continues to expand, the impact of the wake on the power generation and load of wind turbines is becoming increasingly prominent. This study delves into the wake characteristics, power generation, load, and fatigue behavior of an ideal layout wind farm using a dynamic wake meandering model, accounting for diverse spacing configurations, inflow conditions, and yaw angles. The findings reveal that increasing the axial spacing facilitates superior wake recovery, promoting a more evenly distributed load and reducing fatigue damage throughout the wind farm. Moreover, higher inflow wind speed or turbulence intensity exacerbates fatigue damage, especially at elevated wind, where the damage accumulates exponentially. Under yaw conditions, the wake deviates, reducing its adverse impact on downstream turbines and subsequently boosting the overall power generation of the wind farm. However, this deflection simultaneously introduces intricate and detrimental effects on the fatigue load. Notably, optimal yaw achieves a 7.9 % improvement in power generation, also resulting in the most significant fatigue damage at the blade root and tower base. This study provides theoretical and technical underpinnings for the layout optimization and control strategies of wind farms or wind farm clusters.

Self-refined variational transformer for image-conditioned layout generation

Self-refined variational transformer for image-conditioned layout generation

Layout Generation: Automated Components Placement for Advertising Poster using Transformer-Based from Layout Graph

In the digital era, graphic design plays an important role in a company's marketing strategy, especially advertising posters that can convey messages to the audience. However, the process of making attractive and informative posters takes a long time, especially in the placement of components in the layout. This research aims to develop a layout generator system that focuses on placing components on the layout using the SGTransformer method. SGTransformer is one of the transformer-based methods that can be used to generate layouts. This research utilizes the capabilities of Laplacian positional encoding (LapPE) and edge features in SGTransformer that can improve the model's performance in developing a more structured layout based on the input layout graph. A layout graph describes the spatial relationship between components in a layout. The SGTransformer model is trained using advertising poster datasets collected from social media. Once trained, the model automatically places components based on the description of the spatial relationship between components represented by the layout graph. Evaluation of the model shows that the SGTransformer method can produce structured and more diverse layouts although there are still challenges such as overlap between components. This research contributes to the efficiency of the advertising poster design process and provides practical solutions to the creation of advertising posters in the digital marketing industry. Code and other materials will be released at https://github.com/syahdeee/Layout-Generator.

Automated layout generation from sites to flats using GAN and transfer learning

Generating architectural layouts from sites to flats, encompassing site layouts (SLs), building layouts (BLs), and flat layouts (FLs), presents a complex process. Notably, the BL generation is challenging due to the small scale of data, making it difficult to train effective neural networks. This paper introduces an approach for generating layouts throughout the complete process. Initially, it proposes an enhanced generative adversarial network (GAN) combined with the transformer for Stable Diffusion (TranSD-GAN), considering design boundaries and requirements. Subsequently, for generating BLs with small-scale datasets, the paper proposes a stacking transfer learning method. Following this, image operations are conducted to support the flow of building information. Ultimately, BIM models are created at each stage. Through comparative experiments involving neural networks and generation cases, it is demonstrated that the proposed method significantly improves the generative capabilities of small-scale datasets and effectively aids designers throughout the layout design from sites to flats.

Automated building layout generation: Implementation and comparison of STREAMER Early Design Configurator and SDaC Layout Designer

When designing new buildings, various constraints must be taken into account. These constraints encompass factors such as the overall building size, the number, function, and size of required spaces, and even the surrounding neighborhood's existing buildings, infrastructure, and natural environment. Particularly the requirements of interior layout design are usually coupled with each other, making the design process more complex and time-consuming. This paper proposes a novel manner to consider the requirements, aiming to facilitate the generation process of building's interior layout during the early design phase. Besides, this paper aims to generate vectored output, which is more applicable to export BIM model, which also benefits the subsequent design and construct steps. This paper introduces STREAMER Early Design Configurator (EDC) and SDaC Layout Designer, both of which formats the user's requirements, semi-automatically generates interior layouts and exports IFC models of the floor plans. STREAMER EDC is used to discover the potential layouts of large-scale buildings with many rooms such as hospitals, while SDaC Layout Designer focuses on residential houses with comparatively fewer rooms. These drafts are enriched with additional data, including placement information, volume details, and space utilization. Ultimately, the floor plans are exported into three-dimensional objects and exported in the IFC format. This comprehensive step paves the way for more efficient and informed building design, promoting greater flexibility and adaptability in the early stages of the architectural process.

Iris: a multi-constraint graphic layout generation system

Iris: a multi-constraint graphic layout generation system

Conditional room layout generation based on graph neural networks

In this paper, we present a novel end-to-end variational generative model that utilizes graph-based latent representations for indoor scene synthesis at one time. In contrast to prior research, our method deviates from the practice of gradually introducing and arranging furniture in an empty room using autoregression. Instead, it focuses on acquiring a comprehensive implicit representation of the room’s original architectural structure and the placement of furniture. We initially transform the 3D room scene into a dense scene graph, where nodes correspond to the objects present in the room while edges reflect the spatial location links and functional correlations between objects. Then, a neural network is trained to acquire the graph-based latent representation of the room scene through iterative message passing, ultimately resulting in the acquisition of the data distribution on the latent space of the room layout. Given the architectural structure of an empty room as a prerequisite for scene synthesis, the generative model has the ability to sample from the prior distribution of the room’s latent representation. This allows the model to then decode and generate a variety of room layouts. We evaluate our method with the state-of-the-art 3D indoor scene dataset and generation methods. The experimental results demonstrate that our method achieves more rational and diverse outcomes in the context of generating scenes under specific conditions.

A domain knowledge-informed design space exploration methodology for mechanical layout design

Layout designs have a large number of design variables and various physical constraints. Conventional design exploration approaches are time-consuming and may require human intervention. A unique feature about physical layout designs is the availability of domain knowledge, which can be utilised to speed up the design process. In this paper, we propose a generative deep learning-based design space exploration (DSE) methodology that is capable of learning design constraints in the layout design without explicit supervision. Moreover, it can incorporate domain knowledge in the generated layouts, thereby speeding up the design process. This is realised by constructing a layout generation variational autoencoder (LGVAE) model, which uses a latent space as an interface to generate the layouts. By training the LGVAE model, significantly lower-dimensional representations can be learned compared to the original dimensionality of the design space. Therefore, the number of design variables is greatly reduced. We showcase the performance of the proposed DSE approach by solving the heat source layout design problem encountered in thermal management of chips. Experiments demonstrate that the LGVAE model is capable of generating compressed latent representations that capture the characteristics of the input samples, which makes the DSE cost-effective.

Learning layout generation for virtual worlds

The emergence of the metaverse has led to the rapidly increasing demand for the generation of extensive 3D worlds. We consider that an engaging world is built upon a rational layout of multiple land-use areas (e.g., forest, meadow, and farmland). To this end, we propose a generative model of land-use distribution that learns from geographic data. The model is based on a transformer architecture that generates a 2D map of the land-use layout, which can be conditioned on spatial and semantic controls, depending on whether either one or both are provided. This model enables diverse layout generation with user control and layout expansion by extending borders with partial inputs. To generate high-quality and satisfactory layouts, we devise a geometric objective function that supervises the model to perceive layout shapes and regularize generations using geometric priors. Additionally, we devise a planning objective function that supervises the model to perceive progressive composition demands and suppress generations deviating from controls. To evaluate the spatial distribution of the generations, we train an autoencoder to embed land-use layouts into vectors to enable comparison between the real and generated data using the Wasserstein metric, which is inspired by the Fréchet inception distance.

StructDiffusion: End-to-end intelligent shear wall structure layout generation and analysis using diffusion model

Shear wall structures are widely used in high-rise buildings. However, the design process of shear wall structures suffers from inefficiencies. This paper introduces StructDiffusion, an end-to-end intelligent system for shear wall layout generation and analysis. The proposed approach tackles the layout design task by employing a novel diffusion model architecture to generate conditional images. Key components of the proposed method include pretrained diffusion models, ControlNet for conditional control, and LoRA for efficient adaptation. Notably, the method allows for architectural plan images and basic textual design conditions as inputs, enabling manipulation of the generated layouts by adjusting attributes such as building height and seismic intensity. To evaluate the quality of the model's design swiftly, this paper presents a comprehensive evaluation framework that incorporates perceptual and structural validity metrics. Through experimental analyses, this paper demonstrates the effectiveness of the proposed model in generating layouts, surpassing the capabilities of GAN-based methods. Furthermore, our study investigates model-specific parameter configuration, text-guided capabilities, code compliance, and computational efficiency. StructDiffusion significantly enhances the automation and intelligence of structural engineering workflows. The framework effectively addresses challenges associated with instability, data scarcity, and limitations in assessment. This pioneering application of diffusion models opens up promising avenues for advancing data-driven structural design.